WO2021073434A1 - Object behavior recognition method and apparatus, and terminal device - Google Patents

Abstract

An object behavior recognition method, comprising: obtaining a first description file of an object in a first historical time period and a second description file of the object in a second historical time period; for each type of first attribute description text in the first description file, obtaining a preset number of first keywords, converting each of the first keywords into a first feature vector, and averaging the preset number of first feature vectors to obtain a first average feature vector corresponding to each type of first attribute description text; combining the first average feature vectors corresponding to multiple types of first attribute description texts in the first description file to generate a first historical feature matrix; similarly, generating a second historical feature matrix of the second description file; and combining the first historical feature matrix and the second historical feature matrix and inputting same into a model to obtain a behavior recognition result of the object.

Description

对象行为的识别方法、装置及终端设备Recognition method, device and terminal equipment of object behavior

本申请要求于2019年10月16日在国家专利局提交的、申请号为201910981827.8、发明名称为“对象行为的识别方法、装置及终端设备”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。This application claims the priority of a Chinese patent application filed at the National Patent Office on October 16, 2019, with application number 201910981827.8, and the title of the invention "Methods, devices and terminal equipment for identifying object behaviors", the entire contents of which are incorporated by reference Incorporated in this application.

技术领域Technical field

本申请属于计算机技术领域，尤其涉及一种对象行为的识别方法、装置、终端设备及计算机可读存储介质。This application belongs to the field of computer technology, and in particular relates to a method, device, terminal device, and computer-readable storage medium for identifying object behavior.

背景技术Background technique

精确的行为识别具有挑战性，因为人类行为具有复杂性、高度多样化等特征。如何识别对象行为以完成对对象的评价，这种需求目前也非常普遍，特别是在人员招聘时显得尤为突出。Accurate behavior recognition is challenging because human behavior is complex and highly diverse. How to identify the behavior of the object in order to complete the evaluation of the object, this kind of demand is also very common at present, especially in the recruitment of personnel.

发明人意识到人员招聘是企业一项至关重要的活动，企业需要对新进人员进行评价以判定其是否满足岗位需求，因此，亟需一种对象行为的识别方案。The inventor realizes that personnel recruitment is a vital activity of an enterprise. The enterprise needs to evaluate new recruits to determine whether they meet the job requirements. Therefore, there is an urgent need for an object behavior identification scheme.

技术解决方案Technical solutions

本申请实施例提供了一种对象行为的识别方法、装置、终端设备及计算机可读存储介质，可以解决上述技术问题。The embodiments of the present application provide a method, a device, a terminal device, and a computer-readable storage medium for identifying object behaviors, which can solve the above technical problems.

第一方面，本申请实施例提供了一种对象行为的识别方法，包括：In the first aspect, an embodiment of the present application provides a method for identifying object behavior, including:

获取对象在第一历史时间段的第一描述文件和在第二历史时间段的第二描述文件；其中，所述第一描述文件包括多类第一属性描述文字，每类所述第一属性描述文字针对所述对象在所述第一历史时间段的一个属性进行描述；所述第二描述文件包括多类第二属性描述文字，每类所述第二属性描述文字针对所述对象在所述第二历史时间段的一个属性进行描述；Acquire the first description file of the object in the first historical time period and the second description file in the second historical time period; wherein, the first description file includes multiple types of first attribute description text, and each type of the first attribute The description text describes one attribute of the object in the first historical time period; the second description file includes multiple types of second attribute description text, and each type of the second attribute description text is specific to the object in the place. Describe an attribute of the second historical time period;

针对所述第一描述文件中每类所述第一属性描述文字，获取每类所述第一属性描述文字中预设数量个第一关键词，将每个所述第一关键词转化成第一特征向量，对预设数量个所述第一特征向量求均值，得到每类所述第一属性描述文字对应的第一均值特征向量；For each type of the first attribute description text in the first description file, obtain a preset number of first keywords in each type of the first attribute description text, and convert each of the first keywords into a A feature vector, averaging a preset number of the first feature vectors to obtain the first average feature vector corresponding to each type of the first attribute description text;

针对所述第二描述文件中每类所述第二属性描述文字，获取每类所述第二属性描述文字中预设数量个第二关键词，将每个所述第二关键词转化成第二特征向量，对预设数量个所述第二特征向量求均值，得到每类所述第二属性描述文字对应的第二均值特征向量；For each type of the second attribute description text in the second description file, obtain a preset number of second keywords in each type of the second attribute description text, and convert each of the second keywords into the first Two feature vectors, averaging a preset number of the second feature vectors to obtain a second average feature vector corresponding to each type of the second attribute description text;

将所述第一描述文件中多类所述第一属性描述文字对应的第一均值特征向量进行组合，生成第一历史特征矩阵；Combining the first mean value feature vectors corresponding to the multiple types of the first attribute description text in the first description file to generate a first historical feature matrix;

将所述第二描述文件中每类所述第二属性描述文字对应的第二均值特征向量进行组合，生成第二历史特征矩阵；Combine the second mean eigenvectors corresponding to each type of the second attribute description text in the second description file to generate a second historical feature matrix;

将所述第一历史特征矩阵和所述第二历史特征矩阵组合后输入模型得到所述对象的行为识别结果，其中，所述模型为使用多组数据通过机器学习训练得出的，所述多组数据中的每组数据包括第一历史时间段的第一历史特征矩阵样本，第二历史时间段的第二历史特征矩阵样本和行为标签，每种所述行为标签表征一种行为识别结果。The first historical feature matrix and the second historical feature matrix are combined and input into a model to obtain the behavior recognition result of the object, wherein the model is obtained through machine learning training using multiple sets of data, and the multiple Each group of data in the group of data includes a first historical feature matrix sample in a first historical time period, a second historical feature matrix sample in a second historical time period, and behavior labels, each of the behavior labels represents a behavior recognition result.

第二方面，本申请实施例提供了一种对象行为的识别装置，包括：In the second aspect, an embodiment of the present application provides an object behavior recognition device, including:

包括：文档获取模块、特征向量获取模块、矩阵生成模块和行为识别模块；Including: document acquisition module, feature vector acquisition module, matrix generation module and behavior recognition module;

所述文档获取模块，用于：The document acquisition module is used to:

获取对象在第一历史时间段的第一描述文件和在第二历史时间段的第二描述文件；其中，所述第一描述文件包括多类第一属性描述文字，每类所述第一属性描述文字针对所述对象在所述第一历史时间段的一个属性进行描述；所述第二描述文件包括多类第二属性描述文字，每类所述第二属性描述文字针对所述对象在所述第二历史时间段的一个属性进行描述；Acquire the first description file of the object in the first historical time period and the second description file in the second historical time period; wherein, the first description file includes multiple types of first attribute description text, and each type of the first attribute The description text describes one attribute of the object in the first historical time period; the second description file includes multiple types of second attribute description text, and each type of the second attribute description text is specific to the object in the place. Describe an attribute of the second historical time period;

所述特征向量获取模块，用于：The feature vector acquisition module is used to:

针对所述第一描述文件中每类所述第一属性描述文字，获取每类所述第一属性描述文字中预设数量个第一关键词，将每个所述第一关键词转化成第一特征向量，对预设数量个所述 第一特征向量求均值，得到每类所述第一属性描述文字对应的第一均值特征向量；For each type of the first attribute description text in the first description file, obtain a preset number of first keywords in each type of the first attribute description text, and convert each of the first keywords into a A feature vector, averaging a preset number of the first feature vectors to obtain the first average feature vector corresponding to each type of the first attribute description text;

针对所述第二描述文件中每类所述第二属性描述文字，获取每类所述第二属性描述文字中预设数量个第二关键词，将每个所述第二关键词转化成第二特征向量，对预设数量个所述第二特征向量求均值，得到每类所述第二属性描述文字对应的第二均值特征向量；For each type of the second attribute description text in the second description file, obtain a preset number of second keywords in each type of the second attribute description text, and convert each of the second keywords into the first Two feature vectors, averaging a preset number of the second feature vectors to obtain a second average feature vector corresponding to each type of the second attribute description text;

所述矩阵生成模块，用于：The matrix generation module is used to:

将所述第一描述文件中多类所述第一属性描述文字对应的第一均值特征向量进行组合，生成第一历史特征矩阵；Combining the first mean value feature vectors corresponding to the multiple types of the first attribute description text in the first description file to generate a first historical feature matrix;

将所述第二描述文件中每类所述第二属性描述文字对应的第二均值特征向量进行组合，生成第二历史特征矩阵；Combine the second mean eigenvectors corresponding to each type of the second attribute description text in the second description file to generate a second historical feature matrix;

所述行为识别模块，用于：The behavior recognition module is used to:

将所述第一历史特征矩阵和所述第二历史特征矩阵组合后输入模型得到所述对象的行为识别结果，其中，所述模型为使用多组数据通过机器学习训练得出的，所述多组数据中的每组数据包括第一历史时间段的第一历史特征矩阵样本，第二历史时间段的第二历史特征矩阵样本和行为标签，每种所述行为标签表征一种行为识别结果。The first historical feature matrix and the second historical feature matrix are combined and input into a model to obtain the behavior recognition result of the object, wherein the model is obtained through machine learning training using multiple sets of data, and the multiple Each group of data in the group of data includes a first historical feature matrix sample in a first historical time period, a second historical feature matrix sample in a second historical time period, and behavior labels, each of the behavior labels represents a behavior recognition result.

第三方面，本申请实施例提供了一种终端设备，包括：存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现如第一方面所述的识别方法。In a third aspect, an embodiment of the present application provides a terminal device, including: a memory, a processor, and a computer program stored in the memory and running on the processor, and the processor executes the computer program When realizing the identification method as described in the first aspect.

第四方面，本申请实施例提供了一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，所述计算机程序被处理器执行时实现如第一方面所述的识别方法。In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the recognition method as described in the first aspect is implemented.

第五方面，本申请实施例提供了一种计算机程序产品，当计算机程序产品在终端设备上运行时，使得终端设备执行如第一方面所述的识别方法。In the fifth aspect, the embodiments of the present application provide a computer program product, which when the computer program product runs on a terminal device, causes the terminal device to execute the identification method as described in the first aspect.

有益效果Beneficial effect

在本申请实施例中，利用机器学习的方式预先训练模型，基于对象两个不同时间段的描述文件，分别提取描述文件的特征矩阵，将两个特征矩阵组合后输入模型获得对象的行为识别结果，提供了一种对象行为的识别方案，一方面，该识别方案利用更多的资料对对象进行行为识别，增大了输入模型的信息量，提高了对象行为识别的精度，另一方面，该识别方案通过将描述文件中多类属性分别对应的特征向量进行组合形成描述文件对应的特征矩阵，提取了高质量的数据，也减少了噪音，在保证识别结果高精度的同时，也减少了数据处理量，减少了***资源占用。In the embodiment of this application, the model is pre-trained by means of machine learning, and the feature matrix of the description file is extracted based on the description files of the object in two different time periods, and the two feature matrices are combined and input into the model to obtain the behavior recognition result of the object , Provides an object behavior recognition program. On the one hand, the recognition program uses more data to recognize the behavior of the object, increases the amount of information input to the model, and improves the accuracy of object behavior recognition. On the other hand, the The recognition scheme combines the feature vectors corresponding to the multiple types of attributes in the description file to form a feature matrix corresponding to the description file, which extracts high-quality data and reduces noise. While ensuring the high accuracy of the recognition results, it also reduces data The processing volume reduces the system resource occupation.

附图说明Description of the drawings

图1是本申请一实施例提供的对象行为的识别方法所适用于的手机的结构示意图；FIG. 1 is a schematic structural diagram of a mobile phone to which the method for identifying object behaviors provided by an embodiment of the present application is applicable;

图2是本申请一实施例提供的对象行为的识别方法的流程示意图；2 is a schematic flowchart of a method for identifying object behaviors provided by an embodiment of the present application;

图3是本申请一实施例提供的对象行为的识别方法中步骤202的流程示意图；FIG. 3 is a schematic flowchart of step 202 in the method for identifying object behaviors provided by an embodiment of the present application;

图4是本申请一实施例提供的对象行为的识别方法中步骤203的流程示意图；FIG. 4 is a schematic flowchart of step 203 in the method for identifying object behaviors provided by an embodiment of the present application;

图5是本申请一实施例提供的对象行为的识别装置的结构示意图；FIG. 5 is a schematic structural diagram of an object behavior recognition device provided by an embodiment of the present application;

图6是本申请一实施例提供的对象行为的识别方法所适用于的终端设备的结构示意图。FIG. 6 is a schematic structural diagram of a terminal device to which the method for identifying object behaviors provided by an embodiment of the present application is applicable.

本发明的实施方式Embodiments of the present invention

以下描述中，为了说明而不是为了限定，提出了诸如特定***结构、技术之类的具体细节，以便透彻理解本申请实施例。In the following description, for the purpose of illustration rather than limitation, specific details such as a specific system structure and technology are proposed for a thorough understanding of the embodiments of the present application.

为了使本技术领域的人员更好地理解本申请方案，下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚，完整地描述，显然，所描述的实施例仅仅是本申请一部分的实施例，而不是全部的实施例。基于本申请中的实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，所获得的所有其他实施例，都应当属于本申请保护的范围。需要说明的是，在不冲突的情况下，本申请中的实施例及实施例中的特征可以相互组合。In order to enable those skilled in the art to better understand the solution of the application, the technical solutions in the embodiments of the application will be clearly and completely described below in conjunction with the drawings in the embodiments of the application. Obviously, the described embodiments are only It is a part of the embodiments of this application, not all the embodiments. Based on the embodiments in this application, for those of ordinary skill in the art, all other embodiments obtained without creative labor should fall within the protection scope of this application. It should be noted that the embodiments in this application and the features in the embodiments can be combined with each other if there is no conflict.

以下描述中，为了说明而不是为了限定，提出了诸如特定***结构、技术之类的具体细节，以便透彻理解本申请实施例。然而，本领域的技术人员应当清楚，在没有这些具体细节的其它实施例中也可以实现本申请。在其它情况中，省略对众所周知的***、装置、电路以及方法的详细说明，以免不必要的细节妨碍本申请的描述。In the following description, for the purpose of illustration rather than limitation, specific details such as a specific system structure and technology are proposed for a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from obstructing the description of this application.

应当理解，当在本申请说明书和所附权利要求书中使用时，术语“包括”指示所描述特征、整体、步骤、操作、元素和/或组件的存在，但并不排除一个或多个其它特征、整体、步骤、操作、元素、组件和/或其集合的存在或添加。It should be understood that when used in the specification and appended claims of this application, the term "comprising" indicates the existence of the described features, wholes, steps, operations, elements and/or components, but does not exclude one or more other The existence or addition of features, wholes, steps, operations, elements, components, and/or collections thereof.

还应当理解，在本申请说明书和所附权利要求书中使用的术语“和/或”是指相关联列出的项中的一个或多个的任何组合以及所有可能组合，并且包括这些组合。It should also be understood that the term "and/or" used in the specification and appended claims of this application refers to any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.

如在本申请说明书和所附权利要求书中所使用的那样，术语“如果”可以依据上下文被解释为“当...时”或“一旦”或“响应于确定”或“响应于检测到”。类似地，短语“如果确定”或“如果检测到[所描述条件或事件]”可以依据上下文被解释为意指“一旦确定”或“响应于确定”或“一旦检测到[所描述条件或事件]”或“响应于检测到[所描述条件或事件]”。As used in the description of this application and the appended claims, the term "if" can be construed as "when" or "once" or "in response to determination" or "in response to detecting ". Similarly, the phrase "if determined" or "if detected [described condition or event]" can be interpreted as meaning "once determined" or "in response to determination" or "once detected [described condition or event]" depending on the context ]" or "in response to detection of [condition or event described]".

另外，在本申请说明书和所附权利要求书的描述中，术语“第一”、“第二”、“第三”等仅用于区分描述，而不能理解为指示或暗示相对重要性。In addition, in the description of the specification of this application and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

在本申请说明书中描述的参考“一个实施例”或“一些实施例”等意味着在本申请的一个或多个实施例中包括结合该实施例描述的特定特征、结构或特点。由此，在本说明书中的不同之处出现的语句“在一个实施例中”、“在一些实施例中”、“在其他一些实施例中”、“在另外一些实施例中”等不是必然都参考相同的实施例，而是意味着“一个或多个但不是所有的实施例”，除非是以其他方式另外特别强调。术语“包括”、“包含”、“具有”及它们的变形都意味着“包括但不限于”，除非是以其他方式另外特别强调。Reference to "one embodiment" or "some embodiments" described in the specification of this application means that one or more embodiments of this application include a specific feature, structure, or characteristic described in combination with the embodiment. Therefore, the sentences "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. appearing in different places in this specification are not necessarily All refer to the same embodiment, but mean "one or more but not all embodiments" unless it is specifically emphasized otherwise. The terms "including", "including", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

本申请实施例提供的对象行为的识别方法可以应用于手机、平板电脑、可穿戴设备、车载设备、增强现实(augmented reality，AR)/虚拟现实(virtual reality，VR)设备、笔记本电脑、超级移动个人计算机(ultra-mobile personal computer，UMPC)、上网本、个人数字助理(personal digital assistant，PDA)、服务器或云端服务器等终端设备上，本申请实施例对终端设备的具体类型不作任何限制。The object behavior recognition method provided by the embodiments of this application can be applied to mobile phones, tablet computers, wearable devices, in-vehicle devices, augmented reality (AR)/virtual reality (VR) devices, notebook computers, and ultra mobile devices. For terminal devices such as ultra-mobile personal computers (UMPC), netbooks, personal digital assistants (personal digital assistants, PDAs), servers, or cloud servers, the embodiments of this application do not impose any restrictions on the specific types of terminal devices.

例如，所述终端设备可以是WLAN中的站点(STAION，ST)，可以是蜂窝电话、无绳电话、会话启动协议(Session InitiationProtocol，SIP)电话、无线本地环路(Wireless Local Loop，WLL)站、PDA、具有无线通信功能的手持设备、计算设备或连接到无线调制解调器的其它处理设备、车载设备、车联网终端、电脑、膝上型计算机、手持式通信设备、手持式计算设备、卫星无线设备、无线调制解调器卡、电视机顶盒(set top box，STB)、用户驻地设备(customer premise equipment，CPE)和/或用于在无线***上进行通信的其它设备以及下一代通信***，例如，5G网络中的移动终端或者未来演进的公共陆地移动网络(Public Land Mobile Network，PLMN)网络中的移动终端等。For example, the terminal device may be a station (STAION, ST) in a WLAN, a cellular phone, a cordless phone, a Session Initiation Protocol (SIP) phone, a wireless local loop (Wireless Local Loop, WLL) station, PDAs, handheld devices with wireless communication capabilities, computing devices or other processing devices connected to wireless modems, vehicle-mounted devices, car networking terminals, computers, laptop computers, handheld communication devices, handheld computing devices, satellite wireless devices, Wireless modem cards, TV set top boxes (STB), customer premise equipment (CPE) and/or other equipment used for communication on wireless systems, and next-generation communication systems, such as those in 5G networks A mobile terminal or a mobile terminal in the future evolved Public Land Mobile Network (PLMN) network, etc.

作为示例而非限定，当所述终端设备为可穿戴设备时，该可穿戴设备还可以是应用穿戴式技术对日常穿戴进行智能化设计、开发出可以穿戴的设备的总称，如眼镜、手套、手表、服饰及鞋等。可穿戴设备即直接穿在身上，或是整合到用户的衣服或配件的一种便携式设备。可穿戴设备不仅仅是一种硬件设备，更是通过软件支持以及数据交互、云端交互来实现强大的功能。广义穿戴式智能设备包括功能全、尺寸大、可不依赖智能手机实现完整或者部分的功能，如智能手表或智能眼镜等，以及只专注于某一类应用功能，需要和其它设备如智能手机配合使用，如各类进行体征监测的智能手环、智能首饰等。As an example and not a limitation, when the terminal device is a wearable device, the wearable device can also be a general term for applying wearable technology to intelligently design daily wear and develop wearable devices, such as glasses, gloves, Watches, clothing and shoes, etc. A wearable device is a portable device that is directly worn on the body or integrated into the user's clothes or accessories. Wearable devices are not only a kind of hardware device, but also realize powerful functions through software support, data interaction, and cloud interaction. In a broad sense, wearable smart devices include full-featured, large-sized, complete or partial functions that can be implemented without relying on smart phones, such as smart watches or smart glasses, and only focus on a certain type of application function, and need to be used in conjunction with other devices such as smart phones. , Such as all kinds of smart bracelets and smart jewelry for physical sign monitoring.

以所述终端设备为手机为例。图1示出的是与本申请实施例提供的手机的部分结构的框图。参考图1，手机包括：射频(Radio Frequency，RF)电路110、存储器120、输入单元130、显示单元140、传感器150、音频电路160、无线保真(wireless fidelity，WiFi)模块170、处理器180、以及电源190等部件。本领域技术人员可以理解，图1中示出的手机结构并不构成对手机的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同 的部件布置。Take the terminal device as a mobile phone as an example. Fig. 1 shows a block diagram of a part of the structure of a mobile phone provided in an embodiment of the present application. 1, the mobile phone includes: a radio frequency (RF) circuit 110, a memory 120, an input unit 130, a display unit 140, a sensor 150, an audio circuit 160, a wireless fidelity (WiFi) module 170, and a processor 180 , And power supply 190 and other components. Those skilled in the art can understand that the structure of the mobile phone shown in FIG. 1 does not constitute a limitation on the mobile phone, and may include more or less components than those shown in the figure, or a combination of certain components, or different component arrangements.

下面结合图1对手机的各个构成部件进行具体的介绍：The following is a detailed introduction to each component of the mobile phone in conjunction with Figure 1:

RF电路110可用于收发信息或通话过程中，信号的接收和发送，特别地，将基站的下行信息接收后，给处理器180处理；另外，将设计上行的数据发送给基站。通常，RF电路包括但不限于天线、至少一个放大器、收发信机、耦合器、低噪声放大器(Low Noise Amplifier，LNA)、双工器等。此外，RF电路110还可以通过无线通信与网络和其他设备通信。上述无线通信可以使用任一通信标准或协议，包括但不限于全球移动通讯***(Global System of Mobile communication，GSM)、通用分组无线服务(General Packet Radio Service，GPRS)、码分多址(Code Division Multiple Access，CDMA)、宽带码分多址(Wideband Code Division Multiple Access，WCDMA)、长期演进(Long Term Evolution，LTE)、电子邮件、以及短消息服务(Short Messaging Service，SMS)等。The RF circuit 110 can be used for receiving and sending signals during information transmission or communication. In particular, after receiving the downlink information of the base station, it is processed by the processor 180; in addition, the designed uplink data is sent to the base station. Generally, the RF circuit includes, but is not limited to, an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier (LNA), a duplexer, and the like. In addition, the RF circuit 110 may also communicate with the network and other devices through wireless communication. The above-mentioned wireless communication can use any communication standard or protocol, including but not limited to Global System of Mobile Communication (GSM), General Packet Radio Service (GPRS), Code Division Multiple Access (Code Division) Multiple Access (CDMA), Wideband Code Division Multiple Access (WCDMA), Long Term Evolution (LTE), Email, and Short Messaging Service (SMS), etc.

存储器120可用于存储软件程序以及模块，处理器180通过运行存储在存储器120的软件程序以及模块，从而执行手机的各种功能应用以及数据处理。存储器120可主要包括存储程序区和存储数据区，其中，存储程序区可存储操作***、至少一个功能所需的应用程序(比如声音播放功能、图像播放功能等)、引导装载程序(Boot Loader)等；存储数据区可存储根据手机的使用所创建的数据(比如音频数据、电话本等)等。此外，存储器120可以包括高速随机存取存储器，还可以包括非易失性存储器，例如至少一个磁盘存储器件、闪存器件、或其他易失性固态存储器件。可以理解的是，本申请实施例中，存储器120中存储有对象行为识别的程序。The memory 120 may be used to store software programs and modules. The processor 180 executes various functional applications and data processing of the mobile phone by running the software programs and modules stored in the memory 120. The memory 120 may mainly include a program storage area and a data storage area. The storage program area may store an operating system, an application program required by at least one function (such as a sound playback function, an image playback function, etc.), and a boot loader (BootLoader). Etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, phone book, etc.) and so on. In addition, the memory 120 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other volatile solid-state storage devices. It can be understood that, in this embodiment of the present application, a program for object behavior recognition is stored in the memory 120.

输入单元130可用于接收输入的数字或字符信息，以及产生与手机100的用户设置以及功能控制有关的键信号输入。具体地，输入单元130可包括触控面板131以及其他输入设备132。触控面板131，也称为触摸屏，可收集用户在其上或附近的触摸操作(比如用户使用手指、触笔等任何适合的物体或附件在触控面板131上或在触控面板131附近的操作)，并根据预先设定的程式驱动相应的连接装置。可选的，触控面板131可包括触摸检测装置和触摸控制器两个部分。其中，触摸检测装置检测用户的触摸方位，并检测触摸操作带来的信号，将信号传送给触摸控制器；触摸控制器从触摸检测装置上接收触摸信息，并将它转换成触点坐标，再送给处理器180，并能接收处理器180发来的命令并加以执行。此外，可以采用电阻式、电容式、红外线以及表面声波等多种类型实现触控面板131。除了触控面板131，输入单元130还可以包括其他输入设备132。具体地，其他输入设备132可以包括但不限于物理键盘、功能键(比如音量控制按键、开关按键等)、轨迹球、鼠标、操作杆等中的一种或多种。The input unit 130 may be used to receive inputted numeric or character information, and generate key signal input related to user settings and function control of the mobile phone 100. Specifically, the input unit 130 may include a touch panel 131 and other input devices 132. The touch panel 131, also known as a touch screen, can collect user touch operations on or near it (for example, the user uses any suitable objects or accessories such as fingers, stylus, etc.) on the touch panel 131 or near the touch panel 131. Operation), and drive the corresponding connection device according to the preset program. Optionally, the touch panel 131 may include two parts: a touch detection device and a touch controller. Among them, the touch detection device detects the user's touch position, detects the signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts it into contact coordinates, and then sends it To the processor 180, and can receive and execute the commands sent by the processor 180. In addition, the touch panel 131 can be implemented in multiple types such as resistive, capacitive, infrared, and surface acoustic wave. In addition to the touch panel 131, the input unit 130 may also include other input devices 132. Specifically, the other input device 132 may include, but is not limited to, one or more of a physical keyboard, function keys (such as volume control buttons, switch buttons, etc.), trackball, mouse, and joystick.

显示单元140可用于显示由用户输入的信息或提供给用户的信息以及手机的各种菜单。显示单元140可包括显示面板141，可选的，可以采用液晶显示器(Liquid Crystal Display，LCD)、有机发光二极管(Organic Light-Emitting Diode,OLED)等形式来配置显示面板141。进一步的，触控面板131可覆盖显示面板141，当触控面板131检测到在其上或附近的触摸操作后，传送给处理器180以确定触摸事件的类型，随后处理器180根据触摸事件的类型在显示面板141上提供相应的视觉输出。虽然在图1中，触控面板131与显示面板141是作为两个独立的部件来实现手机的输入和输入功能，但是在某些实施例中，可以将触控面板131与显示面板141集成而实现手机的输入和输出功能。The display unit 140 may be used to display information input by the user or information provided to the user and various menus of the mobile phone. The display unit 140 may include a display panel 141. Optionally, the display panel 141 may be configured in the form of a liquid crystal display (LCD), an organic light-emitting diode (OLED), or the like. Further, the touch panel 131 can cover the display panel 141. When the touch panel 131 detects a touch operation on or near it, it transmits it to the processor 180 to determine the type of the touch event, and then the processor 180 responds to the touch event. The type provides corresponding visual output on the display panel 141. Although in FIG. 1, the touch panel 131 and the display panel 141 are used as two independent components to realize the input and input functions of the mobile phone, but in some embodiments, the touch panel 131 and the display panel 141 can be integrated. Realize the input and output functions of the mobile phone.

手机100还可包括至少一种传感器150，比如光传感器、运动传感器以及其他传感器。具体地，光传感器可包括环境光传感器及接近传感器，其中，环境光传感器可根据环境光线的明暗来调节显示面板141的亮度，接近传感器可在手机移动到耳边时，关闭显示面板141和/或背光。作为运动传感器的一种，加速计传感器可检测各个方向上(一般为三轴)加速度的大小，静止时可检测出重力的大小及方向，可用于识别手机姿态的应用(比如横竖屏切换、相关游戏、磁力计姿态校准)、振动识别相关功能(比如计步器、敲击)等；至于手机还可配置的陀螺仪、气压计、湿度计、温度计、红外线传感器等其他传感器，在此不再赘述。The mobile phone 100 may also include at least one sensor 150, such as a light sensor, a motion sensor, and other sensors. Specifically, the light sensor may include an ambient light sensor and a proximity sensor. The ambient light sensor can adjust the brightness of the display panel 141 according to the brightness of the ambient light. The proximity sensor can close the display panel 141 and/or when the mobile phone is moved to the ear. Or backlight. As a kind of motion sensor, the accelerometer sensor can detect the magnitude of acceleration in various directions (usually three-axis), and can detect the magnitude and direction of gravity when it is stationary. It can be used to identify mobile phone posture applications (such as horizontal and vertical screen switching, related Games, magnetometer posture calibration), vibration recognition related functions (such as pedometer, percussion), etc.; as for other sensors such as gyroscopes, barometers, hygrometers, thermometers, infrared sensors, etc., which can also be configured in mobile phones, I will not here Go into details.

音频电路160，扬声器161，传声器162可提供用户与手机之间的音频接口。音频电路160可将接收到的音频数据转换后的电信号，传输到扬声器161，由扬声器161转换为声音信号输出；另一方面，传声器162将收集的声音信号转换为电信号，由音频电路160接收后转换为音频数据，再将音频数据输出处理器180处理后，经RF电路110以发送给比如另一手机，或者将音频数据输出至存储器120以便进一步处理。The audio circuit 160, the speaker 161, and the microphone 162 can provide an audio interface between the user and the mobile phone. The audio circuit 160 can transmit the electrical signal converted from the received audio data to the speaker 161, which is converted into a sound signal for output by the speaker 161; on the other hand, the microphone 162 converts the collected sound signal into an electrical signal, which is then output by the audio circuit 160. After being received, it is converted into audio data, and then processed by the audio data output processor 180, and sent to, for example, another mobile phone via the RF circuit 110, or the audio data is output to the memory 120 for further processing.

WiFi属于短距离无线传输技术，手机通过WiFi模块170可以帮助用户收发电子邮件、浏览网页和访问流式媒体等，它为用户提供了无线的宽带互联网访问。虽然图1示出了WiFi模块170，但是可以理解的是，其并不属于手机100的必须构成，完全可以根据需要在不改变发明的本质的范围内而省略。WiFi is a short-distance wireless transmission technology. The mobile phone can help users send and receive emails, browse web pages, and access streaming media through the WiFi module 170. It provides users with wireless broadband Internet access. Although FIG. 1 shows the WiFi module 170, it is understandable that it is not a necessary component of the mobile phone 100, and can be omitted as needed without changing the essence of the invention.

处理器180是手机的控制中心，利用各种接口和线路连接整个手机的各个部分，通过运行或执行存储在存储器120内的软件程序和/或模块，以及调用存储在存储器120内的数据，执行手机的各种功能和处理数据，从而对手机进行整体监控。可选的，处理器180可包括一个或多个处理单元；优选的，处理器180可集成应用处理器和调制解调处理器，其中，应用处理器主要处理操作***、用户界面和应用程序等，调制解调处理器主要处理无线通信。可以理解的是，上述调制解调处理器也可以不集成到处理器180中。可以理解的是，本申请实施例中，存储器120中存储有对象行为识别的程序，而处理器180可以用于调用存储器120中存储的对象行为识别的程序并执行，以实现本申请实施例的对象行为的识别方法。The processor 180 is the control center of the mobile phone. It uses various interfaces and lines to connect various parts of the entire mobile phone. It executes by running or executing software programs and/or modules stored in the memory 120 and calling data stored in the memory 120. Various functions and processing data of the mobile phone can be used to monitor the mobile phone as a whole. Optionally, the processor 180 may include one or more processing units; preferably, the processor 180 may integrate an application processor and a modem processor, where the application processor mainly processes an operating system, a user interface, and application programs, etc. , The modem processor mainly deals with wireless communication. It can be understood that the foregoing modem processor may not be integrated into the processor 180. It can be understood that, in the embodiment of the present application, the memory 120 stores an object behavior recognition program, and the processor 180 may be used to call and execute the object behavior recognition program stored in the memory 120, so as to implement the object behavior recognition program of the embodiment of the present application. The method of identifying the behavior of the object.

手机100还包括给各个部件供电的电源190(比如电池)，优选的，电源可以通过电源管理***与处理器180逻辑相连，从而通过电源管理***实现管理充电、放电、以及功耗管理等功能。The mobile phone 100 also includes a power source 190 (such as a battery) for supplying power to various components. Preferably, the power source can be logically connected to the processor 180 through a power management system, so that functions such as charging, discharging, and power consumption management can be managed through the power management system.

尽管未示出，手机100还可以包括摄像头。可选地，摄像头在手机100的上的位置可以为前置的，也可以为后置的，还可以为内置的(在使用时可伸出机身)，本申请实施例对此不作限定。Although not shown, the mobile phone 100 may also include a camera. Optionally, the position of the camera on the mobile phone 100 may be front-mounted, rear-mounted, or built-in (that can extend out of the body when in use), which is not limited in the embodiment of the present application.

可选地，手机100可以包括单摄像头、双摄像头或三摄像头等，本申请实施例对此不作限定。摄像头包括但不限于广角摄像头、长焦摄像头或深度摄像头等。Optionally, the mobile phone 100 may include a single camera, a dual camera, or a triple camera, etc., which is not limited in the embodiment of the present application. Cameras include, but are not limited to, wide-angle cameras, telephoto cameras, or depth cameras.

例如，手机100可以包括三摄像头，其中，一个为主摄像头、一个为广角摄像头、一个为长焦摄像头。For example, the mobile phone 100 may include three cameras, of which one is a main camera, one is a wide-angle camera, and one is a telephoto camera.

可选地，当手机100包括多个摄像头时，这多个摄像头可以全部前置，或者全部后置，或者全部内置，或者至少部分前置，或者至少部分后置，或者至少部分内置等，本申请实施例对此不作限定。Optionally, when the mobile phone 100 includes multiple cameras, the multiple cameras may be all front-mounted, or all rear-mounted, or all built-in, or at least partially front-mounted, or at least partially rear-mounted, or at least partially built-in, etc. The application embodiment does not limit this.

另外，尽管未示出，手机100还可以包括蓝牙模块等，在此不再赘述。In addition, although not shown, the mobile phone 100 may also include a Bluetooth module, etc., which will not be repeated here.

图2示出了本申请实施例提供的一种对象行为的识别方法的实现流程图。所述识别方法应用于终端设备。作为示例而非限定，该方法可以应用于具有上述硬件结构的手机100中，以下实施例将以手机100为例，对本申请实施例提供的对象行为的识别方法进行说明。所述方法包括步骤S201至步骤S206。Fig. 2 shows an implementation flowchart of an object behavior recognition method provided by an embodiment of the present application. The identification method is applied to terminal equipment. As an example and not a limitation, the method can be applied to the mobile phone 100 having the above hardware structure. The following embodiments will take the mobile phone 100 as an example to describe the object behavior identification method provided in the embodiments of the present application. The method includes step S201 to step S206.

S201，获取对象在第一历史时间段的第一描述文件和在第二历史时间段的第二描述文件。S201: Acquire a first description file of an object in a first historical time period and a second description file of an object in a second historical time period.

在本申请实施例中，所述第一描述文件包括多类第一属性描述文字，每类所述第一属性描述文字针对所述对象在所述第一历史时间段的一个属性进行描述；所述第二描述文件包括多类第二属性描述文字，每类所述第二属性描述文字针对所述对象在所述第二历史时间段的一个属性进行描述。In the embodiment of the present application, the first description file includes multiple types of first attribute description text, and each type of the first attribute description text describes an attribute of the object in the first historical time period; The second description file includes multiple types of second attribute description text, and each type of the second attribute description text describes an attribute of the object in the second historical time period.

其中，对象为待进行行为识别的对象，例如公司新进员工等。属性是与对象行为识别相关的属性。Among them, the object is an object to be recognized for behavior, such as a new employee of the company. Attributes are attributes related to object behavior recognition.

可以理解的是，第一历史时间段和第二历史时间段可以在时间上具有连续性，也可以不在时间上具有连续性，两者用于表示两个不同的时间段。It is understandable that the first historical time period and the second historical time period may or may not have continuity in time, and both are used to indicate two different time periods.

第一描述文件和第二描述文件可以是用户通过终端设备的摄像头即时拍摄到的包括描述文件内容的图片；还可以是用户通过终端设备的扫描装置即时扫描到的包括描述文件的图 片；还可以是原本已存储在所述终端设备中的文件；甚至可以是所述终端设备通过有线或无线网络从服务器(包括独立的服务器、云端服务器、分布式服务器和服务器集群等)或者其它终端设备处所获取到的文件等。The first description file and the second description file may be pictures including description files that are instantly captured by the user through the camera of the terminal device; they may also be pictures including description files that are instantly scanned by the user through the scanning device of the terminal device; or It is a file originally stored in the terminal device; it can even be obtained by the terminal device from a server (including independent servers, cloud servers, distributed servers, server clusters, etc.) or other terminal devices through a wired or wireless network The documents that arrive, etc.

在本申请一种非限定性使用场景中，当用户想要直接进行对象行为识别时，通过点击终端设备特定的物理按键或者虚拟按键的方式启用终端设备的行为识别功能，在这种模式下，所述终端设备会对用户拍摄的第一描述文件以及第二描述文件自动按照步骤S202及步骤S206的过程进行处理，得到行为识别结果。In a non-limiting use scenario of the present application, when the user wants to directly perform object behavior recognition, the behavior recognition function of the terminal device is activated by clicking a specific physical button or virtual button of the terminal device. In this mode, The terminal device automatically processes the first description file and the second description file taken by the user according to the process of step S202 and step S206 to obtain the behavior recognition result.

在本申请另一种非限定性使用场景中，当用户想要对已经存储的第一描述文件以及第二描述文件进行行为识别时，可以通过点击特定的物理按键或者虚拟按键的方式启用终端设备的行为识别功能，并选定第一描述文件以及第二描述文件，则所述终端设备会对第一描述文件以及第二描述文件自动按照步骤S202及步骤S206的过程进行处理，得到行为识别结果。此处需要说明的是，点击按键和选定描述文件的顺序可以互换，即也可以先选定描述文件，再打开终端设备的行为识别功能。In another non-limiting use scenario of this application, when the user wants to perform behavior recognition on the stored first description file and the second description file, the terminal device can be activated by clicking a specific physical button or virtual button. If the first description file and the second description file are selected, the terminal device will automatically process the first description file and the second description file according to the process of step S202 and step S206 to obtain the behavior recognition result . What needs to be explained here is that the order of clicking the button and selecting the description file can be interchanged, that is, you can also select the description file first, and then turn on the behavior recognition function of the terminal device.

作为本申请实施例的一示例而非限定，当对象为公司新进员工时，第一历史时间段可以为员工在入公司之前的历史时间段；第二历史时间段可以为员工从入公司起至对对象进行行为识别的当下这个历史时间段，例如试用期等。As an example and not a limitation of the embodiment of the present application, when the object is a new employee of the company, the first historical time period may be the historical time period before the employee enters the company; the second historical time period may be the employee since entering the company This historical period of time to the current time when the behavior of the object is recognized, such as a trial period.

从终端设备的存储器中获取预先存储的或者用户(例如企业负责招聘的人员等)实时输入的新进员工入职简历，或者从网站爬取新进员工的历史求职简历等，作为第一描述文件。将新进员工入职公司之后的所有职业记录作为第二描述文件，例如针对试用期表现描述的记录文档。Obtain from the memory of the terminal device a pre-stored or real-time entry resume of a new employee (such as a company responsible for recruiting), or crawl the historical resume of a new employee from a website, as the first description file. Use all occupation records of new employees after joining the company as the second description file, such as a record file describing performance during the probation period.

第一描述文件和第二描述文件中包括针对对象的多个属性的描述文字；第一描述文件和第二描述文件中的属性包括但不限于工龄、晋升次数、项目经验、职业技能、和薪资涨幅等。The first description file and the second description file include description text for multiple attributes of the object; the attributes in the first description file and the second description file include but are not limited to length of service, number of promotions, project experience, professional skills, and salary Increases and so on.

S202，针对所述第一描述文件中每类所述第一属性描述文字，获取每类所述第一属性描述文字中预设数量个第一关键词，将每个所述第一关键词转化成第一特征向量，对预设数量个所述第一特征向量求均值，得到每类所述第一属性描述文字对应的第一均值特征向量。S202: For each type of the first attribute description text in the first description file, obtain a preset number of first keywords in each type of the first attribute description text, and convert each of the first keywords A first feature vector is generated, and a preset number of the first feature vectors are averaged to obtain the first average feature vector corresponding to each type of the first attribute description text.

S203，针对所述第二描述文件中每类所述第二属性描述文字，获取每类所述第二属性描述文字中预设数量个第二关键词，将每个所述第二关键词转化成第二特征向量，对预设数量个所述第二特征向量求均值，得到每类所述第二属性描述文字对应的第二均值特征向量。S203: For each type of the second attribute description text in the second description file, obtain a preset number of second keywords in each type of the second attribute description text, and convert each of the second keywords A second feature vector is formed, and a preset number of the second feature vectors are averaged to obtain a second average feature vector corresponding to each type of the second attribute description text.

在本申请实施例中，针对第一描述文件中每类第一属性描述文字，获取每类第一属性描述文字中预设数量个第一关键词，将预设数量个第一关键词分别表达成第一特征向量，再对预设数量个第一特征向量求均值，得到每类第一属性描述文字对应的第一均值特征向量。In the embodiment of the present application, for each type of first attribute description text in the first description file, a preset number of first keywords in each type of first attribute description text are obtained, and the preset number of first keywords are separately listed The first feature vector is reached, and the preset number of first feature vectors are averaged to obtain the first average feature vector corresponding to each type of first attribute description text.

在本申请实施例中，预先建立关键词与特征向量的对应关系，对应关系的建立方法如下：In the embodiment of the present application, the corresponding relationship between the keyword and the feature vector is established in advance, and the method for establishing the corresponding relationship is as follows:

首先，通过网络爬虫技术爬取各种渠道发布的求职简历信息和员工评价信息整理成为文档集合。First, the job application resume information and employee evaluation information released by various channels are crawled through web crawler technology to organize them into a document collection.

然后，运用开源的分词工具，对每篇文档进行分词和词性标注，然后根据预设的停用词词典去除停用词，并且根据分词后的词语的词性，去掉介词、方位词和语气词等非特征词，得到关键词集合。Then, use the open source word segmentation tool to perform word segmentation and part-of-speech tagging for each document, and then remove the stop words according to the preset stop word dictionary, and remove the prepositions, localizers and modal particles according to the part of speech of the words after the word segmentation Non-feature words, get the keyword set.

最后，利用开源的词向量训练工具Word2Vec(word to vector)，训练上述关键词集合，得到不同的关键词对应的特征向量，将关键词与特征向量的对应关系进行存储，存储于词向量数据库。示例性的，每个特征向量都具有相同的维度，利用N维(N为正整数)的词向量，每个词向量的数值均在0至1，或-1至1之间。Finally, use the open source word vector training tool Word2Vec (word to vector) to train the above keyword set to obtain feature vectors corresponding to different keywords, store the corresponding relationship between the keywords and feature vectors, and store them in the word vector database. Exemplarily, each feature vector has the same dimension, and an N-dimensional (N is a positive integer) word vector is used, and the value of each word vector is between 0 to 1, or -1 to 1.

通过上述方法建立好了关键词与特征向量的对应关系。通过查找对应关系，就可以获取到第一关键词对应的第一特征向量，从而将第一关键词转化成第一特征向量；同理，获取到第二关键词对应的第二特征向量，从而将第二关键词转化成第二特征向量。Through the above method, the corresponding relationship between keywords and feature vectors is established. By searching the corresponding relationship, the first feature vector corresponding to the first keyword can be obtained, and the first keyword can be converted into the first feature vector; in the same way, the second feature vector corresponding to the second keyword can be obtained, thereby Convert the second keyword into a second feature vector.

将F个(F为正整数)第一关键词表达成F个1×N维(N为正整数)的第一特征向量，再对这F个1×N维的第一特征向量求均值，得到每类第一属性描述文字对应的1×N维第 一均值特征向量。Express the F (F is a positive integer) first keyword into F 1×N-dimensional (N is a positive integer) first eigenvector, and then calculate the average of these F 1×N-dimensional first eigenvectors, The 1×N-dimensional first mean eigenvector corresponding to each type of first attribute description text is obtained.

作为一非限制性示例，针对职业经历这一属性，预设数量F为3，职业经历对应的3个第一特征向量依次为(1，0，0)，(1，1，0)和(0，0，1)，此时第一均值特征向量为((1+1+0)/3，(0+1+0)/3，(0+0+1)/3)＝(0.6667，0.3333，0.3333)。As a non-limiting example, for the attribute of occupational experience, the preset number F is 3, and the three first feature vectors corresponding to occupational experience are (1, 0, 0), (1, 1, 0) and ( 0, 0, 1), at this time the first mean eigenvector is ((1+1+0)/3, (0+1+0)/3, (0+0+1)/3)=(0.6667, 0.3333, 0.3333).

将第一描述文件中的每类第一属性描述文字表达成第一均值特征向量，一方面，将数据处理成可由机器处理的数据，使得本申请能够实施；另一方面，获取每类第一属性描述文字中预设数量个第一关键词，基于预设数量个第一特征向量得到每类第一属性描述文字的特征向量，通过过滤一些噪音，保证结果精度的同时，也适当的减少了数据量，提高了处理效率，减少了***资源占用，降低了算力成本。Express each type of first attribute description text in the first description file as a first mean feature vector. On the one hand, the data is processed into machine-processable data so that this application can be implemented; on the other hand, the first attribute of each type is obtained. There are a preset number of first keywords in the attribute description text, and the feature vector of each type of first attribute description text is obtained based on the preset number of first feature vectors. By filtering some noises, the accuracy of the results is guaranteed and the result is appropriately reduced The amount of data improves processing efficiency, reduces system resource occupation, and reduces computing power costs.

针对第二描述文件中每类第二属性描述文字，也进行与第一属性描述文字相同的过程，以得到每类第二属性描述文字对应的第二均值特征向量。此处不再赘述，请参见上述。For each type of second attribute description text in the second description file, the same process as the first attribute description text is also performed to obtain the second mean value feature vector corresponding to each type of second attribute description text. I won't repeat them here, please refer to the above.

需要说明的是，预设数量为经验值，可以根据实际需要进行选择设置，本申请实施例对此不予具体限制。It should be noted that the preset number is an empirical value, which can be selected and set according to actual needs, and the embodiment of the present application does not specifically limit this.

还需要说明的是，虽然步骤S202和步骤S203在描述上有前后之分，标号也有大小之分，但是描述上的前后之分和标号的大小之分都不代表具体限制了步骤的先后时序关系。在本申请实施例中，步骤S202可以在步骤S203之前执行，还可以在步骤S203之后执行，还可以与步骤S203同时执行，本申请不具体限定步骤S202和S203之间的时序关系。It should also be noted that although steps S202 and S203 are described in front and back, and the labels are also different in size, but the front and back in the description and the size of the labels do not mean that the sequence relationship of the steps is specifically limited. . In the embodiment of the present application, step S202 may be performed before step S203, may also be performed after step S203, or simultaneously with step S203. The present application does not specifically limit the timing relationship between steps S202 and S203.

S204，将所述第一描述文件中多类所述第一属性描述文字对应的第一均值特征向量进行组合，生成第一历史特征矩阵。S204: Combine first average feature vectors corresponding to multiple types of the first attribute description text in the first description file to generate a first historical feature matrix.

通过执行步骤S202，得到了每类第一属性描述文字对应的第一均值特征向量，在步骤S204中，将第一描述文件中多类第一属性描述文字对应的第一均值特征向量进行组合，从而生成第一历史特征矩阵。By executing step S202, the first mean value feature vector corresponding to each type of first attribute description text is obtained. In step S204, the first mean value feature vectors corresponding to the multiple types of first attribute description text in the first description file are combined. Thus, the first historical feature matrix is generated.

其中，将多个第一均值特征向量进行组合，是将多个第一特征向量进行拼接，以生成第一历史特征矩阵。作为一非限制性示例，由M个(M为正整数)1×N维的第一均值特征向量进行拼接，生成M×N维的第一历史特征矩阵。Wherein, combining the multiple first mean eigenvectors is to combine the multiple first eigenvectors to generate the first historical feature matrix. As a non-limiting example, M (M is a positive integer) 1×N-dimensional first mean eigenvectors are spliced to generate an M×N-dimensional first historical feature matrix.

本申请实施例中，通过步骤S202和步骤S204，将第一描述文件表达成了第一历史特征矩阵。In the embodiment of the present application, through step S202 and step S204, the first description file is expressed as a first historical feature matrix.

S205，将所述第二描述文件中每类所述第二属性描述文字对应的第二均值特征向量进行组合，生成第二历史特征矩阵。S205: Combine the second mean feature vectors corresponding to each type of the second attribute description text in the second description file to generate a second historical feature matrix.

通过执行步骤S203，得到了每类第二属性描述文字对应的第二均值特征向量，在步骤S205中，将第二描述文件中多类第二属性描述文字对应的第二均值特征向量进行组合，从而生成第二历史特征矩阵。By performing step S203, the second mean value feature vector corresponding to each type of second attribute description text is obtained. In step S205, the second mean value feature vectors corresponding to the multiple types of second attribute description text in the second description file are combined. Thus, the second historical feature matrix is generated.

其中，将多个第二均值特征向量进行组合，是将多个第二特征向量进行拼接，以生成第二历史特征矩阵。作为一非限制性示例，由M个1×N维的第二均值特征向量进行拼接，生成M×N维的第二历史特征矩阵。Wherein, combining the multiple second mean eigenvectors is to combine the multiple second eigenvectors to generate the second historical feature matrix. As a non-limiting example, M x N-dimensional second mean eigenvectors are spliced to generate an M x N-dimensional second historical feature matrix.

本申请实施例中，通过步骤S203和步骤S205，将第二描述文件表达成了第二历史特征矩阵。In the embodiment of the present application, through step S203 and step S205, the second description file is expressed as a second historical feature matrix.

需要说明的是，虽然步骤S202、步骤S203、步骤S204和步骤S205在描述上有前后之分，标号也有大小之分，但是描述上的前后之分和标号的大小之分都不代表具体限制了四个步骤的先后时序关系。在本申请实施例中，步骤S204可以在步骤S205之前执行，还可以在步骤S205之后执行，还可以与步骤S205同时执行；步骤S204可以在步骤S203之后执行，还可以在步骤S203之前执行，还可以与步骤S203同时执行。本申请限定步骤S204在步骤S202之后执行，步骤S205在步骤S203之后执行，并不具体限定步骤S204和步骤S205之间的时序关系，也不具体限定步骤S202与步骤S203，步骤S202与步骤S205之间的时序关系，也不具体限定步骤S203与步骤S202，步骤S203与步骤S204之间的时序关系。It should be noted that although step S202, step S203, step S204, and step S205 are described in front and back, and the labels are also different in size, but the front and back in the description and the size of the labels do not represent specific restrictions. The sequence of the four steps. In the embodiment of the present application, step S204 may be executed before step S205, may also be executed after step S205, or may be executed simultaneously with step S205; step S204 may be executed after step S203, or may be executed before step S203. It can be executed simultaneously with step S203. This application defines that step S204 is executed after step S202, and step S205 is executed after step S203. It does not specifically limit the timing relationship between step S204 and step S205, nor does it specifically limit step S202 and step S203, and step S202 and step S205. The timing relationship between steps S203 and S202, and the timing relationship between step S203 and step S204 are not specifically limited.

S206，将所述第一历史特征矩阵和所述第二历史特征矩阵组合后输入模型得到所述对象 的行为识别结果。S206: Combine the first historical feature matrix and the second historical feature matrix and input a model to obtain a behavior recognition result of the object.

在本申请实施例中，所述模型为使用多组数据通过机器学习训练得出的，所述多组数据中的每组数据包括第一历史时间段的第一历史特征矩阵样本，第二历史时间段的第二历史特征矩阵样本和行为标签，每种所述行为标签表征一种行为识别结果。In the embodiment of the present application, the model is obtained through machine learning training using multiple sets of data. Each set of data in the multiple sets of data includes a first historical feature matrix sample of a first historical time period, and a second historical time period. The second historical feature matrix sample and behavior label of the time period, each of the behavior labels represents a kind of behavior recognition result.

其中，模型可以为采用使用机器学习中的贝叶斯分类、决策树、随机森林、支持向量机、人工神经网络等分类方法构建的模型。Among them, the model may be a model constructed by using classification methods such as Bayesian classification, decision tree, random forest, support vector machine, artificial neural network, etc. in machine learning.

将第一历史特征矩阵HisTarVecI和第二历史特征矩阵HisTarVecII组合得到组合历史特征矩阵(HisVecI，HisVecII)，将组合历史特征矩阵(HisVecI，HisVecII)输入模型得到对象的行为识别结果。Combine the first historical feature matrix HisTarVecI and the second historical feature matrix HisTarVecII to obtain a combined historical feature matrix (HisVecI, HisVecII), and input the combined historical feature matrix (HisVecI, HisVecII) into the model to obtain the behavior recognition result of the object.

需要说明的是，当第一历史特征矩阵和第一历史特征矩阵均为M×N维，组合得到的矩阵可以为2M×N维，也可以为M×2N维。It should be noted that when the first historical feature matrix and the first historical feature matrix are both M×N dimensions, the combined matrix may be 2M×N dimensions or M×2N dimensions.

行为识别结果具体的形式跟训练模型时使用的行为标签相关，一种行为标签表征一种行为识别结果。若用T个(T为正整数)不同的行为标签标识T种不同的行为识别结果，那么模型输出的行为标签表征的结果，为对象的行为识别结果。The specific form of the behavior recognition result is related to the behavior label used when training the model, and a kind of behavior label represents a kind of behavior recognition result. If T (T is a positive integer) different behavior labels are used to identify T different behavior recognition results, then the result of the behavior label characterization output by the model is the behavior recognition result of the object.

作为一非限制性示例，行为识别结果为用户对新进员工入职一段时间后结合入职简历和入职一段时间内的工作表现进行评价的评价结果。可以用任一一个数值来表示这个评价结果，作为行为识别结果，例如，用1至6这6个数字标签分别表示风险员工，不合格员工，合格员工，一般员工，优良员工和优秀员工。在该非限制性示例中，针入职简历和入职情况这两份历史数据，实现对新进人员的行为评价，提高了评价结果的准确度，帮助企业招聘人员对新进人员做出接受或者拒绝的正确决策。As a non-limiting example, the behavior recognition result is an evaluation result of the user evaluating a new employee after a period of entry, combined with the entry resume and the performance of the entry for a period of time. Any numerical value can be used to represent the evaluation result as a behavior recognition result. For example, the 6 digital labels from 1 to 6 are used to represent risky employees, unqualified employees, qualified employees, general employees, excellent employees, and excellent employees. In this non-limiting example, the two historical data of entry resume and entry status can be used to evaluate the behavior of new recruits, improve the accuracy of the evaluation results, and help corporate recruiters accept or reject new recruits The right decision.

应当理解的是，第一历史特征矩阵样本与第二历史特征矩阵样本的获取可参见前述对第一历史特征矩阵和第二历史特征矩阵的获取过程，两者思路一致。在本申请实施例中，采用多样的大量样本进行模型训练，以得到鲁棒性更好的模型。It should be understood that the acquisition of the first historical feature matrix sample and the second historical feature matrix sample can refer to the aforementioned acquisition process of the first historical feature matrix and the second historical feature matrix, and the ideas of the two are the same. In the embodiments of the present application, a large number of diverse samples are used for model training to obtain a more robust model.

本申请实施例利用机器学习的方式预先训练模型，基于对象两个不同时间段的描述文件，分别提取描述文件的特征矩阵，将两个特征矩阵组合后输入模型获得对象的行为识别结果，提供了一种对象行为的识别方案。一方面，该识别方案利用更多的资料对对象进行行为识别，增大了输入模型的信息量，提高了对象行为识别的精度；另一方面，该识别方案通过将描述文件中多类属性分别对应的特征向量进行组合形成描述文件对应的特征矩阵，提取了高质量的数据，也减少了噪音，在保证识别结果高精度的同时，也减少了数据处理量，减少了***资源占用。The embodiment of this application uses machine learning to pre-train the model, extracts the feature matrix of the description file based on the description files of the object in two different time periods, and combines the two feature matrices into the model to obtain the behavior recognition result of the object. A recognition scheme of object behavior. On the one hand, the recognition scheme uses more data to recognize the behavior of objects, increases the amount of information input to the model, and improves the accuracy of object behavior recognition; on the other hand, the recognition scheme separates multiple types of attributes in the description file. The corresponding feature vectors are combined to form the feature matrix corresponding to the description file, which extracts high-quality data and reduces noise. While ensuring the high accuracy of the recognition result, it also reduces the amount of data processing and reduces the system resource occupation.

在上述图2所示实施例的基础上，本申请实施例提供了另一种对象的行为识别方法，本申请实施例在图2所示的实施例的基础上，对步骤S202中，获取每类所述第一属性描述文字中预设数量个第一关键词，进行了具体优化。如图3所示，获取每类所述第一属性描述文字中预设数量个第一关键词，包括步骤S301至步骤S303。On the basis of the above-mentioned embodiment shown in FIG. 2, the embodiment of the present application provides another object behavior recognition method. The embodiment of the present application is based on the embodiment shown in FIG. The preset number of first keywords in the description text of the first attribute of the class is specifically optimized. As shown in FIG. 3, obtaining a preset number of first keywords in each type of the first attribute description text includes step S301 to step S303.

S301，对每类所述第一属性描述文字进行分词、去停用词和去非特征词处理，得到所述第一属性描述文字对应的第一关键词集合。S301: Perform word segmentation, remove stop words and remove non-characteristic words for each type of the first attribute description text to obtain a first keyword set corresponding to the first attribute description text.

针对第一描述文档中每类第一属性描述文字，先进行分词和词性标注，然后根据预设的停用词词典去除停用词，并且根据分词后的词语的词性，去掉介词、方位词和语气词等非特征词，得到每类第一属性描述文字对应的第二关键词集合。For each type of first attribute description text in the first description document, first perform word segmentation and part-of-speech tagging, then remove the stop words according to the preset stop word dictionary, and remove the prepositions, localizers and words according to the part of speech of the words after the word segmentation. For non-characteristic words such as modal particles, the second keyword set corresponding to each type of first attribute description text is obtained.

S302，计算所述第一关键词集合中每个第一关键词的相关度。S302: Calculate the relevance of each first keyword in the first keyword set.

在本申请实施例中，所述相关度表征所述第一关键词与所述第一关键词集合中其他第一关键词之间的关联程度。In the embodiment of the present application, the degree of relevance represents the degree of relevance between the first keyword and other first keywords in the first keyword set.

具体地，计算所述第一关键词集合中每个第一关键词的相关度，包括：针对所述第一关键词集合中每个第一关键词，分别获取所述第一关键词与所述第一关键词集合中其他第一关键词之间的相关度；对所述第一关键词与其他第一关键词之间的相关度求和，作为所述第一关键词集合中每个所述第一关键词的相关度。Specifically, calculating the relevance of each first keyword in the first keyword set includes: for each first keyword in the first keyword set, respectively obtaining the first keyword and all the keywords. The relevance between the other first keywords in the first keyword set; sum the relevance between the first keyword and the other first keywords as each of the first keyword sets The relevance of the first keyword.

其中，第一关键词集合中每个第一关键词的相关度，等于每个第一关键词与每一个其他第一关键词的相关度之和。Wherein, the relevance of each first keyword in the first keyword set is equal to the sum of the relevance of each first keyword and every other first keyword.

也就是说，第i个第一关键字与其他第一关键字的相关度RelKeywordi的计算公式为：In other words, the calculation formula of RelKeywordi between the i-th first keyword and other first keywords is:

其中，RelKeywordi,j表示第i个第一关键字与第j个第一关键字的相关度，i和j取值为1至W，且不等于i，W为正整数，W表示第一关键词集合中第一关键词的总数量。Among them, RelKeywordi,j represents the degree of relevance between the i-th first keyword and the j-th first keyword, i and j range from 1 to W, and are not equal to i, W is a positive integer, and W represents the first key The total number of the first keyword in the word set.

可选地，第i个第一关键字与第j个第一关键字的相关度RelKeywordi,j的计算方式为：Optionally, the calculation method of the relevance RelKeywordi,j between the i-th first keyword and the j-th first keyword is:

其中，NumProSeni为第i个第一关键词所在的第一属性描述文字的总句子数，NumProSenj为第j个第一关键词所在的第一属性描述文字的总句子数，可以以逗号断句或者句号断句作为一句。显然的，由于第i个第一关键字与第j个第一关键字对应同一个第一属性描述文字，因而NumProSeni等于NumProSenj。Among them, NumProSeni is the total number of sentences of the first attribute description text where the i-th first keyword is located, and NumProSenj is the total number of sentences of the first attribute description text where the j-th first keyword is located. It can be a comma or a period. Sentence as a sentence. Obviously, since the i-th first keyword and the j-th first keyword correspond to the same first attribute description text, NumProSeni is equal to NumProSenj.

NumKeywordSeni为第一属性描述文字的总句子数中出现第i个第一关键词的句子的句子次数。NumKeywordSeni is the number of sentences with the i-th first keyword in the total sentences of the first attribute description text.

NumKeywordSenj为第一属性描述文字的总句子数中出现第j个第一关键词的句子的句子次数。NumKeywordSenj is the number of sentences in which the j-th sentence of the first keyword appears in the total number of sentences of the first attribute description text.

NumKeywordSeni,j为第一属性描述文字的总句子数中同时出现第i个和第j个第一关键词的句子的句子次数。NumKeywordSeni,j is the number of sentences in which the i-th and j-th sentences of the first keyword appear simultaneously in the total number of sentences of the first attribute description text.

S303，将所述第一关键词集合中相关度排名靠前的预设数量个第一关键词，作为每类所述第一属性描述文字对应的预设数量个第一关键词。S303: Use a preset number of first keywords with a high relevance ranking in the first keyword set as a preset number of first keywords corresponding to each type of the first attribute description text.

在本申请实施例中，将相关度排名靠前的预设数量个，例如N个，第一关键词，作为每类第一属性描述文字对应的N个第一关键词。提取出了文件中的关键信息，也减少了噪声数据，保证后续识别结果精度的同时，也减少了数据处理量，减少了***资源占用，此外，提供了一种定量的筛选关键词的方式，使得本申请实施例易于实现。In this embodiment of the present application, a preset number of top relevance rankings, such as N, first keywords, are used as the N first keywords corresponding to each type of first attribute description text. The key information in the file is extracted and the noise data is reduced. While ensuring the accuracy of subsequent recognition results, it also reduces the amount of data processing and the system resource occupation. In addition, it provides a quantitative way to filter keywords. This makes the embodiments of the present application easy to implement.

可以理解的，基于上述图3所示实施例，同理，图2所示实施例的步骤S203中，获取每类所述第二属性描述文字中预设数量个第二关键词，如图4所示，包括步骤S401至步骤S403。It is understandable that, based on the embodiment shown in FIG. 3, and similarly, in step S203 of the embodiment shown in FIG. 2, a preset number of second keywords in each type of the second attribute description text are obtained, as shown in FIG. 4 As shown, it includes step S401 to step S403.

S401，对每类所述第二属性描述文字进行分词、去停用词和去非特征词处理，得到所述第二属性描述文字对应的第二关键词集合。S401: Perform word segmentation, remove stop words and remove non-characteristic words for each type of the second attribute description text to obtain a second keyword set corresponding to the second attribute description text.

S402，计算所述第二关键词集合中每个第二关键词的相关度。S402: Calculate the relevance of each second keyword in the second keyword set.

其中，所述相关度表征所述第二关键词与所述第二关键词集合中其他第二关键词之间的关联程度。Wherein, the degree of relevance represents the degree of relevance between the second keyword and other second keywords in the second keyword set.

可选地，计算所述第二关键词集合中每个第二关键词的相关度，包括：Optionally, calculating the relevance of each second keyword in the second keyword set includes:

针对所述第二关键词集合中每个第二关键词，分别获取所述第二关键词与所述第二关键词集合中其他第二关键词之间的相关度；对所述第二关键词与其他第二关键词之间的相关度求和，作为所述第二关键词集合中每个所述第二关键词的相关度。For each second keyword in the second keyword set, the relevance between the second keyword and other second keywords in the second keyword set is obtained; for the second key The sum of the relevance between a word and other second keywords is used as the relevance of each second keyword in the second keyword set.

S403，将所述第二关键词集合中相关度排名靠前的预设数量个第二关键词，作为每类所述第二属性描述文字对应的预设数量个第二关键词。S403: Use a preset number of second keywords with a high relevance ranking in the second keyword set as a preset number of second keywords corresponding to each type of the second attribute description text.

应理解，上述实施例中各步骤的序号的大小并不意味着执行顺序的先后，各过程的执行顺序应以其功能和内在逻辑确定，而不应对本申请实施例的实施过程构成任何限定。It should be understood that the size of the sequence number of each step in the foregoing embodiment does not mean the order of execution. The execution sequence of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present application.

对应于上文实施例所述的对象行为的识别方法，图5示出了本申请实施例提供的对象行为的识别装置的结构框图，为了便于说明，仅示出了与本申请实施例相关的部分。Corresponding to the object behavior recognition method described in the above embodiment, FIG. 5 shows a structural block diagram of the object behavior recognition device provided in an embodiment of the present application. For ease of description, only the information related to the embodiment of the present application is shown. section.

参照图5，该装置包括：Referring to Figure 5, the device includes:

文档获取模块51、特征向量获取模块52、矩阵生成模块53和行为识别模块54； Document acquisition module 51, feature vector acquisition module 52, matrix generation module 53, and behavior recognition module 54;

所述文档获取模块51，用于：The document obtaining module 51 is configured to:

获取对象在第一历史时间段的第一描述文件和在第二历史时间段的第二描述文件；其中，所述第一描述文件包括多类第一属性描述文字，每类所述第一属性描述文字针对所述对象在所述第一历史时间段的一个属性进行描述；所述第二描述文件包括多类第二属性描述文字，每类所述第二属性描述文字针对所述对象在所述第二历史时间段的一个属性进行描述；Acquire the first description file of the object in the first historical time period and the second description file in the second historical time period; wherein, the first description file includes multiple types of first attribute description text, and each type of the first attribute The description text describes one attribute of the object in the first historical time period; the second description file includes multiple types of second attribute description text, and each type of the second attribute description text is specific to the object in the place. Describe an attribute of the second historical time period;

所述特征向量获取模块52，用于：The feature vector obtaining module 52 is configured to:

针对所述第一描述文件中每类所述第一属性描述文字，获取每类所述第一属性描述文字中预设数量个第一关键词，将每个所述第一关键词转化成第一特征向量，对预设数量个所述第一特征向量求均值，得到每类所述第一属性描述文字对应的第一均值特征向量；For each type of the first attribute description text in the first description file, obtain a preset number of first keywords in each type of the first attribute description text, and convert each of the first keywords into a A feature vector, averaging a preset number of the first feature vectors to obtain the first average feature vector corresponding to each type of the first attribute description text;

针对所述第二描述文件中每类所述第二属性描述文字，获取每类所述第二属性描述文字中预设数量个第二关键词，将每个所述第二关键词转化成第二特征向量，对预设数量个所述第二特征向量求均值，得到每类所述第二属性描述文字对应的第二均值特征向量；For each type of the second attribute description text in the second description file, obtain a preset number of second keywords in each type of the second attribute description text, and convert each of the second keywords into the first Two feature vectors, averaging a preset number of the second feature vectors to obtain a second average feature vector corresponding to each type of the second attribute description text;

所述矩阵生成模块53，用于：The matrix generation module 53 is used for:

将所述第一描述文件中多类所述第一属性描述文字对应的第一均值特征向量进行组合，生成第一历史特征矩阵；Combining the first mean value feature vectors corresponding to the multiple types of the first attribute description text in the first description file to generate a first historical feature matrix;

将所述第二描述文件中每类所述第二属性描述文字对应的第二均值特征向量进行组合，生成第二历史特征矩阵；Combine the second mean eigenvectors corresponding to each type of the second attribute description text in the second description file to generate a second historical feature matrix;

所述行为识别模块54，用于：The behavior recognition module 54 is used to:

将所述第一历史特征矩阵和所述第二历史特征矩阵组合后输入模型得到所述对象的行为识别结果，其中，所述模型为使用多组数据通过机器学习训练得出的，所述多组数据中的每组数据包括第一历史时间段的第一历史特征矩阵样本，第二历史时间段的第二历史特征矩阵样本和行为标签，每种所述行为标签表征一种行为识别结果。The first historical feature matrix and the second historical feature matrix are combined and input into a model to obtain the behavior recognition result of the object, wherein the model is obtained through machine learning training using multiple sets of data, and the multiple Each group of data in the group of data includes a first historical feature matrix sample in a first historical time period, a second historical feature matrix sample in a second historical time period, and behavior labels, each of the behavior labels represents a behavior recognition result.

需要说明的是，上述装置/单元之间的信息交互、执行过程等内容，由于与本申请方法实施例基于同一构思，其具体功能及带来的技术效果，具体可参见方法实施例部分，此处不再赘述。It should be noted that the information interaction and execution process between the above-mentioned devices/units are based on the same concept as the method embodiment of this application, and its specific functions and technical effects can be found in the method embodiment section. I won't repeat it here.

所属领域的技术人员可以清楚地了解到，为了描述的方便和简洁，仅以上述各功能单元、模块的划分进行举例说明，实际应用中，可以根据需要而将上述功能分配由不同的功能单元、模块完成，即将所述装置的内部结构划分成不同的功能单元或模块，以完成以上描述的全部或者部分功能。实施例中的各功能单元、模块可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中，上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。另外，各功能单元、模块的具体名称也只是为了便于相互区分，并不用于限制本申请的保护范围。上述***中单元、模块的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。Those skilled in the art can clearly understand that, for the convenience and conciseness of description, only the division of the above functional units and modules is used as an example. In practical applications, the above functions can be allocated to different functional units and modules as needed. Module completion, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist alone physically, or two or more units can be integrated into one unit. The above-mentioned integrated units can be hardware-based Formal realization can also be realized in the form of software functional units. In addition, the specific names of the functional units and modules are only used to facilitate distinguishing each other, and are not used to limit the protection scope of the present application. For the specific working process of the units and modules in the foregoing system, reference may be made to the corresponding process in the foregoing method embodiment, which will not be repeated here.

图6为本申请一实施例提供的终端设备的结构示意图。如图6所示，该实施例的终端设备6包括：至少一个处理器60(图6中仅示出一个处理器)、存储器61以及存储在所述存储器61中并可在所述至少一个处理器60上运行的计算机程序62，所述处理器60执行所述计算机程序62时实现上述各个方法实施例中的步骤。FIG. 6 is a schematic structural diagram of a terminal device provided by an embodiment of this application. As shown in FIG. 6, the terminal device 6 of this embodiment includes: at least one processor 60 (only one processor is shown in FIG. 6), a memory 61, and a memory 61 that is stored in the memory 61 and can be processed in the at least one processor. The computer program 62 running on the processor 60 implements the steps in the foregoing method embodiments when the processor 60 executes the computer program 62.

本申请实施例还提供了一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，所述计算机程序被处理器执行时实现可实现上述各个方法实施例中的步骤。The embodiments of the present application also provide a computer-readable storage medium, where the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in each of the foregoing method embodiments can be realized.

本申请实施例提供了一种计算机程序产品，当计算机程序产品在移动终端上运行时，使 得移动终端执行时实现可实现上述各个方法实施例中的步骤。The embodiments of the present application provide a computer program product. When the computer program product runs on a mobile terminal, the steps in the foregoing method embodiments can be realized when the mobile terminal is executed.

所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请实现上述实施例方法中的全部或部分流程，可以通过计算机程序来指令相关的硬件来完成，所述的计算机程序可存储于一计算机可读存储介质中，该计算机程序在被处理器执行时，可实现上述各个方法实施例的步骤。其中，所述计算机程序包括计算机程序代码，所述计算机程序代码可以为源代码形式、对象代码形式、可执行文件或某些中间形式等。所述计算机可读介质至少可以包括：能够将计算机程序代码携带到拍照装置/终端设备的任何实体或装置、记录介质、计算机存储器、只读存储器(Read-Only Memory，ROM)、随机存取存储器(Random Access Memory，RAM)、电载波信号、电信信号以及软件分发介质。例如U盘、移动硬盘、磁碟或者光盘等。在某些司法管辖区，根据立法和专利实践，计算机可读介质不可以是电载波信号和电信信号。If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the implementation of all or part of the processes in the above-mentioned embodiment methods in this application can be accomplished by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When executed by the processor, the steps of the foregoing method embodiments can be implemented. Wherein, the computer program includes computer program code, and the computer program code may be in the form of source code, object code, executable file, or some intermediate forms. The computer-readable medium may at least include: any entity or device capable of carrying computer program code to the photographing device/terminal device, recording medium, computer memory, read-only memory (ROM), random access memory (Random Access Memory, RAM), electric carrier signal, telecommunications signal, and software distribution medium. Such as U disk, mobile hard disk, floppy disk or CD-ROM, etc. In some jurisdictions, according to legislation and patent practices, computer-readable media cannot be electrical carrier signals and telecommunication signals.

在上述实施例中，对各个实施例的描述都各有侧重，某个实施例中没有详述或记载的部分，可以参见其它实施例的相关描述。In the above-mentioned embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail or recorded in an embodiment, reference may be made to related descriptions of other embodiments.

本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。A person of ordinary skill in the art may realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software depends on the specific application and design constraint conditions of the technical solution. Professionals and technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered beyond the scope of this application.

在本申请所提供的实施例中，应该理解到，所揭露的装置/网络设备和方法，可以通过其它的方式实现。例如，以上所描述的装置/网络设备实施例仅仅是示意性的，例如，所述模块或单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个***，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通讯连接可以是通过一些接口，装置或单元的间接耦合或通讯连接，可以是电性，机械或其它的形式。In the embodiments provided in this application, it should be understood that the disclosed apparatus/network equipment and method may be implemented in other ways. For example, the device/network device embodiments described above are only illustrative. For example, the division of the modules or units is only a logical function division, and there may be other divisions in actual implementation, such as multiple units. Or components can be combined or integrated into another system, or some features can be omitted or not implemented. In addition, the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical, mechanical or other forms.

所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.

以上所述实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的精神和范围，均应包含在本申请的保护范围之内。The above-mentioned embodiments are only used to illustrate the technical solutions of the present application, not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, a person of ordinary skill in the art should understand that it can still implement the foregoing The technical solutions recorded in the examples are modified, or some of the technical features are equivalently replaced; these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the application, and should be included in Within the scope of protection of this application.

Claims (20)

1. 一种对象行为的识别方法，包括：A method of identifying object behavior, including:

   获取对象在第一历史时间段的第一描述文件和在第二历史时间段的第二描述文件；其中，所述第一描述文件包括多类第一属性描述文字，每类所述第一属性描述文字针对所述对象在所述第一历史时间段的一个属性进行描述；所述第二描述文件包括多类第二属性描述文字，每类所述第二属性描述文字针对所述对象在所述第二历史时间段的一个属性进行描述；Acquire the first description file of the object in the first historical time period and the second description file in the second historical time period; wherein, the first description file includes multiple types of first attribute description text, and each type of the first attribute The description text describes one attribute of the object in the first historical time period; the second description file includes multiple types of second attribute description text, and each type of the second attribute description text is specific to the object in the place. Describe an attribute of the second historical time period;

   针对所述第一描述文件中每类所述第一属性描述文字，获取每类所述第一属性描述文字中预设数量个第一关键词，将每个所述第一关键词转化成第一特征向量，对预设数量个所述第一特征向量求均值，得到每类所述第一属性描述文字对应的第一均值特征向量；For each type of the first attribute description text in the first description file, obtain a preset number of first keywords in each type of the first attribute description text, and convert each of the first keywords into a A feature vector, averaging a preset number of the first feature vectors to obtain the first average feature vector corresponding to each type of the first attribute description text;

   针对所述第二描述文件中每类所述第二属性描述文字，获取每类所述第二属性描述文字中预设数量个第二关键词，将每个所述第二关键词转化成第二特征向量，对预设数量个所述第二特征向量求均值，得到每类所述第二属性描述文字对应的第二均值特征向量；For each type of the second attribute description text in the second description file, obtain a preset number of second keywords in each type of the second attribute description text, and convert each of the second keywords into the first Two feature vectors, averaging a preset number of the second feature vectors to obtain a second average feature vector corresponding to each type of the second attribute description text;

   将所述第一描述文件中多类所述第一属性描述文字对应的第一均值特征向量进行组合，生成第一历史特征矩阵；Combining the first mean value feature vectors corresponding to the multiple types of the first attribute description text in the first description file to generate a first historical feature matrix;

   将所述第二描述文件中每类所述第二属性描述文字对应的第二均值特征向量进行组合，生成第二历史特征矩阵；Combine the second mean eigenvectors corresponding to each type of the second attribute description text in the second description file to generate a second historical feature matrix;

   将所述第一历史特征矩阵和所述第二历史特征矩阵组合后输入模型得到所述对象的行为识别结果，其中，所述模型为使用多组数据通过机器学习训练得出的，所述多组数据中的每组数据包括第一历史时间段的第一历史特征矩阵样本，第二历史时间段的第二历史特征矩阵样本和行为标签，每种所述行为标签表征一种行为识别结果。The first historical feature matrix and the second historical feature matrix are combined and input into a model to obtain the behavior recognition result of the object, wherein the model is obtained through machine learning training using multiple sets of data, and the multiple Each group of data in the group of data includes a first historical feature matrix sample in a first historical time period, a second historical feature matrix sample in a second historical time period, and behavior labels, each of the behavior labels represents a behavior recognition result.
2. 如权利要求1所述的识别方法，其中，所述获取每类所述第一属性描述文字中预设数量个第一关键词，包括：5. The recognition method according to claim 1, wherein said obtaining a preset number of first keywords in each type of said first attribute description text comprises:

   对每类所述第一属性描述文字进行分词、去停用词和去非特征词处理，得到所述第一属性描述文字对应的第一关键词集合；Performing word segmentation, removing stop words and removing non-characteristic words for each type of the first attribute description text to obtain the first keyword set corresponding to the first attribute description text;

   计算所述第一关键词集合中每个第一关键词的相关度；所述相关度表征所述第一关键词与所述第一关键词集合中其他第一关键词之间的关联程度；Calculate the relevance of each first keyword in the first keyword set; the relevance represents the degree of association between the first keyword and other first keywords in the first keyword set;

   将所述第一关键词集合中相关度排名靠前的预设数量个第一关键词，作为每类所述第一属性描述文字对应的预设数量个第一关键词；Taking the preset number of first keywords with the highest relevance ranking in the first keyword set as the preset number of first keywords corresponding to each type of the first attribute description text;

   所述获取每类所述第二属性描述文字中预设数量个第二关键词，包括：The acquiring a preset number of second keywords in the description text of each type of the second attribute includes:

   对每类所述第二属性描述文字进行分词、去停用词和去非特征词处理，得到所述第二属性描述文字对应的第二关键词集合；Performing word segmentation, removing stop words and removing non-characteristic words for each type of the second attribute description text to obtain a second keyword set corresponding to the second attribute description text;

   计算所述第二关键词集合中每个第二关键词的相关度；所述相关度表征所述第二关键词与所述第二关键词集合中其他第二关键词之间的关联程度；Calculate the relevance of each second keyword in the second keyword set; the relevance represents the degree of association between the second keyword and other second keywords in the second keyword set;

   将所述第二关键词集合中相关度排名靠前的预设数量个第二关键词，作为每类所述第二属性描述文字对应的预设数量个第二关键词。The preset number of second keywords with the highest relevance ranking in the second keyword set are used as the preset number of second keywords corresponding to each type of the second attribute description text.
3. 如权利要求2所述的识别方法，其中，所述计算所述第一关键词集合中每个第一关键词的相关度，包括：3. The recognition method of claim 2, wherein the calculating the relevance of each first keyword in the first keyword set comprises:

   针对所述第一关键词集合中每个第一关键词，分别获取所述第一关键词与所述第一关键词集合中其他第一关键词之间的相关度；对所述第一关键词与其他第一关键词之间的相关度求和，作为所述第一关键词集合中每个所述第一关键词的相关度；For each first keyword in the first keyword set, the correlation between the first keyword and other first keywords in the first keyword set is obtained; for the first keyword The sum of the relevance between a word and other first keywords is used as the relevance of each of the first keywords in the first keyword set;

   所述计算所述第二关键词集合中每个第二关键词的相关度，包括：The calculating the relevance of each second keyword in the second keyword set includes:

   针对所述第二关键词集合中每个第二关键词，分别获取所述第二关键词与所述第二关键词集合中其他第二关键词之间的相关度；对所述第二关键词与其他第二关键词之间的相关度求和，作为所述第二关键词集合中每个所述第二关键词的相关度。For each second keyword in the second keyword set, the relevance between the second keyword and other second keywords in the second keyword set is obtained; for the second key The sum of the relevance between a word and other second keywords is used as the relevance of each second keyword in the second keyword set.
4. 如权利要求1所述的识别方法，其中，所述将每个所述第一关键词转化成第一特征向量，包括：5. The recognition method according to claim 1, wherein said converting each of said first keywords into a first feature vector comprises:

   从预先建立的关键词与特征向量的对应关系中，获取每个所述第一关键词对应的第一特征向量；Acquiring the first feature vector corresponding to each of the first keywords from the pre-established correspondence between keywords and feature vectors;

   所述将每个所述第二关键词转化成第二特征向量，包括：The converting each of the second keywords into a second feature vector includes:

   从预先建立的关键词与特征向量的对应关系中，获取每个所述第二关键词对应的第二特征向量。From the pre-established correspondence between keywords and feature vectors, a second feature vector corresponding to each of the second keywords is obtained.
5. 如权利要求3所述的识别方法，其中，通过以下公式获取第i个第一关键词与第j个第一关键词之间的相关度RelKeywordi,j：The recognition method according to claim 3, wherein the correlation degree RelKeywordi,j between the i-th first keyword and the j-th first keyword is obtained by the following formula:

   

   

   其中，NumProSeni为第i个第一关键词所在的第一属性描述文字的总句子数；Among them, NumProSeni is the total number of sentences of the first attribute description text where the i-th first keyword is located;

   NumProSenj为第j个第一关键词所在的第一属性描述文字的总句子数，NumProSeni等于NumProSenj；NumProSenj is the total number of sentences of the first attribute description text where the j-th first keyword is located, and NumProSeni is equal to NumProSenj;

   NumKeywordSeni为所述第一属性描述文字的总句子数中出现第i个第一关键词的句子的句子次数；NumKeywordSeni is the number of sentences in which the sentence of the i-th first keyword appears in the total number of sentences of the first attribute description text;

   NumKeywordSenj为所述第一属性描述文字的总句子数中出现第j个第一关键词的句子的句子次数；NumKeywordSenj is the number of sentences in which the j-th sentence of the first keyword appears in the total number of sentences of the first attribute description text;

   NumKeywordSeni,j为所述第一属性描述文字的总句子数中同时出现第i个和第j个第一关键词的句子的句子次数。NumKeywordSeni,j is the number of sentences in which the i-th and j-th sentences of the first keyword appear simultaneously in the total number of sentences of the first attribute description text.
6. 如权利要求3所述的识别方法，其中，通过以下公式获取第k个第二关键词与第l个第二关键词之间的相关度RelKeywordk,l：The recognition method according to claim 3, wherein the correlation degree RelKeywordk,l between the kth second keyword and the lth second keyword is obtained by the following formula:

   

   

   其中，NumProSenk为第k个第二关键词所在的第二属性描述文字的总句子数；Among them, NumProSenk is the total number of sentences of the second attribute description text where the k-th second keyword is located;

   NumProSenl为第l个第二关键词所在的第二属性描述文字的总句子数，NumProSenk等于NumProSenl；NumProSenl is the total number of sentences of the second attribute description text where the lth second keyword is located, and NumProSenk is equal to NumProSenl;

   NumKeywordSenk为所述第二属性描述文字的总句子数中出现第k个第二关键词的句子的句子次数；NumKeywordSenk is the number of sentences in which the k-th second keyword appears in the total number of sentences of the second attribute description text;

   NumKeywordSenl为所述第二属性描述文字的总句子数中出现第l个第二关键词的句子的句子次数；NumKeywordSenl is the number of sentences in which the sentence of the l-th second keyword appears in the total number of sentences of the second attribute description text;

   NumKeywordSenk,l为所述第二属性描述文字的总句子数中同时出现第k个和第l个第二关键词的句子的句子次数。NumKeywordSenk,l is the number of sentences in which the k-th and l-th sentences of the second keyword appear simultaneously in the total number of sentences of the second attribute description text.
7. 一种对象行为的识别装置，包括：文档获取模块、特征向量获取模块、矩阵生成模块和行为识别模块；An object behavior recognition device, including: a document acquisition module, a feature vector acquisition module, a matrix generation module, and a behavior recognition module;

   所述文档获取模块，用于：The document acquisition module is used to:

   获取对象在第一历史时间段的第一描述文件和在第二历史时间段的第二描述文件；其中，所述第一描述文件包括多类第一属性描述文字，每类所述第一属性描述文字针对所述对象在所述第一历史时间段的一个属性进行描述；所述第二描述文件包括多类第二属性描述文字，每类所述第二属性描述文字针对所述对象在所述第二历史时间段的一个属性进行描述；Acquire the first description file of the object in the first historical time period and the second description file in the second historical time period; wherein, the first description file includes multiple types of first attribute description text, and each type of the first attribute The description text describes one attribute of the object in the first historical time period; the second description file includes multiple types of second attribute description text, and each type of the second attribute description text is specific to the object in the place. Describe an attribute of the second historical time period;

   所述特征向量获取模块，用于：The feature vector acquisition module is used to:

   针对所述第一描述文件中每类所述第一属性描述文字，获取每类所述第一属性描述文字中预设数量个第一关键词，将每个所述第一关键词转化成第一特征向量，对预设数量个所述第一特征向量求均值，得到每类所述第一属性描述文字对应的第一均值特征向量；For each type of the first attribute description text in the first description file, obtain a preset number of first keywords in each type of the first attribute description text, and convert each of the first keywords into a A feature vector, averaging a preset number of the first feature vectors to obtain the first average feature vector corresponding to each type of the first attribute description text;

   针对所述第二描述文件中每类所述第二属性描述文字，获取每类所述第二属性描述文字中预设数量个第二关键词，将每个所述第二关键词转化成第二特征向量，对预设数量个所述第二特征向量求均值，得到每类所述第二属性描述文字对应的第二均值特征向量；For each type of the second attribute description text in the second description file, obtain a preset number of second keywords in each type of the second attribute description text, and convert each of the second keywords into the first Two feature vectors, averaging a preset number of the second feature vectors to obtain a second average feature vector corresponding to each type of the second attribute description text;

   所述矩阵生成模块，用于：The matrix generation module is used to:

   将所述第一描述文件中多类所述第一属性描述文字对应的第一均值特征向量进行组合，生成第一历史特征矩阵；Combining the first mean value feature vectors corresponding to the multiple types of the first attribute description text in the first description file to generate a first historical feature matrix;

   将所述第二描述文件中每类所述第二属性描述文字对应的第二均值特征向量进行组合，生成第二历史特征矩阵；Combine the second mean eigenvectors corresponding to each type of the second attribute description text in the second description file to generate a second historical feature matrix;

   所述行为识别模块，用于：The behavior recognition module is used to:

   将所述第一历史特征矩阵和所述第二历史特征矩阵组合后输入模型得到所述对象的行为识别结果，其中，所述模型为使用多组数据通过机器学习训练得出的，所述多组数据中的每组数据包括第一历史时间段的第一历史特征矩阵样本，第二历史时间段的第二历史特征矩阵样本和行为标签，每种所述行为标签表征一种行为识别结果。The first historical feature matrix and the second historical feature matrix are combined and input into a model to obtain the behavior recognition result of the object, wherein the model is obtained through machine learning training using multiple sets of data, and the multiple Each group of data in the group of data includes a first historical feature matrix sample in a first historical time period, a second historical feature matrix sample in a second historical time period, and behavior labels, each of the behavior labels represents a behavior recognition result.
8. 如权利要求7所述的识别装置，其中，所述获取每类所述第一属性描述文字中预设数量个第一关键词，包括：8. The recognition device according to claim 7, wherein said obtaining a preset number of first keywords in each type of said first attribute description text comprises:

   对每类所述第一属性描述文字进行分词、去停用词和去非特征词处理，得到所述第一属性描述文字对应的第一关键词集合；Performing word segmentation, removing stop words and removing non-characteristic words for each type of the first attribute description text to obtain the first keyword set corresponding to the first attribute description text;

   计算所述第一关键词集合中每个第一关键词的相关度；所述相关度表征所述第一关键词与所述第一关键词集合中其他第一关键词之间的关联程度；Calculate the relevance of each first keyword in the first keyword set; the relevance represents the degree of association between the first keyword and other first keywords in the first keyword set;

   将所述第一关键词集合中相关度排名靠前的预设数量个第一关键词，作为每类所述第一属性描述文字对应的预设数量个第一关键词；Taking the preset number of first keywords with the highest relevance ranking in the first keyword set as the preset number of first keywords corresponding to each type of the first attribute description text;

   所述获取每类所述第二属性描述文字中预设数量个第二关键词，包括：The acquiring a preset number of second keywords in the description text of each type of the second attribute includes:

   对每类所述第二属性描述文字进行分词、去停用词和去非特征词处理，得到所述第二属性描述文字对应的第二关键词集合；Performing word segmentation, removing stop words and removing non-characteristic words for each type of the second attribute description text to obtain a second keyword set corresponding to the second attribute description text;

   计算所述第二关键词集合中每个第二关键词的相关度；所述相关度表征所述第二关键词与所述第二关键词集合中其他第二关键词之间的关联程度；Calculate the relevance of each second keyword in the second keyword set; the relevance represents the degree of association between the second keyword and other second keywords in the second keyword set;

   将所述第二关键词集合中相关度排名靠前的预设数量个第二关键词，作为每类所述第二属性描述文字对应的预设数量个第二关键词。The preset number of second keywords with the highest relevance ranking in the second keyword set are used as the preset number of second keywords corresponding to each type of the second attribute description text.
9. 一种终端设备，包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现如下步骤：A terminal device includes a memory, a processor, and a computer program that is stored in the memory and can run on the processor, and the processor implements the following steps when the processor executes the computer program:

   获取对象在第一历史时间段的第一描述文件和在第二历史时间段的第二描述文件；其中，所述第一描述文件包括多类第一属性描述文字，每类所述第一属性描述文字针对所述对象在所述第一历史时间段的一个属性进行描述；所述第二描述文件包括多类第二属性描述文字，每类所述第二属性描述文字针对所述对象在所述第二历史时间段的一个属性进行描述；Acquire the first description file of the object in the first historical time period and the second description file in the second historical time period; wherein, the first description file includes multiple types of first attribute description text, and each type of the first attribute The description text describes one attribute of the object in the first historical time period; the second description file includes multiple types of second attribute description text, and each type of the second attribute description text is specific to the object in the place. Describe an attribute of the second historical time period;

   针对所述第一描述文件中每类所述第一属性描述文字，获取每类所述第一属性描述文字中预设数量个第一关键词，将每个所述第一关键词转化成第一特征向量，对预设数量个所述第一特征向量求均值，得到每类所述第一属性描述文字对应的第一均值特征向量；For each type of the first attribute description text in the first description file, obtain a preset number of first keywords in each type of the first attribute description text, and convert each of the first keywords into a A feature vector, averaging a preset number of the first feature vectors to obtain the first average feature vector corresponding to each type of the first attribute description text;

   针对所述第二描述文件中每类所述第二属性描述文字，获取每类所述第二属性描述文字中预设数量个第二关键词，将每个所述第二关键词转化成第二特征向量，对预设数量个所述第二特征向量求均值，得到每类所述第二属性描述文字对应的第二均值特征向量；For each type of the second attribute description text in the second description file, obtain a preset number of second keywords in each type of the second attribute description text, and convert each of the second keywords into the first Two feature vectors, averaging a preset number of the second feature vectors to obtain a second average feature vector corresponding to each type of the second attribute description text;

   将所述第一描述文件中多类所述第一属性描述文字对应的第一均值特征向量进行组合，生成第一历史特征矩阵；Combining the first mean value feature vectors corresponding to the multiple types of the first attribute description text in the first description file to generate a first historical feature matrix;

   将所述第二描述文件中每类所述第二属性描述文字对应的第二均值特征向量进行组合，生成第二历史特征矩阵；Combine the second mean eigenvectors corresponding to each type of the second attribute description text in the second description file to generate a second historical feature matrix;

   将所述第一历史特征矩阵和所述第二历史特征矩阵组合后输入模型得到所述对象的行为识别结果，其中，所述模型为使用多组数据通过机器学习训练得出的，所述多组数据中的每组数据包括第一历史时间段的第一历史特征矩阵样本，第二历史时间段的第二历史特征矩阵样本和行为标签，每种所述行为标签表征一种行为识别结果。The first historical feature matrix and the second historical feature matrix are combined and input into a model to obtain the behavior recognition result of the object, wherein the model is obtained through machine learning training using multiple sets of data, and the multiple Each group of data in the group of data includes a first historical feature matrix sample in a first historical time period, a second historical feature matrix sample in a second historical time period, and behavior labels, each of the behavior labels represents a behavior recognition result.
10. 如权利要求9所述的终端设备，其中，所述获取每类所述第一属性描述文字中预设数量个第一关键词，包括：The terminal device according to claim 9, wherein said obtaining a preset number of first keywords in each type of said first attribute description text comprises:

    对每类所述第一属性描述文字进行分词、去停用词和去非特征词处理，得到所述第一属性描述文字对应的第一关键词集合；Performing word segmentation, removing stop words and removing non-characteristic words for each type of the first attribute description text to obtain the first keyword set corresponding to the first attribute description text;

    计算所述第一关键词集合中每个第一关键词的相关度；所述相关度表征所述第一关键词与所述第一关键词集合中其他第一关键词之间的关联程度；Calculate the relevance of each first keyword in the first keyword set; the relevance represents the degree of association between the first keyword and other first keywords in the first keyword set;

    将所述第一关键词集合中相关度排名靠前的预设数量个第一关键词，作为每类所述第一属性描述文字对应的预设数量个第一关键词；Taking the preset number of first keywords with the highest relevance ranking in the first keyword set as the preset number of first keywords corresponding to each type of the first attribute description text;

    所述获取每类所述第二属性描述文字中预设数量个第二关键词，包括：The acquiring a preset number of second keywords in the description text of each type of the second attribute includes:

    对每类所述第二属性描述文字进行分词、去停用词和去非特征词处理，得到所述第二属性描述文字对应的第二关键词集合；Performing word segmentation, removing stop words and removing non-characteristic words for each type of the second attribute description text to obtain a second keyword set corresponding to the second attribute description text;

    计算所述第二关键词集合中每个第二关键词的相关度；所述相关度表征所述第二关键词与所述第二关键词集合中其他第二关键词之间的关联程度；Calculate the relevance of each second keyword in the second keyword set; the relevance represents the degree of association between the second keyword and other second keywords in the second keyword set;

    将所述第二关键词集合中相关度排名靠前的预设数量个第二关键词，作为每类所述第二属性描述文字对应的预设数量个第二关键词。The preset number of second keywords with the highest relevance ranking in the second keyword set are used as the preset number of second keywords corresponding to each type of the second attribute description text.
11. 如权利要求10所述的终端设备，其中，所述计算所述第一关键词集合中每个第一关键词的相关度，包括：The terminal device according to claim 10, wherein the calculating the relevance of each first keyword in the first keyword set comprises:

    针对所述第一关键词集合中每个第一关键词，分别获取所述第一关键词与所述第一关键词集合中其他第一关键词之间的相关度；对所述第一关键词与其他第一关键词之间的相关度求和，作为所述第一关键词集合中每个所述第一关键词的相关度；For each first keyword in the first keyword set, the relevance between the first keyword and the other first keywords in the first keyword set is obtained; for the first keyword The sum of the relevance between a word and other first keywords is used as the relevance of each of the first keywords in the first keyword set;

    所述计算所述第二关键词集合中每个第二关键词的相关度，包括：The calculating the relevance of each second keyword in the second keyword set includes:

    针对所述第二关键词集合中每个第二关键词，分别获取所述第二关键词与所述第二关键词集合中其他第二关键词之间的相关度；对所述第二关键词与其他第二关键词之间的相关度求和，作为所述第二关键词集合中每个所述第二关键词的相关度。For each second keyword in the second keyword set, the relevance between the second keyword and other second keywords in the second keyword set is obtained; for the second key The sum of the relevance between a word and other second keywords is used as the relevance of each second keyword in the second keyword set.
12. 如权利要求9所述的终端设备，其中，所述将每个所述第一关键词转化成第一特征向量，包括：9. The terminal device of claim 9, wherein said converting each of said first keywords into a first feature vector comprises:

    从预先建立的关键词与特征向量的对应关系中，获取每个所述第一关键词对应的第一特征向量；Acquiring the first feature vector corresponding to each of the first keywords from the pre-established correspondence between keywords and feature vectors;

    所述将每个所述第二关键词转化成第二特征向量，包括：The converting each of the second keywords into a second feature vector includes:

    从预先建立的关键词与特征向量的对应关系中，获取每个所述第二关键词对应的第二特征向量。From the pre-established correspondence between keywords and feature vectors, a second feature vector corresponding to each of the second keywords is obtained.
13. 如权利要求11所述的终端设备，其中，通过以下公式获取第i个第一关键词与第j个第一关键词之间的相关度RelKeywordi,j：The terminal device according to claim 11, wherein the correlation degree RelKeywordi,j between the i-th first keyword and the j-th first keyword is obtained by the following formula:

    

    

    其中，NumProSeni为第i个第一关键词所在的第一属性描述文字的总句子数；Among them, NumProSeni is the total number of sentences of the first attribute description text where the i-th first keyword is located;

    NumProSenj为第j个第一关键词所在的第一属性描述文字的总句子数，NumProSeni等于NumProSenj；NumProSenj is the total number of sentences of the first attribute description text where the j-th first keyword is located, and NumProSeni is equal to NumProSenj;

    NumKeywordSeni为所述第一属性描述文字的总句子数中出现第i个第一关键词的句子的句子次数；NumKeywordSeni is the number of sentences in which the sentence of the i-th first keyword appears in the total number of sentences of the first attribute description text;

    NumKeywordSenj为所述第一属性描述文字的总句子数中出现第j个第一关键词的句子的句子次数；NumKeywordSenj is the number of sentences in which the j-th sentence of the first keyword appears in the total number of sentences of the first attribute description text;

    NumKeywordSeni,j为所述第一属性描述文字的总句子数中同时出现第i个和第j个第一关键词的句子的句子次数。NumKeywordSeni,j is the number of sentences in which the i-th and j-th sentences of the first keyword appear simultaneously in the total number of sentences of the first attribute description text.
14. 如权利要求11所述的终端设备，其中，通过以下公式获取第k个第二关键词与第l个第二关键词之间的相关度RelKeywordk,l：The terminal device according to claim 11, wherein the correlation degree RelKeywordk,l between the kth second keyword and the lth second keyword is obtained by the following formula:

    

    

    其中，NumProSenk为第k个第二关键词所在的第二属性描述文字的总句子数；Among them, NumProSenk is the total number of sentences of the second attribute description text where the k-th second keyword is located;

    NumProSenl为第l个第二关键词所在的第二属性描述文字的总句子数，NumProSenk等于NumProSenl；NumProSenl is the total number of sentences of the second attribute description text where the lth second keyword is located, and NumProSenk is equal to NumProSenl;

    NumKeywordSenk为所述第二属性描述文字的总句子数中出现第k个第二关键词的句子的句子次数；NumKeywordSenk is the number of sentences in which the k-th second keyword appears in the total number of sentences of the second attribute description text;

    NumKeywordSenl为所述第二属性描述文字的总句子数中出现第l个第二关键词的句子的句子次数；NumKeywordSenl is the number of sentences in which the sentence of the l-th second keyword appears in the total number of sentences of the second attribute description text;

    NumKeywordSenk,l为所述第二属性描述文字的总句子数中同时出现第k个和第l个第二关键词的句子的句子次数。NumKeywordSenk,l is the number of sentences in which the k-th and l-th sentences of the second keyword appear simultaneously in the total number of sentences of the second attribute description text.
15. 一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，所述计算机程序被处理器执行时实现如下步骤：A computer-readable storage medium, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented:

    获取对象在第一历史时间段的第一描述文件和在第二历史时间段的第二描述文件；其中，所述第一描述文件包括多类第一属性描述文字，每类所述第一属性描述文字针对所述对象在所述第一历史时间段的一个属性进行描述；所述第二描述文件包括多类第二属性描述文字，每类所述第二属性描述文字针对所述对象在所述第二历史时间段的一个属性进行描述；Acquire the first description file of the object in the first historical time period and the second description file in the second historical time period; wherein, the first description file includes multiple types of first attribute description text, and each type of the first attribute The description text describes one attribute of the object in the first historical time period; the second description file includes multiple types of second attribute description text, and each type of the second attribute description text is specific to the object in the place. Describe an attribute of the second historical time period;

    针对所述第一描述文件中每类所述第一属性描述文字，获取每类所述第一属性描述文字中预设数量个第一关键词，将每个所述第一关键词转化成第一特征向量，对预设数量个所述第一特征向量求均值，得到每类所述第一属性描述文字对应的第一均值特征向量；For each type of the first attribute description text in the first description file, obtain a preset number of first keywords in each type of the first attribute description text, and convert each of the first keywords into a A feature vector, averaging a preset number of the first feature vectors to obtain the first average feature vector corresponding to each type of the first attribute description text;

    针对所述第二描述文件中每类所述第二属性描述文字，获取每类所述第二属性描述文字中预设数量个第二关键词，将每个所述第二关键词转化成第二特征向量，对预设数量个所述 第二特征向量求均值，得到每类所述第二属性描述文字对应的第二均值特征向量；For each type of the second attribute description text in the second description file, obtain a preset number of second keywords in each type of the second attribute description text, and convert each of the second keywords into the first Two feature vectors, averaging a preset number of the second feature vectors to obtain a second average feature vector corresponding to each type of the second attribute description text;

    将所述第一描述文件中多类所述第一属性描述文字对应的第一均值特征向量进行组合，生成第一历史特征矩阵；Combining the first mean value feature vectors corresponding to the multiple types of the first attribute description text in the first description file to generate a first historical feature matrix;

    将所述第二描述文件中每类所述第二属性描述文字对应的第二均值特征向量进行组合，生成第二历史特征矩阵；Combine the second mean eigenvectors corresponding to each type of the second attribute description text in the second description file to generate a second historical feature matrix;

    将所述第一历史特征矩阵和所述第二历史特征矩阵组合后输入模型得到所述对象的行为识别结果，其中，所述模型为使用多组数据通过机器学习训练得出的，所述多组数据中的每组数据包括第一历史时间段的第一历史特征矩阵样本，第二历史时间段的第二历史特征矩阵样本和行为标签，每种所述行为标签表征一种行为识别结果。The first historical feature matrix and the second historical feature matrix are combined and input into a model to obtain the behavior recognition result of the object, wherein the model is obtained through machine learning training using multiple sets of data, and the multiple Each group of data in the group of data includes a first historical feature matrix sample in a first historical time period, a second historical feature matrix sample in a second historical time period, and behavior labels, each of the behavior labels represents a behavior recognition result.
16. 如权利要求15所述的计算机可读存储介质，其中，所述获取每类所述第一属性描述文字中预设数量个第一关键词，包括：15. The computer-readable storage medium of claim 15, wherein said obtaining a preset number of first keywords in each type of said first attribute description text comprises:

    对每类所述第一属性描述文字进行分词、去停用词和去非特征词处理，得到所述第一属性描述文字对应的第一关键词集合；Performing word segmentation, removing stop words and removing non-characteristic words for each type of the first attribute description text to obtain the first keyword set corresponding to the first attribute description text;

    计算所述第一关键词集合中每个第一关键词的相关度；所述相关度表征所述第一关键词与所述第一关键词集合中其他第一关键词之间的关联程度；Calculate the relevance of each first keyword in the first keyword set; the relevance represents the degree of association between the first keyword and other first keywords in the first keyword set;

    将所述第一关键词集合中相关度排名靠前的预设数量个第一关键词，作为每类所述第一属性描述文字对应的预设数量个第一关键词；Taking the preset number of first keywords with the highest relevance ranking in the first keyword set as the preset number of first keywords corresponding to each type of the first attribute description text;

    所述获取每类所述第二属性描述文字中预设数量个第二关键词，包括：The acquiring a preset number of second keywords in the description text of each type of the second attribute includes:

    对每类所述第二属性描述文字进行分词、去停用词和去非特征词处理，得到所述第二属性描述文字对应的第二关键词集合；Performing word segmentation, removing stop words and removing non-characteristic words for each type of the second attribute description text to obtain a second keyword set corresponding to the second attribute description text;

    计算所述第二关键词集合中每个第二关键词的相关度；所述相关度表征所述第二关键词与所述第二关键词集合中其他第二关键词之间的关联程度；Calculate the relevance of each second keyword in the second keyword set; the relevance represents the degree of association between the second keyword and other second keywords in the second keyword set;

    将所述第二关键词集合中相关度排名靠前的预设数量个第二关键词，作为每类所述第二属性描述文字对应的预设数量个第二关键词。The preset number of second keywords with the highest relevance ranking in the second keyword set are used as the preset number of second keywords corresponding to each type of the second attribute description text.
17. 如权利要求16所述的计算机可读存储介质，其中，所述计算所述第一关键词集合中每个第一关键词的相关度，包括：16. The computer-readable storage medium of claim 16, wherein the calculating the relevance of each first keyword in the first keyword set comprises:

    针对所述第一关键词集合中每个第一关键词，分别获取所述第一关键词与所述第一关键词集合中其他第一关键词之间的相关度；对所述第一关键词与其他第一关键词之间的相关度求和，作为所述第一关键词集合中每个所述第一关键词的相关度；For each first keyword in the first keyword set, the relevance between the first keyword and the other first keywords in the first keyword set is obtained; for the first keyword The sum of the relevance between a word and other first keywords is used as the relevance of each of the first keywords in the first keyword set;

    所述计算所述第二关键词集合中每个第二关键词的相关度，包括：The calculating the relevance of each second keyword in the second keyword set includes:

    针对所述第二关键词集合中每个第二关键词，分别获取所述第二关键词与所述第二关键词集合中其他第二关键词之间的相关度；对所述第二关键词与其他第二关键词之间的相关度求和，作为所述第二关键词集合中每个所述第二关键词的相关度。For each second keyword in the second keyword set, the relevance between the second keyword and other second keywords in the second keyword set is obtained; for the second key The sum of the relevance between a word and other second keywords is used as the relevance of each second keyword in the second keyword set.
18. 如权利要求17所述的计算机可读存储介质，其中，所述将每个所述第一关键词转化成第一特征向量，包括：17. The computer-readable storage medium of claim 17, wherein said converting each of said first keywords into a first feature vector comprises:

    从预先建立的关键词与特征向量的对应关系中，获取每个所述第一关键词对应的第一特征向量；Acquiring the first feature vector corresponding to each of the first keywords from the pre-established correspondence between keywords and feature vectors;

    所述将每个所述第二关键词转化成第二特征向量，包括：The converting each of the second keywords into a second feature vector includes:

    从预先建立的关键词与特征向量的对应关系中，获取每个所述第二关键词对应的第二特征向量。From the pre-established correspondence between keywords and feature vectors, a second feature vector corresponding to each of the second keywords is obtained.
19. 如权利要求17所述的计算机可读存储介质，其中，通过以下公式获取第i个第一关键词与第j个第一关键词之间的相关度RelKeywordi,j：17. The computer-readable storage medium of claim 17, wherein the correlation degree RelKeywordi,j between the i-th first keyword and the j-th first keyword is obtained by the following formula:

    

    

    其中，NumProSeni为第i个第一关键词所在的第一属性描述文字的总句子数；Among them, NumProSeni is the total number of sentences of the first attribute description text where the i-th first keyword is located;

    NumProSenj为第j个第一关键词所在的第一属性描述文字的总句子数，NumProSeni等于NumProSenj；NumProSenj is the total number of sentences of the first attribute description text where the j-th first keyword is located, and NumProSeni is equal to NumProSenj;

    NumKeywordSeni为所述第一属性描述文字的总句子数中出现第i个第一关键词的句子的句子次数；NumKeywordSeni is the number of sentences in which the sentence of the i-th first keyword appears in the total number of sentences of the first attribute description text;

    NumKeywordSenj为所述第一属性描述文字的总句子数中出现第j个第一关键词的句子的句子次数；NumKeywordSenj is the number of sentences in which the j-th sentence of the first keyword appears in the total number of sentences of the first attribute description text;

    NumKeywordSeni,j为所述第一属性描述文字的总句子数中同时出现第i个和第j个第一关键词的句子的句子次数。NumKeywordSeni,j is the number of sentences in which the i-th and j-th sentences of the first keyword appear simultaneously in the total number of sentences of the first attribute description text.
20. 如权利要求17所述的计算机可读存储介质，其中，通过以下公式获取第k个第二关键词与第l个第二关键词之间的相关度RelKeywordk,l：The computer-readable storage medium according to claim 17, wherein the correlation degree RelKeywordk,l between the kth second keyword and the lth second keyword is obtained by the following formula:

    

    

    其中，NumProSenk为第k个第二关键词所在的第二属性描述文字的总句子数；Among them, NumProSenk is the total number of sentences of the second attribute description text where the k-th second keyword is located;

    NumProSenl为第l个第二关键词所在的第二属性描述文字的总句子数，NumProSenk等于NumProSenl；NumProSenl is the total number of sentences of the second attribute description text where the lth second keyword is located, and NumProSenk is equal to NumProSenl;

    NumKeywordSenk为所述第二属性描述文字的总句子数中出现第k个第二关键词的句子的句子次数；NumKeywordSenk is the number of sentences in which the k-th second keyword appears in the total number of sentences of the second attribute description text;

    NumKeywordSenl为所述第二属性描述文字的总句子数中出现第l个第二关键词的句子的句子次数；NumKeywordSenl is the number of sentences in which the sentence of the l-th second keyword appears in the total number of sentences of the second attribute description text;

    NumKeywordSenk,l为所述第二属性描述文字的总句子数中同时出现第k个和第l个第二关键词的句子的句子次数。NumKeywordSenk,l is the number of sentences in which the k-th and l-th sentences of the second keyword appear simultaneously in the total number of sentences of the second attribute description text.

Images