WO2020088058A1

WO2020088058A1 - Information generating method and device

Info

Publication number: WO2020088058A1
Application number: PCT/CN2019/102176
Authority: WO
Inventors: 谷长胜; 杨希旺
Original assignee: 北京字节跳动网络技术有限公司
Priority date: 2018-10-31
Filing date: 2019-08-23
Publication date: 2020-05-07
Also published as: CN111125574A; CN111125574B

Abstract

Disclosed are an information generating method and device. An embodiment of the method comprises: acquiring user information of a target user and at least one information to be displayed of a first type, said information of the first type being information to be pushed to a terminal of the target user so as to be displayed on the terminal; for each of the at least one said information of the first type, acquiring characteristic information of each of said information of the first type; inputting the characteristic information and user information into a pre-trained attention score model to obtain an attention score used for characterizing the degree of attention paid by the target user to each of said information of the first type; and acquiring a preset compensation display cost value and generating a first type display cost value of each of said information of the first type on the basis of the compensation display cost value and the attention score.

Description

生成信息的方法和装置Method and device for generating information

本申请要求在2018年10月31日提交中国专利局、申请号为201811290026.9的中国专利申请的优先权，该申请的全部内容通过引用结合在本申请中。This application requires the priority of the Chinese patent application with the application number 201811290026.9 filed by the China Patent Office on October 31, 2018. The entire content of this application is incorporated by reference in this application.

技术领域Technical field

本申请实施例涉及计算机技术领域，例如生成信息的方法和装置。The embodiments of the present application relate to the field of computer technology, such as a method and an apparatus for generating information.

背景技术Background technique

相关技术中的展示信息的方法是将各种类别的信息分别在对应的展示位中展示。The method of displaying information in the related art is to display various types of information in corresponding display positions.

随着互联网技术的发展，用户通过终端浏览信息越来越频繁，为了更有针对性地在用户终端上显示不同类型的展示信息，通常将不同类型的展示信息分别展示在终端设备的不同显示区域内。为了尽可能有针对性地向用户推送展示信息，通常需要得到各个展示信息的展示代价值(即用于表征展示信息的提供者为向用户推送展示信息所付出的代价的数值)，从而根据展示代价值确定展示信息的排序。With the development of Internet technology, users browse information more and more frequently through the terminal. In order to display different types of display information on the user terminal in a more targeted manner, different types of display information are usually displayed in different display areas of the terminal device Inside. In order to push the display information to users as targeted as possible, it is usually necessary to obtain the display generation value of each display information (that is, the value used to characterize the price paid by the provider of the display information to push the display information to the user), according to the display The generation value determines the order of the displayed information.

发明内容Summary of the invention

本申请实施例提出了生成信息的方法和装置。The embodiments of the present application provide a method and a device for generating information.

第一方面，本申请实施例提供了一种生成信息的方法，该方法包括：获取目标用户的用户信息和至少一个第一类待展示信息，所述第一类待展示信息为待向目标用户的终端推送以在终端上展示的信息；对于至少一个第一类待展示信息中的每个第一类待展示信息，获取每个第一类待展示信息的特征信息；将所述特征信息和用户信息输入预先训练的被关注评分模型，得到用于表征每个第一类待展示信息被目标用户关注的程度的被关注评分；获取预设的补偿展示代价值，并基于所述补偿展示代价值和所述被关注评分，生成每个第一类待展示信息的第一类展示代价值。In a first aspect, an embodiment of the present application provides a method for generating information. The method includes: acquiring user information of a target user and at least one first type of information to be displayed, where the first type of information to be displayed is a target user to be displayed The terminal pushes the information to be displayed on the terminal; for each first-type to-be-displayed information in at least one first-type to-be-displayed information, obtain feature information of each first-type to-be-displayed information; compare the feature information with The user information is input into the pre-trained scoring model of attention, to obtain a scoring of interest that characterizes the degree to which each first-type information to be displayed is focused by the target user; obtain a preset value of the compensation display generation, and based on the compensation display generation The value and the focused score generate the first-generation display generation value of each first-type information to be displayed.

第二方面，本申请实施例提供了一种生成信息的装置，该装置包括：第一获取单元，被配置成获取目标用户的用户信息和至少一个待向目标用户的终端推送以在终端上展示的第一类待展示信息；生成单元，被配置成对于至少一个第一类待展示信息中的每个第一类待展示信息，获取每个第一类待展示信息的特征信息；将所述特征信息和用户信息输入预先训练的被关注评分模型，得到用于表征每个第一类待展示信息被目标用户关注的程度的被关注评分；获取预设的补偿展示代价值，并基于所述补偿展示代价值和所述被关注评分，生成每个第一类待展示信息的第一类展示代价值。In a second aspect, an embodiment of the present application provides an apparatus for generating information. The apparatus includes: a first acquiring unit configured to acquire user information of a target user and at least one to be pushed to a terminal of the target user to be displayed on the terminal The first type of information to be displayed; the generating unit is configured to obtain the characteristic information of each first type of information to be displayed for each first type of information to be displayed in at least one first type of information to be displayed; The feature information and user information are input into the pre-trained scoring model of attention, to obtain a scoring of attention that characterizes the degree to which each first type of information to be displayed is focused by the target user; obtain a preset compensation display generation value, and based Compensate the display generation value and the focused score to generate the first type of display generation value for each first type of information to be displayed.

第三方面，本申请实施例提供了一种服务器，该服务器包括：至少一个处理器；存储装置，其上存储有至少一个程序；当至少一个程序被至少一个处理器执行，使得至少一个处理器实现如第一方面中任一实现方式描述的方法。In a third aspect, an embodiment of the present application provides a server including: at least one processor; a storage device on which at least one program is stored; when at least one program is executed by at least one processor, such that at least one processor Implement the method as described in any one of the implementation manners of the first aspect.

第四方面，本申请实施例提供了一种计算机可读介质，其上存储有计算机程序，该计算机程序被处理器执行时实现如第一方面中任一实现方式描述的方法。According to a fourth aspect, an embodiment of the present application provides a computer-readable medium on which a computer program is stored. When the computer program is executed by a processor, the method described in any one of the implementation manners of the first aspect is implemented.

附图说明BRIEF DESCRIPTION

图1是本申请的一个实施例可以应用于其中的示例性***架构图；FIG. 1 is an exemplary system architecture diagram to which an embodiment of the present application can be applied;

图2是根据本申请实施例的生成信息的方法的一个实施例的流程图；2 is a flowchart of an embodiment of a method for generating information according to an embodiment of the present application;

图3是根据本申请实施例的生成信息的方法的一个应用场景的示意图；3 is a schematic diagram of an application scenario of a method for generating information according to an embodiment of the present application;

图4是根据本申请实施例的生成信息的方法的又一个实施例的流程图；4 is a flowchart of another embodiment of a method for generating information according to an embodiment of the present application;

图5是根据本申请实施例的生成信息的装置的一个实施例的结构示意图；5 is a schematic structural diagram of an embodiment of an apparatus for generating information according to an embodiment of the present application;

图6是适于用来实现本申请实施例的服务器的计算机***的结构示意图。6 is a schematic structural diagram of a computer system suitable for implementing the server of the embodiment of the present application.

具体实施方式detailed description

下面结合附图和实施例对本申请作详细说明。可以理解的是，此处所描述的具体实施例仅仅用于解释相关发明，而非对该发明的限定。另外还需要说明的是，为了便于描述，附图中仅示出了与有关发明相关的部分。The application will be described in detail below with reference to the drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the related invention, but not to limit the invention. It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings.

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。下面将参考附图并结合实施例来详细说明本申请。It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other if there is no conflict. The present application will be described in detail below with reference to the drawings and in conjunction with the embodiments.

图1示出了可以应用本申请实施例的生成信息的方法或生成信息的装置的示例性***架构100。FIG. 1 shows an exemplary system architecture 100 to which the method for generating information or the device for generating information of the embodiments of the present application can be applied.

如图1所示，***架构100可以包括终端设备101、102、103，网络104和服务器105。网络104用以在终端设备101、102、103和服务器105之间提供通信链路的介质。网络104可以包括各种连接类型，例如有线、无线通信链路或者光纤电缆等等。As shown in FIG. 1, the system architecture 100 may include

terminal devices

101, 102, 103, a network 104, and a server 105. The network 104 is a medium used to provide a communication link between the

terminal devices

101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, and so on.

用户可以使用终端设备101、102、103通过网络104与服务器105交互，以接收或发送消息等。终端设备101、102、103上可以安装有各种通讯客户端应用，例如网页浏览器应用、购物类应用、搜索类应用、即时通信工具、邮箱客户端、社交平台软件等。The user can use the

terminal devices

101, 102, 103 to interact with the server 105 through the network 104 to receive or send messages, and so on. Various communication client applications may be installed on the

terminal devices

101, 102, and 103, such as web browser applications, shopping applications, search applications, instant communication tools, email clients, and social platform software.

终端设备101、102、103可以是硬件，也可以是软件。在终端设备101、102、103为硬件的情况下，可以是具有显示屏的各种电子设备，包括但不限于智能手机、平板电脑、膝上型便携计算机和台式计算机等等。在终端设备101、102、103为软件的情况下，可以安装在上述所列举的电子设备中。其可以实现成多个软件或软件模块(例如用来提供分布式服务的软件或软件模块)，也可以实现成单个软件或软件模块。在此不做具体限定。The

terminal devices

101, 102, and 103 may be hardware or software. When the

terminal devices

101, 102, and 103 are hardware, they may be various electronic devices with display screens, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and so on. When the

terminal devices

101, 102, and 103 are software, they can be installed in the electronic devices listed above. It can be implemented as multiple software or software modules (for example, software or software modules used to provide distributed services) or as a single software or software module. There is no specific limit here.

服务器105可以是提供各种服务的服务器，例如对终端设备101、102、103上展示的信息提供支持的后台信息处理服务器。后台信息处理可以对获取的第一类待展示信息和待使用终端设备浏览第一类待展示信息的用户的用户信息等信息进行处理，生成第一类待展示信息的第一类展示代价值。The server 105 may be a server that provides various services, such as a background information processing server that supports information displayed on the

terminal devices

101, 102, and 103. The background information processing can process the acquired first-type information to be displayed and the user information of the user who uses the terminal device to browse the first-type information to be displayed, and generate the first-type display substitute value of the first-type information to be displayed.

需要说明的是，本申请实施例所提供的生成信息的方法一般由服务器105执行，相应地，生成信息的装置一般设置于服务器105中。It should be noted that the method for generating information provided by the embodiments of the present application is generally executed by the server 105, and accordingly, the device for generating information is generally provided in the server 105.

需要说明的是，服务器可以是硬件，也可以是软件。在服务器为硬件的情况下，可以实现成多个服务器组成的分布式服务器集群，也可以实现成单个服务器。在服务器为软件的情况下，可以实现成多个软件或软件模块(例如用来提供分布式服务的软件或软件模块)，也可以实现成单个软件或软件模块。在此不做具体限定。It should be noted that the server can be hardware or software. When the server is hardware, it can be implemented as a distributed server cluster composed of multiple servers, or as a single server. When the server is software, it may be implemented as multiple software or software modules (for example, software or software modules for providing distributed services), or may be implemented as a single software or software module. There is no specific limit here.

应该理解，图1中的终端设备、网络和服务器的数目仅仅是示意性的。根据实现需要，可以具有任意数目的终端设备、网络和服务器。It should be understood that the numbers of terminal devices, networks, and servers in FIG. 1 are only schematic. According to the implementation needs, there can be any number of terminal devices, networks and servers.

继续参考图2，图2示出了根据本申请的生成信息的方法的一个实施例的流程200。该生成信息的方法，包括以下步骤：With continued reference to FIG. 2, FIG. 2 illustrates a flow 200 of an embodiment of a method for generating information according to the present application. The method for generating information includes the following steps:

在步骤201中，获取目标用户的用户信息和至少一个第一类待展示信息，第一类展示信息为待向目标用户的终端推送、以在终端上展示的信息。In step 201, the user information of the target user and at least one first type of information to be displayed are obtained. The first type of display information is information to be pushed to the terminal of the target user to be displayed on the terminal.

在本实施例中，生成信息的方法的执行主体(例如图1所示的服务器)可以通过有线连接方式或者无线连接方式从远程或从本地获取至少一个待向目标用户的终端推送以在终端上展示的第一类待展示信息和目标用户的用户信息。其中，目标用户可以是待利用其使用的终端(例如图1所示的终端设备)浏览上述执行主体推送的信息的用户。目标用户的用户信息可以用于表征目标用户的特征，目标用户的特征包括但不限于以下至少一种：目标用户的性别、年龄、兴趣等。In this embodiment, the execution subject of the method for generating information (such as the server shown in FIG. 1) may obtain at least one terminal to be pushed to the target user from a remote or local via wired connection or wireless connection to be pushed on the terminal The first type of information to be displayed and the user information of the target user. Wherein, the target user may be a user to browse the information pushed by the above-mentioned execution subject using the terminal (eg, the terminal device shown in FIG. 1) to be used by it. The user information of the target user can be used to characterize the characteristics of the target user, and the characteristics of the target user include but are not limited to at least one of the following: the target user's gender, age, and interests.

第一类待展示信息可以是属于预设类别的信息，例如文章类信息、视频类信息等。第一类待展示信息可以包括但不限于以下至少一项：图片、文字、音频、视频、链接地址等。The first type of information to be displayed may be information belonging to a preset category, such as article type information, video type information, and the like. The first type of information to be displayed may include, but is not limited to, at least one of the following: pictures, text, audio, video, link addresses, and so on.

在步骤202中，对于至少一个第一类待展示信息中的每个第一类待展示信息，获取每个第一类待展示信息的特征信息；将特征信息和用户信息输入预先训练的被关注评分模型，得到用于表征每个第一类待展示信息被目标用户关注的程度的被关注评分；获取预设的补偿展示代价值，并基于补偿展示代价值和被关注评分，生成每个第一类待展示信息的第一类展示代价值。In step 202, for each first-type to-be-displayed information in at least one first-type to-be-displayed information, obtain feature information of each first-type to-be-displayed information; input the feature information and user information into the pre-trained attention Scoring model to obtain the attention score that characterizes the degree to which each first-type information to be displayed is focused by the target user; obtain the preset compensation display generation value, and generate each ranking based on the compensation display generation value and attention score The first type of display value of a type of information to be displayed.

在本实施例中，对于步骤201中获取的至少一个第一类待展示信息中的第一类待展示信息，上述执行主体可以针对该第一类待展示信息执行如下步骤：In this embodiment, for the first type of to-be-displayed information among the at least one first type of to-be-displayed information acquired in step 201, the above-mentioned execution subject may perform the following steps for the first type of to-be-displayed information:

在步骤2021中，获取该第一类待展示信息的特征信息。In step 2021, the characteristic information of the first type of information to be displayed is obtained.

在一实施例中，上述执行主体可以从远程或从本地获取该第一类待展示信息的特征信息。其中，特征信息用于表征第一类待展示信息的特征。第一类待展示信息的特征可以包括但不限于以下至少一种：第一类待展示信息的标题、第一类待展示信息所属的类型、第一类待展示信息包括的链接地址等。特征信息可以是第一类待展示信息包括的信息(例如第一类待展示信息的标题)，也可以是与第一类待展示信息相关的信息(例如表征第一类待展示信息所属的类型的类型信息、第一类待展示信息的生成时间等)。In an embodiment, the above-mentioned execution subject may obtain the feature information of the first type of information to be displayed remotely or locally. Among them, the characteristic information is used to characterize the characteristics of the first type of information to be displayed. The characteristics of the first type of information to be displayed may include, but are not limited to, at least one of the following: the title of the first type of information to be displayed, the type of the first type of information to be displayed, the link address included in the first type of information to be displayed, and so on. The feature information may be information included in the first type of information to be displayed (such as the title of the first type of information to be displayed), or may be information related to the first type of information to be displayed (such as characterizing the type of the first type of information to be displayed) Type information, the generation time of the first type of information to be displayed, etc.).

在步骤2022中，将特征信息和用户信息输入预先训练的被关注评分模型，得到用于表征每个第一类待展示信息被目标用户关注的程度的被关注评分。In step 2022, the feature information and the user information are input into the pre-trained scoring model of attention, to obtain a scoring attention that is used to characterize the degree to which each first type of information to be displayed is focused by the target user.

其中，被关注评分可以用于表征第一类待展示信息被目标用户关注的程度。通常，被关注评分可以是大于等于0小于等于1的数值，被关注评分越大，表示第一类待展示信息被目标用户关注的程度越大。Among them, the attention score can be used to characterize the degree to which the first type of information to be displayed is focused by the target user. Generally, the attention score may be a value greater than or equal to 0 and less than or equal to 1, and the greater the attention score, the greater the degree of attention of the first type of information to be displayed by the target user.

上述被关注评分模型设置为表征展示信息的特征信息、浏览展示信息的用户的用户信息与用于表征展示信息被用户关注的关注程度的被关注评分的对应关系。The above-mentioned attention-scoring model is set to correspond to the feature information characterizing the display information, the user information of the user browsing the display information, and the attention score used to characterize the degree of attention of the display information to the user.

在一实施例中，被关注评分模型可以是技术人员预先基于对大量的特征信息、用户信息与被关注评分的统计而预先制定的、存储有多个特征信息、用户信息与被关注评分的对应关系表。其中，对应关系表中的被关注评分可以是技术人员预先针对用户信息表征的用户所属的用户群体(例如可以用预设的用户信息集合表征)对第一类待展示信息进行的评分所做的统计，并计算(例如将各个用户的评分取平均)得到的被关注评分。上述执行主体可以根据输入的特征信息和用户信息，从对应关系表中查找对应的被关注评分。上述被关注评分模型也可以为基于预设的训练样本，利用机器学习方法对初始模型(例如神经网络)进行训练后得到的模型。通过使用被关注评分模型，可以根据不同的第一类待展示信息的特征信息和用户信息，得到不同的被关注评分，有助于提高生成被关注评分的准确性。In an embodiment, the focused scoring model may be pre-formulated by a technician based on statistics of a large amount of feature information, user information, and focused scores, and stores a plurality of feature information, user information, and correspondence of the focused scores Relational tables. Wherein, the attention score in the correspondence table may be a score made by the technician in advance for the user group to which the user characterized by the user information belongs (for example, it may be characterized by a preset user information set) to the first type of information to be displayed Make statistics and calculate (for example, average the scores of users) the scores of interest. According to the input feature information and user information, the above-mentioned execution subject may search for the corresponding focused score from the correspondence table. The above noted scoring model may also be a model obtained after training an initial model (such as a neural network) using a machine learning method based on preset training samples. By using the focused rating model, different focused ratings can be obtained according to different feature information and user information of the first type of information to be displayed, which helps to improve the accuracy of generating focused ratings.

在本实施例的一些实现方式中，被关注评分模型可以包括结果生成子模型和至少两个评分子模型。其中，至少两个子模型中的子模型分别与结果生成模型连接。被关注评分可以按照如下步骤得到：In some implementations of this embodiment, the focused scoring model may include a result generation sub-model and at least two scoring sub-models. Among them, the sub-models in at least two sub-models are respectively connected to the result generation model. Followed scores can be obtained as follows:

对于至少两个评分子模型中的每个评分子模型，将特征信息和用户信息输入所述至少两个评分子模型中的每个评分子模型，得到每个评分子模型对应的评分。其中，评分子模型设置为表征特征信息、用户信息与评分的对应关系。For each of the at least two rating sub-models, input feature information and user information into each of the at least two rating sub-models to obtain a rating corresponding to each rating sub-model. The scoring sub-model is set to characterize the correspondence between feature information, user information, and scoring.

将所得到的评分输入结果生成子模型，得到用于表征每个第一类待展示信息被目标用户关注的程度的被关注评分。其中，结果生成子模型设置为基于预设的权重，对所得到的评分进行加权求和，得到被关注评分。The obtained score is input into the result to generate a sub-model, and a focused score that is used to characterize the degree to which each first type of information to be displayed is focused by the target user is obtained. The result generation sub-model is set to perform weighted summation of the obtained scores based on preset weights to obtain the attention score.

在一实施例中，评分子模型可以对应于预设的评分种类。例如，评分种类可以包括但不限于：用于表征用户针对特征信息对应的第一类待展示信息的点击概率的评分、用于表征用户针对特征信息对应的第一类待展示信息的转发概率的评分、用户针对特征信息对应的第一类待展示信息的评论概率的评分。其中，上述点击概率、转发概率、评论概率可以是预先对用户的行为进行统计得到的概率。例如，点击概率可以是用户在预设时间段内点击的展示信息的数量与浏览的信息的数量的比值，转发概率可以是用户在预设时间段内转发的展示信息的数量与浏览的信息的数量的比值，评论概率可以是用户在预设时间段内评论的展示信息的数量与浏览的信息的数量的比值。In an embodiment, the scoring sub-model may correspond to a preset scoring category. For example, the score category may include, but is not limited to: a score for characterizing the user's click probability of the first type of information to be displayed corresponding to the feature information, a characterization for the user's forwarding probability of the first type of information to be displayed corresponding to the feature information The score, the user's score for the comment probability of the first type of information to be displayed corresponding to the feature information. Wherein, the above-mentioned click probability, forwarding probability, and comment probability may be probabilities obtained by statistically analyzing the user's behavior. For example, the click probability may be the ratio of the number of displayed information clicked by the user in the preset time period to the number of browsed information, and the forward probability may be the number of displayed information forwarded by the user in the preset time period and the browsed information The ratio of the number and the comment probability may be the ratio of the number of displayed information commented by the user and the number of browsed information within a preset time period.

作为示例，评分子模型可以是技术人员预先基于对大量的特征信息、用户信息与属于评分子模型对应的评分种类的评分的统计而预先制定的、存储有多个特征信息、用户信息与评分的对应关系表。其中，对应关系表中的评分可以是技术人员预先针对用户信息表征的用户对第一类待展示信息进行的操作(例如点击)所做的统计，并计算(例如计算用户在预设时间段内点击的展示信息的数量与浏览的信息的数量的比值)得到的评分。上述执行主体可以根据输入的特征信息和用户信息，从对应关系表中查找对应的评分。此外，上述评分子模型也可以为基于预设的训练样本，利用机器学习方法对初始模型(例如神经网络)进行训练后得到的模型。As an example, the scoring sub-model may be pre-formulated by a technician based on statistics of a large number of feature information, user information, and ratings belonging to the scoring category corresponding to the scoring sub-model, and storing a plurality of feature information, user information, and ratings Correspondence table. Wherein, the score in the correspondence table may be the statistics made by the technician in advance on the operations (such as clicks) performed by the user represented by the user information on the first type of information to be displayed, and calculated (such as calculating the user within a preset time period) The score obtained by the ratio of the number of displayed information clicked to the number of browsed information). The above-mentioned execution subject can search for the corresponding score from the correspondence table according to the input feature information and user information. In addition, the above-mentioned scoring sub-model may also be a model obtained after training an initial model (such as a neural network) using a machine learning method based on a preset training sample.

在一实施例中，至少两个评分子模型中的每个评分子模型，预先按照如下步骤训练得到：In an embodiment, each of the at least two scoring sub-models is pre-trained according to the following steps:

获取每个评分子模型对应的训练样本集合。其中，训练样本包括预设的样本第一类待展示信息的特征信息、浏览过样本第一类待展示信息的样本用户的样本用户信息，以及预先标注的、用于表征样本用户针对样本第一类待展示信息是否进行目标操作的标注信息。例如，标注信息可以是数字，例如“0”表示样本用户没有进行目标操作，“1”表示样本用户进行了目标操作。其中，目标操作可以包括但不限于以下至少一种：点击操作、评论操作、转发操作等。Obtain the training sample set corresponding to each scoring sub-model. Among them, the training sample includes the preset feature information of the sample first type of information to be displayed, the sample user information of the sample user who has browsed the sample first type of information to be displayed, and the pre-marked characterization of the sample user for the sample first Tag information about whether the information to be displayed is targeted for operation. For example, the label information may be a number, for example, "0" indicates that the sample user has not performed the target operation, and "1" indicates that the sample user has performed the target operation. The target operation may include but is not limited to at least one of the following: click operation, comment operation, forwarding operation, and the like.

利用机器学习方法，将所获取的训练样本集合中的训练样本包括的特征信息和样本用户信息作为输入，将与输入的特征信息和样本用户信息对应的标注信息作为期望输出，训练得到每个评分子模型。需要说明的是，训练得到的评分子模型输出的评分可以是0-1之间的数值，表征样本用户针对样本第一类待展示信息进行目标操作的概率。Using machine learning methods, the training sample in the obtained training sample set includes the feature information and sample user information as input, and the label information corresponding to the input feature information and sample user information is used as the expected output, and each score is obtained by training Submodel. It should be noted that the score output by the training score sub-model obtained may be a value between 0-1, which represents the probability that the sample user performs the target operation on the first type of information to be displayed in the sample.

在一实施例中，评分子模型可以是对初始模型进行训练得到的模型。初始模型可以包括但不限于以下至少一种：因子分解机(Factorization Machine，FM)模型、神经网络模型等。初始模型可以设置有初始参数，参数在训练过程中可以被不断地调整。训练转化率预测模型的执行主体可以基于预设的损失函数计算损失值，根据损失值确定初始模型是否训练完成。在这里，需要说明的是，损失值可以用于表征实际输出与期望输出之间的差异。实践中，可以采用预设的各种损失函数计算实际输出相对于标注的输出的损失值。例如，可以采用对数损失函数、交叉熵损失函数等计算损失值。In an embodiment, the scoring sub-model may be a model obtained by training the initial model. The initial model may include, but is not limited to, at least one of the following: Factorization Machine (FM) model, neural network model, etc. The initial model can be set with initial parameters, which can be adjusted continuously during the training process. The execution subject of the training conversion rate prediction model can calculate the loss value based on the preset loss function, and determine whether the initial model is trained according to the loss value. Here, it should be noted that the loss value can be used to characterize the difference between the actual output and the expected output. In practice, various preset loss functions can be used to calculate the actual output loss relative to the marked output. For example, a log loss function, cross-entropy loss function, etc. can be used to calculate the loss value.

在步骤2023中，获取预设的补偿展示代价值，并基于补偿展示代价值和被关注评分，生成每个第一类待展示信息的第一类展示代价值。In step 2023, a preset compensation display generation value is obtained, and based on the compensation display generation value and the attention score, a first type of display generation value of each first type of information to be displayed is generated.

作为示例，上述执行主体可以将补偿展示代价值和被关注评分相乘，得到第一类展示代价值。或者，上述执行主体可以补偿展示代价值和被关注评分相乘后再进行其他方式的处理(例如将相乘后所得到的乘积与预设数值相加或相减)，得到第一类展示代价值。从而可以使得技术人员对第一类展示代价值进行调整，从而有助于当存在多个类别的待展示信息时，可以通过调整第一类待展示代价值有针对性地调整多个类别的待展示信息的排序。As an example, the above-mentioned executive body may multiply the compensation display generation value and the attention score to obtain the first type of display generation value. Alternatively, the above-mentioned executive body can compensate for the value of the display generation and multiply the score of interest and then perform other processing (such as adding or subtracting the product obtained after the multiplication and the preset value) to obtain the first type of display generation value. Therefore, the technicians can adjust the value of the first category of display generation, which helps to adjust the value of the first category of display to adjust the target value of multiple categories by adjusting the value of the first category of display Display the order of information.

补偿展示代价值可以用于表征推送某个第一类待展示信息集合的推送者为了使用户浏览第一类待展示信息集合中的第一类待展示信息所付出的代价。上述推送者可以是使用上述执行主体向用户推送第一类待展示信息的单位或个人。例如，假设用户使用安装在其使用的终端上的目标应用接收第一类待展示信息，则该目标应用的所有者即为推送者。实践中，推送者为了获得新的用户，会付出一定的代价，基于所付出的代价可以计算得到补偿展示代价值。例如将表征其付出的代价的代价值除以浏览第一类待展示信息集合中的第一类待展示信息的用户的数量，得到补偿展示代价值。通常，被关注评分越高，表示目标用户对该第一类待展示信息的关注程度越高，将其乘以补偿展示代价值，所得到的第一类展示代价值越高，表征上述推送者向目标用户推送该第一类待展示信息所付出的代价越高，同时也表征第一类待展示信息的价值越大。The compensated display generation value can be used to characterize the price paid by a pusher who pushes a certain first-type to-be-displayed information set in order for the user to browse the first-type to-be-displayed information in the first-type to-be-displayed information set. The above-mentioned pusher may be a unit or an individual who uses the above-mentioned execution subject to push the first type of information to be displayed to the user. For example, if a user uses a target application installed on a terminal used by the user to receive the first type of information to be displayed, the owner of the target application is the pusher. In practice, the pusher will pay a certain price in order to obtain new users. Based on the paid price, the compensation display value can be calculated. For example, the generation value representing the price paid by it is divided by the number of users who browse the first type of information to be displayed in the first type of information to be displayed to obtain the compensation display generation value. Generally, the higher the attention rating, the higher the target user ’s attention to the first type of information to be displayed. Multiply it by the compensation display value. The higher the value of the first type of display generation, the higher the pusher ’s value. The higher the cost of pushing the first type of information to be displayed to the target user, it also signifies that the value of the first type of information to be displayed is greater.

在本实施例的一些实现方式中，生成第一类展示代价值所使用的补偿展示代价值可以预先由上述执行主体或其他电子设备通过如下步骤得到：In some implementations of this embodiment, the compensation display generation value used to generate the first type of display generation value may be obtained in advance by the above-mentioned executing body or other electronic device through the following steps:

步骤一，获取预设的比例集合。其中，比例集合中的比例是在预设时间段内向预设的用户群中的用户所使用的终端推送的第一类待展示信息的数量和第二类待展示信息的数量的比例。上述用户集合中的用户可以是利用其使用的终端浏览上述执行主体推送的信息的用户。第二类待展示信息可以是与第一类展示信息的类别不同的展示信息。例如，第一类展示信息是文章类信息，第二类展示信息是产品宣传类信息。Step 1: Obtain a preset ratio set. The ratio in the ratio set is the ratio of the quantity of the first type of information to be displayed and the quantity of the second type of information to be pushed to the terminal used by users in the preset user group within a preset time period. The user in the user set may be a user who browses the information pushed by the execution subject using the terminal used by the user. The second type of information to be displayed may be display information different from the category of the first type of display information. For example, the first type of display information is article type information, and the second type of display information is product promotion type information.

步骤二，对于比例集合中的比例，执行如下步骤：Step 2: For the proportions in the proportion set, perform the following steps:

确定每个比例对应的损失用户的数量。其中，损失用户是在预设时间段内，按照每个比例推送第一类待展示信息和第二类待展示信息，且在预设时间段后不使用终端接收推送的第一类待展示信息和第二类待展示信息的用户。作为示例，假设预设时间段是一个月，按照该比例向上述用户集合中的用户所使用的终端推送第一类待展示信息和第二类待展示信息，一个月后，用户集合中有部分用户不再使用终端接收上述执行主体推送的第一类待展示信息和第二类待展示信息，则这部分用户即损失用户。Determine the number of lost users corresponding to each ratio. Among them, the lost user pushes the first type of information to be displayed and the second type of information to be displayed according to each proportion within a preset time period, and does not use the terminal to receive the pushed first type of information to be displayed after the preset time period And users of the second type of information to be displayed. As an example, assuming that the preset time period is one month, the first type of information to be displayed and the second type of information to be displayed are pushed to the terminal used by the users in the above user set according to this ratio. After one month, a part of the user set The user no longer uses the terminal to receive the first type of information to be displayed and the second type of information to be displayed pushed by the execution subject, and this part of the user loses the user.

基于损失用户的数量，确定每个比例对应的补充代价值。其中，补充代价值用于表征增加与损失用户等量的用户所付出的代价。通常，信息的推送者为了增加用户数量，可以付出一定的代价来增加用户。例如，上述执行主体通过向预先安装了目标应用(即用于接收第一类待展示信息和第二类待展示信息的应用)的终端推送第一类待展示信息和第二类待展示信息，通常，为了使这些终端中预先安装目标应用，信息推送者需要付出一定的代价。通常，可以获取预设的、用于表征增加一个用户所付出的代价的人均补充代价值，将人均补充代价值与损失用户的数量的乘积确定为该比例对应的补充代价值。或者，上述执行主体中可以存储有预设的、表征损失用户的数量与补充代价值的对应关系的对应关系表，从对应关系表中查找与所确定的损失用户的数量对应的补充代价值。Based on the number of lost users, determine the complementary generation value corresponding to each ratio. Among them, the supplementary generation value is used to characterize the price paid by the same number of users as the increased and lost users. Generally, in order to increase the number of users, information pushers can pay a certain price to increase users. For example, the above-mentioned executive agent pushes the first type of information to be displayed and the second type of information to be displayed by a terminal pre-installed with a target application (that is, an application for receiving the first type of information to be displayed and the second type of information to be displayed), Generally, in order to pre-install target applications in these terminals, information pushers need to pay a certain price. Generally, it is possible to obtain a preset supplementary replacement value per capita that characterizes the cost of adding a user, and determine the product of the supplementary replacement value per person and the number of lost users as the supplementary replacement value corresponding to the ratio. Alternatively, the above-mentioned executive body may store a preset correspondence table representing the correspondence between the number of lost users and the supplemental generation value, and look up the supplementary generation value corresponding to the determined number of lost users from the correspondence table.

确定预设时间段内使用终端浏览按照每个比例推送的第一类待展示信息的用户的数量，将所确定的补充代价值除以所确定的用户数量，将所得到的商确定为每个比例对应的补偿展示代价值。Determine the number of users who use the terminal to browse the first type of information to be displayed at each rate during the preset time period, divide the determined supplemental generation value by the determined number of users, and determine the obtained quotient for each The compensation corresponding to the ratio shows the generation value.

步骤三，从所确定的补偿展示代价值中，选择补偿展示代价值作为生成第一类展示代价值所使用的补偿展示代价值。Step 3: From the determined compensation display generation value, select the compensation display generation value as the compensation display generation value used to generate the first type of display generation value.

经过执行上述步骤二，可以得到上述比例集合中的比例对应的补偿展示代价值。上述执行主体可以按照各种方法从各个补偿展示代价值中选择补偿展示代价值以作为生成第一类展示代价值所使用的补偿展示代价值。作为示例，可以从各个补偿展示代价值中选择最小值或中位数等作为生成第一类展示代价值所使用的补偿展示代价值。After performing the above step two, the compensation display generation value corresponding to the proportion in the above proportion set can be obtained. The above-mentioned executive body can select the compensation display generation value from each compensation display generation value according to various methods as the compensation display generation value used to generate the first type of display generation value. As an example, a minimum value or a median value may be selected from each compensation display generation value as the compensation display generation value used to generate the first type of display generation value.

在本实施例的一些实现方式中，上述执行主体或其他电子设备可以按照如下步骤确定生成第一类展示代价值所使用的补偿展示代价值：In some implementations of this embodiment, the above-mentioned executive body or other electronic device may determine the compensation display generation value used to generate the first type of display generation value according to the following steps:

对于所确定的补偿展示代价值中的每个补偿展示代价值，将按照每个补偿展示代价值对应的比例推送的第二类待展示信息的第二类展示代价值的和作为每个补偿展示代价值对应的第二类总展示代价值。再确定第二类总展示代价值与每个补偿展示代价值对应的补充代价值的差值。For each compensated display generation value in the determined compensation display generation value, the sum of the second type of display generation value of the second type of information to be displayed pushed according to the proportion corresponding to each compensation display generation value is taken as each compensation display The second category corresponding to the generation value corresponds to the total display generation value. Then determine the difference between the total display generation value of the second category and the supplemental generation value corresponding to each compensated display generation value.

通常，第二类待展示信息的提供者(例如第二类待展示信息用于表征某产品，则该产品的所有者即为第二类待展示信息的提供者)为了使用户浏览第二类待展示信息，需要付出一定的代价，每个第二类待展示信息对应一个第二类展示代价值，第二类展示代价值用于表征第二类待展示信息的提供者为了向用户推送第二类待展示信息所付出的代价。对于某个补偿展示代价值，将该补偿展示代价值对应的各个第二类展示代价值相加即可得到用于表征各个第二类待展示信息的提供者所付出的总的代价的第二类总展示代价值。Usually, the provider of the second type of information to be displayed (for example, the second type of information to be displayed is used to characterize a product, then the owner of the product is the provider of the second type of information to be displayed) The information to be displayed needs to pay a certain price. Each second type of information to be displayed corresponds to a second type of display generation value. The second type of display generation value is used to characterize the provider of the second type of information to be displayed. The price paid for the second category of information to be displayed. For a certain compensation display generation value, the second type of display compensation value corresponding to the compensation display generation value can be added to obtain the second price that characterizes the total price paid by each provider of the second type of information to be displayed Class total display generation value.

从所确定的差值中选择最大值，将最大值对应的补偿展示代价值作为生成第一类展示代价值所使用的补偿展示代价值。实践中，各个第二类待展示信息的提供者通常向上述推送者付出代价，最大值表征各个第二类待展示信息的提供者付出的总的代价与上述推送者付出的总的代价相差最大。The maximum value is selected from the determined differences, and the compensation display generation value corresponding to the maximum value is used as the compensation display generation value used to generate the first type of display generation value. In practice, each provider of the second type of information to be displayed usually pays a price to the above-mentioned pushers, and the maximum value represents that the total cost paid by each provider of the second type of information to be displayed is the largest difference from the total cost paid by the above-mentioned pushers .

继续参见图3，图3是根据本实施例的生成信息的方法的应用场景的一个示意图。在图3的应用场景中，服务器301首先从本地获取三个待向目标用户的终端推送以在终端上展示的第一类待展示信息3021、3022、3023和目标用户的用户信息304(例如性别信息、所处地理位置信息、年龄信息等)。其中，第一类待展示信息是新闻类信息，包括新闻的标题、图像等。然后，服务器301获取每个第一类待展示信息的特征信息(例如第一类待展示信息的标题、类型名称等)，分别为3031、3032、3033。依次将每个特征信息和用户信息304输入预先训练的被关注评分模型305，得到用于表征该第一类待展示信息被目标用户关注的程度的被关注评分3061(例如“0.9”)、3062(例如“0.5”)、3063(例如“0.6”)。再然后，服务器301获取预设的补偿展示代价值307(例如“0.5”)，将补偿展示代价值307分别和各个被关注评分相乘，得到每个第一类待展示信息的第一类展示代价值3081(例如“0.45”)、3082(例如“0.25”)、3083(例如“0.3”)。Continue to refer to FIG. 3, which is a schematic diagram of an application scenario of the method for generating information according to this embodiment. In the application scenario of FIG. 3, the server 301 first obtains three types of first-to-

be-displayed information

3021, 3022, 3023 and user information 304 (such as gender) of the first user to be pushed to the terminal of the target user to be displayed on the terminal from the local Information, location information, age information, etc.). Among them, the first type of information to be displayed is news information, including news titles, images, and so on. Then, the server 301 obtains the feature information (for example, the title, type name, etc. of the first type of information to be displayed) of each first type of information to be displayed, which is 3031, 3032, and 3033, respectively. Each feature information and user information 304 are input into the pre-trained attention scoring model 305 in turn to obtain attention scores 3061 (for example, "0.9"), 3062, which characterize the degree to which the first type of information to be displayed is focused by the target user (Eg "0.5"), 3063 (eg "0.6"). Then, the server 301 obtains the preset compensation display generation value 307 (for example, "0.5"), and multiplies the compensation display generation value 307 by the scores of each concern respectively to obtain the first type of display of each first type of information to be displayed Generation value 3081 (eg "0.45"), 3082 (eg "0.25"), 3083 (eg "0.3").

本申请的上述实施例提供的方法，通过使用被关注评分模型和预设的补偿展示代价值，得到第一类待展示信息的第一类展示代价值，从而可以对不同的第一类待展示信息和不同的用户信息进行分析，有效地利用被关注评分模型，提高了生成第一类展示代价值的准确性，有助于在需要向用户展示多种待展示信息的情况下，提高将各种待展示信息展示给用户的针对性。The method provided in the above embodiments of the present application obtains the first-type display generation value of the first-type information to be displayed by using the focused scoring model and the preset compensation display generation value, so that different first-type display types can be displayed Analysis of information and different user information, effectively use the scoring model of attention, improve the accuracy of generating the first type of display generation value, and help to improve the various The targeted display of the information to be displayed to the user.

进一步参考图4，其示出了生成信息的方法的又一个实施例的流程400。该生成信息的方法的流程400，包括步骤步骤401至步骤406。With further reference to FIG. 4, it shows a flow 400 of yet another embodiment of a method of generating information. The process 400 of the method for generating information includes steps 401 to 406.

在步骤401中，获取目标用户的用户信息和至少一个第一类待展示信息，第一类待展示信息为待向目标用户的终端推送以在终端上展示的信息。In step 401, the user information of the target user and at least one first type of information to be displayed are obtained. The first type of information to be displayed is information to be pushed to the terminal of the target user to be displayed on the terminal.

在本实施例中，步骤401与图2对应实施例中的步骤201基本一致，这里不再赘述。In this embodiment, step 401 is basically the same as step 201 in the embodiment corresponding to FIG. 2 and will not be repeated here.

在步骤402中，对于至少一个第一类待展示信息中的每个第一类待展示信息，获取每个第一类待展示信息的特征信息；将特征信息和用户信息输入预先训练的被关注评分模型，得到用于表征每个第一类待展示信息被目标用户关注的程度的被关注评分；获取预设的补偿展示代价值，并基于补偿展示代价值和被关注评分，生成每个第一类待展示信息的第一类展示代价值。In step 402, for each first-type to-be-displayed information in at least one first-type to-be-displayed information, obtain feature information of each first-type to-be-displayed information; input the feature information and user information into a pre-trained attention Scoring model to obtain the attention score that characterizes the degree to which each first-type information to be displayed is focused by the target user; obtain the preset compensation display generation value, and generate each ranking based on the compensation display generation value and attention score The first type of display value of a type of information to be displayed.

在本实施例中，步骤402与图2对应实施例中的步骤202基本一致，这里不再赘述。In this embodiment, step 402 is basically the same as step 202 in the embodiment corresponding to FIG. 2 and will not be repeated here.

在步骤403中，获取预设的至少一个第二类待展示信息，以及第二类待展示信息对应的第二类展示代价值。In step 403, at least one preset second type of information to be displayed and the second type of display substitute value corresponding to the second type of information to be displayed are obtained.

在本实施例中，生成信息的方法的执行主体(例如图1所示的服务器)可以通过有线连接方式或者无线连接方式从远程或从本地获取预设的至少一个第二类待展示信息，以及第二类待展示信息对应的第二类展示代价值。In this embodiment, the execution subject of the method for generating information (such as the server shown in FIG. 1) may obtain at least one preset second type of information to be displayed from a remote or local through a wired connection or a wireless connection, and The second type of display substitute value corresponding to the second type of information to be displayed.

其中，第二类待展示信息可以是与第一类展示信息的类别不同的展示信息。例如，第一类展示信息是文章类信息，第二类展示信息是产品宣传类信息。第二类展示代价值用于表征第二类待展示信息的提供者为了向用户推送第二类待展示信息所付出的代价(例如价格、积分等)，第二类展示代价值越大，表征第二类待展示信息的提供者为了向用户推送第二类待展示信息所付出的代价越大，同时也可以表征第二类待展示信息的价值越大。Wherein, the second type of information to be displayed may be display information different from the category of the first type of display information. For example, the first type of display information is article type information, and the second type of display information is product promotion type information. The second type of display generation value is used to characterize the price (such as price, points, etc.) paid by the provider of the second type of information to be displayed in order to push the second type of information to be displayed to the user. The provider of the second type of information to be displayed pays more for pushing the second type of information to be displayed to the user, and at the same time, it can also represent the greater value of the second type of information to be displayed.

在步骤404中，将至少一个第二类展示代价值和至少一个第一类展示代价值合并为展示代价值集合，并将至少一个第一类待展示信息和至少一个第二类待展示信息合并为待展示信息集合。In step 404, at least one second-generation display generation value and at least one first-type display generation value are combined into a display generation value set, and at least one first-type display information and at least one second-type display information are combined It is a collection of information to be displayed.

在本实施例中，上述执行主体可以将所获取的至少一个第二类展示代价值和所得到的至少一个第一类展示代价值合并为展示代价值集合以及将至少一个第一类待展示信息和至少一个第二类待展示信息合并为待展示信息集合。In this embodiment, the above-mentioned execution subject may combine the acquired at least one second-generation display generation value and the obtained at least one first-type display generation value into a display generation value set and combine at least one first-type information to be displayed Merging with at least one second type of information to be displayed into a set of information to be displayed.

在步骤405中，基于展示代价值集合中的展示代价值的大小，对展示代价值集合中的展示代价值进行排序。In step 405, the display generation value in the display generation value set is sorted based on the size of the display generation value in the display generation value set.

在本实施例中，上述执行主体可以基于展示代价值集合中的展示代价值的大小，对展示代价值集合中的展示代价值进行排序。In this embodiment, the execution subject may sort the display generation value in the display generation value set based on the size of the display generation value in the display generation value set.

作为示例，上述执行主体可以按照展示代价值由大到小的顺序进行排序。由于按照上述步骤401-步骤402所得到的第一类展示代价值可以表征第一类待展示信息的价值，因此，展示代价值集合中的展示代价值可以表征不同种类的待展示信息的价值。从而可以有助于丰富对不同类型的待展示信息进行排序的依据。As an example, the above-mentioned execution subjects may be sorted in order of display generation value from largest to smallest. Since the first type of display generation value obtained according to the above steps 401-402 can characterize the value of the first type of information to be displayed, the display generation value in the display generation value set can characterize the value of different types of information to be displayed. This can help enrich the basis for sorting different types of information to be displayed.

在步骤406中，从排序后的展示代价值集合中选择展示代价值，并将所选择的展示代价值对应的、包含于待展示信息集合中的待展示信息推送至终端。In step 406, the display generation value is selected from the sorted display generation value set, and the to-be-displayed information included in the to-be-displayed information set corresponding to the selected display generation value is pushed to the terminal.

在本实施例中，上述执行主体可以首先从排序后的展示代价值集合中按照各种方法选择展示代价值。例如，可以选择全部或部分展示代价值，当选择部分展示代价值时，可以按照预设的选择数量或预设的选择比例从排序后的展示代价值集合中选择排列在前的部分展示代价值。In this embodiment, the above-mentioned execution subject may first select the display generation value from the sorted display generation value set according to various methods. For example, you can select all or part of the display generation value. When you select a partial display generation value, you can select the top part of the display generation value from the sorted display generation value set according to the preset selection number or preset selection ratio. .

然后，上述执行主体可以将所选择的展示代价值对应的、包含于待展示信息集合中的待展示信息推送至终端。作为示例，上述执行主体可以按照所选择的展示代价值由大到小的顺序，对各个展示代价值对应的待展示信息进行排序，按照排序将各个待展示信息依次推送至终端。Then, the above-mentioned execution subject may push the to-be-displayed information corresponding to the selected display generation value included in the to-be-displayed information set to the terminal. As an example, the above-mentioned execution subject may sort the information to be displayed corresponding to each display generation value according to the order of the selected display generation value from large to small, and sequentially push each information to be displayed to the terminal according to the sorting.

从图4中可以看出，与图2对应的实施例相比，本实施例中的生成信息的方法的流程400突出了从至少一个第一类待展示信息和至少一个第二类待展示信息组成的待展示信息集合中选择待展示信息并推送的步骤。由此，本实施例描述的方案可以对不同类型的待展示信息进行混合排序，从而提高了信息推送的针对性，有助于在用户终端上有针对性地顺序展示各种待展示信息。As can be seen from FIG. 4, compared with the embodiment corresponding to FIG. 2, the process 400 of the method for generating information in this embodiment highlights from at least one first type of information to be displayed and at least one second type of information to be displayed The steps of selecting and pushing the information to be displayed in the composed information to be displayed set. Therefore, the solution described in this embodiment can sort and mix different types of information to be displayed, thereby improving the pertinence of information push and helping to display various to-be-displayed information on the user terminal in a targeted order.

进一步参考图5，作为对上述各图所示方法的实现，本申请提供了一种生成信息的装置的一个实施例，该装置实施例与图2所示的方法实施例相对应，该装置具体可以应用于各种电子设备中。With further reference to FIG. 5, as an implementation of the method shown in the above figures, the present application provides an embodiment of a device for generating information, which corresponds to the method embodiment shown in FIG. 2. It can be applied to various electronic devices.

如图5所示，本实施例的生成信息的装置500包括：第一获取单元501，被配置成获取目标用户的用户信息和至少一个第一类待展示信息，第一类待展示信息为待向目标用户的终端推送以在终端上展示的信息和；生成单元502，被配置成对于至少一个第一类待展示信息中的每个第一类待展示信息，获取每个第一类待展示信息的特征信息；将特征信息和用户信息输入预先训练的被关注评分模型，得到用于表征每个第一类待展示信息被目标用户关注的程度的被关注评分；获取预设的补偿展示代价值，并基于补偿展示代价值和被关注评分，生成每个第一类待展示信息的第一类展示代价值。As shown in FIG. 5, the information generating apparatus 500 of this embodiment includes: a first acquiring unit 501 configured to acquire user information of a target user and at least one first-type to-be-displayed information, and the first-type to-be-displayed information is a Push to the terminal of the target user to display the information sum on the terminal; the generating unit 502 is configured to obtain each first-type to-be-displayed information for each first-type to-be-displayed information in at least one first-type to-be-displayed information Feature information of information; input feature information and user information into a pre-trained scoring model of interest to obtain a focus score that characterizes the degree to which each first-type information to be displayed is focused by the target user; obtain a preset compensation display code Value, and based on the compensation display generation value and the attention score, the first type of display generation value of each first type of information to be displayed is generated.

在本实施例中，第一获取单元501可以通过有线连接方式或者无线连接方式从远程或从本地获取至少一个待向目标用户的终端推送以在终端上展示的第一类待展示信息和目标用户的用户信息。其中，目标用户可以是待利用其使用的终端(例如图1所示的终端设备)浏览上述装置500推送的信息的用户。目标用户的用户信息可以用于表征目标用户的特征，目标用户的特征包括但不限于以下至少一种：目标用户的性别、年龄、兴趣等。In this embodiment, the first acquiring unit 501 may acquire at least one first type of information to be displayed and the target user to be pushed to the terminal of the target user to be displayed on the terminal from a remote or local via a wired connection or a wireless connection User information. The target user may be a user to browse the information pushed by the device 500 by using the terminal used by the terminal (for example, the terminal device shown in FIG. 1). The user information of the target user can be used to characterize the characteristics of the target user, and the characteristics of the target user include but are not limited to at least one of the following: the target user's gender, age, and interests.

在本实施例中，对于第一获取单元501获取的至少一个第一类待展示信息中的第一类待展示信息，上述生成单元502可以针对该第一类待展示信息执行步骤5021至步骤5023。In this embodiment, for the first type of to-be-displayed information among the at least one first type of to-be-displayed information acquired by the first obtaining unit 501, the above-mentioned generating unit 502 may perform steps 5021 to 5023 for the first type of to-be-displayed information .

在步骤5021中，获取每个第一类待展示信息的特征信息。In step 5021, feature information of each first type of information to be displayed is obtained.

在一实施例中，上述第一获取单元501可以从远程或从本地获取该第一类待展示信息的特征信息。其中，特征信息用于表征第一类待展示信息的特征。第一类待展示信息的特征可以包括但不限于以下至少一种：第一类待展示信息的标题、第一类待展示信息所属的类型、第一类待展示信息包括的链接地址等。特征信息可以是第一类待展示信息包括的信息(例如第一类待展示信息的标题)，也可以是预设的、与第一类待展示信息相关的信息(例如表征第一类待展示信息所属的类型的类型信息、第一类待展示信息的生成时间等)。In an embodiment, the first obtaining unit 501 may obtain the feature information of the first type of information to be displayed remotely or locally. Among them, the characteristic information is used to characterize the characteristics of the first type of information to be displayed. The characteristics of the first type of information to be displayed may include, but are not limited to, at least one of the following: the title of the first type of information to be displayed, the type of the first type of information to be displayed, the link address included in the first type of information to be displayed, and so on. The feature information may be information included in the first type of information to be displayed (such as the title of the first type of information to be displayed), or may be preset information related to the first type of information to be displayed (such as characterizing the first type of information to be displayed) Type information of the type to which the information belongs, generation time of the first type of information to be displayed, etc.).

在步骤5022中，将特征信息和用户信息输入预先训练的被关注评分模型，得到用于表征每个第一类待展示信息被目标用户关注的程度的被关注评分。In step 5022, the feature information and the user information are input into the pre-trained scoring model of attention, to obtain a scoring attention that is used to characterize the degree to which each first type of information to be displayed is focused by the target user.

上述被关注评分模型用于表征展示信息的特征信息、浏览展示信息的用户的用户信息与用于表征展示信息被用户关注的关注程度的被关注评分的对应关系。在一实施例中，被关注评分模型可以是技术人员预先基于对大量的特征信息、用户信息与被关注评分的统计而预先制定的、存储有多个特征信息、用户信息与被关注评分的对应关系表。其中，对应关系表中的被关注评分可以是技术人员预先针对用户信息表征的用户所属的用户群体(例如可以用预设的用户信息集合表征)对第一类待展示信息进行的评分所做的统计，并计算(例如将各个用户的评分取平均)得到的被关注评分。上述生成单元502可以根据输入的特征信息和用户信息，从对应关系表中查找对应的被关注评分。上述被关注评分模型也可以为基于预设的训练样本，利用机器学习方法对初始模型(例如神经网络)进行训练后得到的模型。The above attention rating model is used to characterize the feature information of the display information, the user information of the user browsing the display information, and the correspondence relationship of the attention ratings used to characterize the degree of attention of the display information to the user. In an embodiment, the focused scoring model may be pre-formulated by a technician based on statistics of a large amount of feature information, user information, and focused scores, and stores a plurality of feature information, user information, and correspondence of the focused scores Relational tables. Wherein, the attention score in the correspondence table may be a score made by the technician in advance for the user group to which the user characterized by the user information belongs (for example, it may be characterized by a preset user information set) to the first type of information to be displayed Make statistics and calculate (for example, average the scores of users) the scores of interest. The above-mentioned generating unit 502 may search for the corresponding score of interest from the correspondence table according to the input feature information and user information. The above noted scoring model may also be a model obtained after training an initial model (such as a neural network) using a machine learning method based on preset training samples.

在步骤5023中，获取预设的补偿展示代价值，并基于补偿展示代价值和被关注评分，生成每个第一类待展示信息的第一类展示代价值。In step 5023, a preset compensation display generation value is obtained, and based on the compensation display generation value and the attention score, a first type of display generation value for each first type of information to be displayed is generated.

作为示例，上述生成单元502可以将所获取的补偿展示代价值和所得到的被关注评分相乘，得到第一类展示代价值。或者，上述生成单元502可以将所获取的补偿展示代价值和所得到的被关注评分相乘后再进行其他方式的处理(例如将相乘后所得到的乘积与预设数值相加或相减)，得到第一类展示代价值。As an example, the above-mentioned generating unit 502 may multiply the obtained compensation display generation value and the obtained attention score to obtain the first type of display generation value. Alternatively, the above generation unit 502 may multiply the obtained compensation display generation value and the obtained attention score and then perform other processing (such as adding or subtracting the product obtained after the multiplication to a preset value) ), Get the first category of display value.

补偿展示代价值可以用于表征推送某个第一类待展示信息集合的推送者为了使用户浏览第一类待展示信息集合中的第一类待展示信息所付出的代价。上述推送者可以是使用上述装置500向用户推送第一类待展示信息的单位或个人。例如，假设用户使用安装在其使用的终端上的目标应用接收第一类待展示信息，则该目标应用的所有者即为推送者。实践中，推送者为了获得新的用户，会付出一定的代价，基于所付出的代价可以计算得到补偿展示代价值。例如将表征其付出的代价的代价值除以浏览第一类待展示信息集合中的第一类待展示信息的用户的数量，得到补偿展示代价值。The compensated display generation value can be used to characterize the price paid by a pusher who pushes a certain first-type to-be-displayed information set in order for the user to browse the first-type to-be-displayed information in the first-type to-be-displayed information set. The above-mentioned pusher may be a unit or an individual who uses the above-mentioned device 500 to push the first type of information to be displayed to the user. For example, if a user uses a target application installed on the terminal used by the user to receive the first type of information to be displayed, the owner of the target application is the pusher. In practice, the pusher will pay a certain price in order to obtain new users. Based on the paid price, the compensation display value can be calculated. For example, the generation value representing the price paid by it is divided by the number of users who browse the first type of information to be displayed in the first type of information to be displayed to obtain the compensation display generation value.

通常，被关注评分越高，表示目标用户对该第一类待展示信息的关注程度越高，将其乘以补偿展示代价值，所得到的第一类展示代价值越高，表征上述推送者向目标用户推送该第一类待展示信息所付出的代价越高，同时也表征第一类待展示信息的价值越大。Generally, the higher the attention rating, the higher the target user ’s attention to the first type of information to be displayed. Multiply it by the compensation display value. The higher the value of the first type of display generation, the higher the pusher ’s value. The higher the cost of pushing the first type of information to be displayed to the target user, it also signifies that the value of the first type of information to be displayed is greater.

在本实施例的一些实现方式中，该装置500还可以包括：第二获取单元(图中未示出)，被配置成获取预设的至少一个第二类待展示信息，以及第二类待展示信息对应的第二类展示代价值；合并单元(图中未示出)，被配置成将至少一个第二类展示代价值和至少一个第一类展示代价值合并为展示代价值集合，并将至少一个第一类待展示信息和至少一个第二类待展示信息合并为待展示信息集合；排序单元(图中未示出)，被配置成基于展示代价值集合中的展示代价值的大小，对展示代价值集合中的展示代价值进行排序；推送单元(图中未示出)，被配置成从排序后的展示代价值集合中选择展示代价值，并将所选择的展示代价值对应的、包含于待展示信息集合中的待展示信息推送至终端。In some implementation manners of this embodiment, the apparatus 500 may further include: a second acquiring unit (not shown in the figure) configured to acquire at least one preset second-type information to be displayed, and a second-type pending The second-generation display generation value corresponding to the display information; the merging unit (not shown in the figure) is configured to merge at least one second-type display generation value and at least one first-type display generation value into a display generation value set, and Combining at least one first-type information to be displayed and at least one second-type information to be displayed into a set of information to be displayed; a sorting unit (not shown in the figure) configured to be based on the size of the display generation value in the display generation value set , To sort the display generation value in the display generation value set; the push unit (not shown in the figure) is configured to select the display generation value from the sorted display generation value set and correspond the selected display generation value The to-be-displayed information contained in the to-be-displayed information set is pushed to the terminal.

在本实施例的一些实现方式中，被关注评分模型可以包括结果生成子模型和至少两个评分子模型；以及生成单元包括：第一生成模块(图中未示出)，被配置成将特征信息和用户信息输入至少两个评分子模型中的评分子模型中的每个评分子模型，得到每个评分子模型对应的评分；第二生成模块(图中未示出)，被配置成将所得到的评分输入结果生成子模型，得到表征该第一类待展示信息被目标用户关注的程度的被关注评分，其中，结果生成子模型设置为基于预设的权重，对所得到的评分进行加权求和，得到被关注评分。In some implementations of this embodiment, the focused scoring model may include a result generation sub-model and at least two scoring sub-models; and the generation unit includes: a first generation module (not shown in the figure) configured to configure features Information and user information are input to each of the scoring sub-models in at least two scoring sub-models to obtain the score corresponding to each scoring sub-model; the second generation module (not shown in the figure) is configured to The obtained score input result generates a sub-model to obtain a focused score that characterizes the degree to which the first type of information to be displayed is focused on by the target user, wherein the result-generated sub-model is set to perform score evaluation based on preset weights Weighted summation to get the attention score.

在本实施例的一些实现方式中，对于至少两个评分子模型中的评分子模型，每个评分子模型可以预先按照如下步骤训练得到：获取每个评分子模型对应的训练样本集合，其中，训练样本包括预设的样本第一类待展示信息的特征信息、浏览过样本第一类待展示信息的样本用户的样本用户信息，以及预先标注的、用于表征样本用户针对样本第一类待展示信息是否进行目标操作的标注信息；利用机器学习方法，将所获取的训练样本集合中的训练样本包括的特征信息和样本用户信息作为输入，将与输入的特征信息和样本用户信息对应的标注信息作为期望输出，训练得到每个评分子模型。In some implementations of this embodiment, for the scoring sub-models of at least two scoring sub-models, each scoring sub-model may be pre-trained according to the following steps: obtain a training sample set corresponding to each scoring sub-model, where, The training sample includes the feature information of the preset sample first type of information to be displayed, the sample user information of the sample user who has browsed the sample first type of information to be displayed, and the pre-marked characterization of the sample user for the sample first type of information Annotated information showing whether the information is performing target operations; using machine learning methods, using the feature information and sample user information included in the training samples in the acquired training sample set as input, and labeling the input feature information and sample user information The information is output as expected, and each scoring sub-model is trained.

在本实施例的一些实现方式中，生成第一类展示代价值所使用的补偿展示代价值可以预先通过如下步骤得到：获取预设的比例集合，其中，比例集合中的比例是在预设时间段内向预设的用户集合中的用户所使用的终端推送的第一类待展示信息的数量和第二类待展示信息的数量的比例；对于比例集合中的比例，确定每个比例对应的损失用户的数量，其中，损失用户是在预设时间段内，按照每个比例推送第一类待展示信息和第二类待展示信息，且在预设时间段后不使用终端接收推送的第一类待展示信息和第二类待展示信息的用户；基损失用户的数量，确定每个比例对应的补充代价值，其中，补充代价值用于表征增加与损失用户等量的用户所付出的代价；确定预设时间段内使用终端浏览按照每个比例推送的第一类待展示信息的用户的数量，将所确定的补充代价值除以所确定的用户数量，将所得到的商确定为每个比例对应的补偿展示代价值；从所确定的补偿展示代价值中，选择补偿展示代价值作为生成第一类展示代价值所使用的补偿展示代价值。In some implementations of this embodiment, the compensation display generation value used to generate the first type of display generation value may be obtained in advance through the following steps: obtaining a preset proportion set, where the proportion in the proportion set is at a preset time The ratio of the amount of information to be displayed in the first category to the amount of information to be displayed in the second category pushed to the terminal used by the user in the preset user set within the segment; for the proportion in the proportion set, determine the loss corresponding to each proportion The number of users, where the lost user is to push the first type of information to be displayed and the second type of information to be displayed according to each proportion within a preset time period, and to receive the pushed first without using the terminal after the preset time period Users of information to be displayed and information to be displayed of the second type; based on the number of users lost, determine the supplemental generation value corresponding to each proportion, where the supplemental generation value is used to characterize the price paid by the user who adds the same amount as the lost user ; Determine the number of users who use the terminal to browse the first type of information to be displayed in accordance with each proportion within the preset time period, and replace the determined The replacement value is divided by the determined number of users, and the obtained quotient is determined as the compensation display generation value corresponding to each proportion; from the determined compensation display generation value, the compensation display generation value is selected as the first type of display generation value The compensation used shows the generational value.

在本实施例的一些实现方式中，从所确定的补偿展示代价值中，选择补偿展示代价值作为生成第一类展示代价值所使用的补偿展示代价值，包括：对于所确定的补偿展示代价值中的每个补偿展示代价值，按照每个补偿展示代价值对应的比例推送的第二类待展示信息的第二类展示代价值的和作为每个补偿展示代价值对应的第二类总展示代价值；确定第二类总展示代价值与每个补偿展示代价值对应的补充代价值的差值；从所确定的差值中选择最大值，将最大值对应的补偿展示代价值作为生成第一类展示代价值所使用的补偿展示代价值。In some implementations of this embodiment, from the determined compensation display generation value, the compensation display generation value is selected as the compensation display generation value used to generate the first type of display generation value, including: for the determined compensation display generation value The value of each compensation display generation value in the value, the sum of the second type of display generation value of the second type of information to be displayed pushed according to the proportion corresponding to each compensation display generation value is taken as the total of the second type corresponding to each compensation display generation value Display generation value; determine the difference between the second type of total display generation value and the supplementary generation value corresponding to each compensated display generation value; select the maximum value from the determined differences, and use the compensation display generation value corresponding to the maximum value as the generation The first type of display generation value is the compensation display generation value used.

本申请的上述实施例提供的装置，通过使用被关注评分模型和预设的补偿展示代价值，得到第一类待展示信息的第一类展示代价值，从而可以对不同的第一类待展示信息和不同的用户信息进行分析，提高了生成第一类展示代价值的准确性，有助于在需要向用户展示多种待展示信息的情况下，提高将各种待展示信息展示给用户的针对性。The device provided by the above embodiment of the present application obtains the first-type display generation value of the first-type information to be displayed by using the focused scoring model and the preset compensation display generation value, so that different first-type display to-be-displayed values The analysis of information and different user information improves the accuracy of generating the first type of display generation value, which helps to improve the display of various types of information to be displayed to users when multiple types of information to be displayed need to be displayed to users. Targeted.

下面参考图6，其示出了适于用来实现本申请实施例的服务器的计算机***600的结构示意图。图6示出的服务器仅仅是一个示例，不应对本申请实施例的功能和使用范围带来任何限制。6, which shows a schematic structural diagram of a computer system 600 suitable for implementing the server of the embodiment of the present application. The server shown in FIG. 6 is only an example, and should not bring any limitation to the functions and usage scope of the embodiments of the present application.

如图6所示，计算机***600包括中央处理单元(Central Processing Unit，CPU)601，其可以根据存储在只读存储器(Read-Only Memory，ROM)602中的程序或者从存储部分608加载到随机访问存储器(Random Access Memory，RAM)603中的程序而执行各种适当的动作和处理。在RAM 603中，还存储有***600操作所需的各种程序和数据。CPU 601、ROM 602以及RAM 603通过总线604彼此相连。输入/输出(Input/Output，I/O)接口605也连接至总线604。As shown in FIG. 6, the computer system 600 includes a central processing unit (Central Processing Unit, CPU) 601, which can be loaded into a random unit according to a program stored in a read-only memory (Read-Only Memory, ROM) 602 or from the storage section 608 The program in the memory (Random Access Memory) 603 is accessed to perform various appropriate actions and processes. In the RAM 603, various programs and data necessary for the operation of the system 600 are also stored. The CPU 601, ROM 602, and RAM 603 are connected to each other through a bus 604. An input / output (Input / Output, I / O) interface 605 is also connected to the bus 604.

以下部件连接至I/O接口605：包括键盘、鼠标等的输入部分606；包括诸如液晶显示器(Liquid Crystal Display，LCD)等以及扬声器等的输出部分607；包括硬盘等的存储部分608；以及包括诸如局域网(Local Area Network，LAN)卡、调制解调器等的网络接口卡的通信部分609。通信部分609经由诸如因特网的网络执行通信处理。驱动器610也根据需要连接至I/O接口605。可拆卸介质611，诸如磁盘、光盘、磁光盘、半导体存储器等等，根据需要安装在驱动器610上，以便于从其上读出的计算机程序根据需要被安装入存储部分608。The following components are connected to the I / O interface 605: an input section 606 including a keyboard, a mouse, etc .; an output section 607 including a liquid crystal display (Liquid Crystal Display, LCD), etc., and a speaker; a storage section 608 including a hard disk, etc .; and including The communication portion 609 of a network interface card such as a local area network (Local Area Network, LAN) card, modem, or the like. The communication section 609 performs communication processing via a network such as the Internet. The driver 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like, is installed on the drive 610 as necessary, so that the computer program read out therefrom is installed into the storage section 608 as needed.

特别地，根据本公开的实施例，上文参考流程图描述的过程可以被实现为计算机软件程序。例如，本公开的实施例包括一种计算机程序产品，其包括承载在计算机可读介质上的计算机程序，该计算机程序包含设置为执行流程图所示的方法的程序代码。在这样的实施例中，该计算机程序可以通过通信部分609从网络上被下载和安装，和从可拆卸介质611被安装其中的至少一种方式被安装。在该计算机程序被中央处理单元(CPU)601执行时，执行本申请的方法中限定的上述功能。In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart may be implemented as a computer software program. For example, embodiments of the present disclosure include a computer program product that includes a computer program carried on a computer-readable medium, the computer program containing program code configured to perform the method shown in the flowchart. In such an embodiment, the computer program may be downloaded and installed from the network through the communication section 609, and installed from at least one of the ways in which the removable medium 611 is installed. When the computer program is executed by the central processing unit (CPU) 601, the above-mentioned functions defined in the method of the present application are executed.

需要说明的是，本申请所述的计算机可读介质可以是计算机可读信号介质或者计算机可读介质或者是上述两者的任意组合。计算机可读介质例如可以是——但不限于——电、磁、光、电磁、红外线、或半导体的***、装置或器件，或者任意以上的组合。计算机可读介质的更具体的例子可以包括但不限于：具有至少一个导线的电连接、便携式计算机磁盘、硬盘、随机访问存储器(Random Access Memory，RAM)、只读存储器(Read Only Memory，ROM)、可擦式可编程只读存储器(Erasable Programmable Read-Only Memory，EPROM)或闪存、光纤、便携式紧凑磁盘只读存储器(Compact Disc Read-Only Memory，CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。在本申请中，计算机可读介质可以是任何包含或存储程序的有形介质，该程序可以被指令执行***、装置或者器件使用或者与其结合使用。而在本申请中，计算机可读的信号介质可以包括在基带中或者作为载波一部分传播的数据信号，其中承载了计算机可读的程序代码。这种传播的数据信号可以采用多种形式，包括但不限于电磁信号、光信号或上述的任意合适的组合。计算机可读的信号介质还可以是计算机可读介质以外的任何计算机可读介质，该计算机可读介质可以发送、传播或者传输用于由指令执行***、装置或者器件使用或者与其结合使用的程序。计算机可读介质上包含的程序代码可以用任何适当的介质传输，包括但不限于：无线、电线、光缆、射频(Radio Frequency，RF)等等，或者上述的任意合适的组合。It should be noted that, the computer-readable medium described in this application may be a computer-readable signal medium or a computer-readable medium, or any combination of the two. The computer-readable medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination of the above. More specific examples of computer-readable media may include, but are not limited to: electrical connections with at least one wire, portable computer disk, hard disk, random access memory (Random Access Memory, RAM), read-only memory (Read Only Memory, ROM) , Erasable programmable read-only memory (Erasable Programmable Read-Only Memory, EPROM) or flash memory, optical fiber, portable compact disk read-only memory (Compact Disc Read-Only Memory, CD-ROM), optical storage devices, magnetic storage devices , Or any suitable combination of the above. In this application, the computer-readable medium may be any tangible medium that contains or stores a program, and the program may be used by or in combination with an instruction execution system, apparatus, or device. In this application, the computer-readable signal medium may include a data signal that is propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal can take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. The computer-readable signal medium may also be any computer-readable medium other than a computer-readable medium, and the computer-readable medium may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted on any appropriate medium, including but not limited to: wireless, wire, optical cable, radio frequency (Radio Frequency, RF), etc., or any suitable combination of the foregoing.

可以以一种或多种程序设计语言或其组合来编写用于执行本申请的操作的计算机程序代码，所述程序设计语言包括面向对象的程序设计语言—诸如Java、Smalltalk、C++，还包括常规的过程式程序设计语言—诸如“C”语言或类似的程序设计语言。程序代码可以完全地在用户计算机上执行、部分地在用户计算机上执行、作为一个独立的软件包执行、部分在用户计算机上部分在远程计算机上执行、或者完全在远程计算机或服务器上执行。在涉及远程计算机的情形中，远程计算机可以通过任意种类的网络——包括局域网(LAN)或广域网(Wide Area Network，WAN)—连接到用户计算机，或者，可以连接到外部计算机(例如利用因特网服务提供商来通过因特网连接)。The computer program code for performing the operations of the present application may be written in one or more programming languages or a combination thereof, the programming languages including object-oriented programming languages such as Java, Smalltalk, C ++, as well as conventional Procedural programming language-such as "C" language or similar programming language. The program code may be executed entirely on the user's computer, partly on the user's computer, as an independent software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server. In situations involving remote computers, the remote computer can be connected to the user's computer through any kind of network, including a local area network (LAN) or a wide area network (Wide Area Network, WAN), or can be connected to an external computer (for example, using Internet services Provider to connect via the Internet).

附图中的流程图和框图，图示了按照本申请各种实施例的***、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上，流程图或框图中的每个方框可以代表一个模块、程序段、或代码的一部分，该模块、程序段、或代码的一部分包含至少一个用于实现规定的逻辑功能的可执行指令。也应当注意，在有些作为替换的实现中，方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如，两个接连地表示的方框实际上可以基本并行地执行，它们有时也可以按相反的顺序执行，这依据所涉及的功能而定。也要注意的是，框图和流程图中至少之一的每个方框、以及框图和流程图中的至少之一方框的组合，可以用执行规定的功能或操作的专用的基于硬件的***来实现，或者可以用专用硬件与计算机指令的组合来实现。The flowchart and block diagrams in the drawings illustrate the possible implementation architecture, functions, and operations of the system, method, and computer program product according to various embodiments of the present application. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the module, program segment, or part of the code contains at least one Execute instructions. It should also be noted that in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks represented in succession may actually be executed in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block of at least one of the block diagrams and flowcharts, and the combination of at least one of the block diagrams and flowcharts, can be implemented with a dedicated hardware-based system that performs the specified function or operation It can be realized by a combination of dedicated hardware and computer instructions.

描述于本申请实施例中所涉及到的单元可以通过软件的方式实现，也可以通过硬件的方式来实现。所描述的单元也可以设置在处理器中，例如，可以描述为：一种处理器包括第一获取单元和生成单元。其中，这些单元的名称在某种情况下并不构成对该单元本身的限定，例如，第一获取单元还可以被描述为“获取至少一个待向目标用户的终端推送以在终端上展示的第一类待展示信息和目标用户的用户信息的单元”。The units described in the embodiments of the present application may be implemented in software or hardware. The described unit may also be provided in the processor. For example, it may be described as: a processor includes a first acquiring unit and a generating unit. Among them, the names of these units do not constitute a limitation on the unit itself under certain circumstances. For example, the first obtaining unit may also be described as “obtaining at least one terminal to be pushed to the terminal of the target user to be displayed on the terminal A unit of information to be displayed and user information of the target user ".

作为另一方面，本申请还提供了一种计算机可读介质，该计算机可读介质可以是上述实施例中描述的服务器中所包含的；也可以是单独存在，而未装配入该服务器中。上述计算机可读介质承载有至少一个程序，当上述至少一个程序被该服务器执行时，使得该服务器：获取至少一个待向目标用户的终端推送以在终端上展示的第一类待展示信息和目标用户的用户信息；对于所获取的至少一个第一类待展示信息中的每个第一类待展示信息，获取该第一类待展示信息的特征信息；将所获取的特征信息和用户信息输入预先训练的被关注评分模型，得到用于表征每个第一类待展示信息被目标用户关注的程度的被关注评分；获取预设的补偿展示代价值，以及基于所获取的补偿展示代价值和所得到的被关注评分，生成每个第一类待展示信息的第一类展示代价值。As another aspect, the present application also provides a computer-readable medium, which may be contained in the server described in the foregoing embodiments; or may exist alone without being assembled into the server. The computer-readable medium carries at least one program, and when the at least one program is executed by the server, the server causes the server to acquire at least one first type of information to be displayed and a target to be pushed to the terminal of the target user to be displayed on the terminal User information of the user; for each first-type to-be-displayed information in the at least one first-type to-be-displayed information obtained, obtain characteristic information of the first-type to-be-displayed information; input the obtained characteristic information and user information A pre-trained attention-scoring model to obtain a attention score that characterizes the degree to which each first-type information to be displayed is focused by the target user; obtain a preset compensation display generation value, and based on the obtained compensation display generation value and The obtained attention score generates the first-generation display generation value of each first-type information to be displayed.

Claims

一种生成信息的方法，包括：A method of generating information, including:

获取所述目标用户的用户信息和至少一个第一类待展示信息，所述第一类待展示信息为待向目标用户的终端推送以在所述终端上展示的信息；Acquiring user information of the target user and at least one first type of information to be displayed, where the first type of information to be displayed is information to be pushed to the terminal of the target user to be displayed on the terminal;

对于至少一个第一类待展示信息中的每个第一类待展示信息，获取每个第一类待展示信息的特征信息；将所述特征信息和所述用户信息输入预先训练的被关注评分模型，得到用于表征每个第一类待展示信息被所述目标用户关注的程度的被关注评分；获取预设的补偿展示代价值，并基于所述补偿展示代价值和所述被关注评分，生成每个第一类待展示信息的第一类展示代价值。For each first-type to-be-displayed information in at least one first-type to-be-displayed information, obtain feature information of each first-type to-be-displayed information; enter the feature information and the user information into a pre-trained attention score A model to obtain a focused score that characterizes the degree to which each first type of information to be displayed is focused by the target user; obtain a preset compensated display generation value, and based on the compensated display generation value and the focused score To generate the first-generation display generation value of each first-type information to be displayed.
根据权利要求1所述的方法，还包括：The method of claim 1, further comprising:

获取预设的至少一个第二类待展示信息，以及所述第二类待展示信息对应的第二类展示代价值；Acquiring at least one preset second-type information to be displayed and the second-type display substitute value corresponding to the second-type information to be displayed;

将所述至少一个第二类展示代价值和至少一个第一类展示代价值合并为展示代价值集合，并将所述至少一个第一类待展示信息和所述至少一个第二类待展示信息合并为待展示信息集合；Combining the at least one second-generation display generation value and at least one first-type display generation value into a display generation value set, and combining the at least one first-type information to be displayed and the at least one second-type information to be displayed Merged into a collection of information to be displayed;

基于所述展示代价值集合中的展示代价值的大小，对所述展示代价值集合中的展示代价值进行排序；Sort the display generation value in the display generation value set based on the size of the display generation value in the display generation value set;

从排序后的展示代价值集合中选择展示代价值，并将所选择的展示代价值对应的包含于所述待展示信息集合中的待展示信息推送至所述终端。Select the display generation value from the sorted display generation value set, and push the to-be-displayed information included in the to-be-displayed information set corresponding to the selected display generation value to the terminal.
根据权利要求1所述的方法，其中，所述被关注评分模型包括结果生成子模型和至少两个评分子模型；The method according to claim 1, wherein the focused scoring model includes a result generation sub-model and at least two scoring sub-models;

所述将所述特征信息和所述用户信息输入预先训练的被关注评分模型，得到用于表征每个第一类待展示信息被所述目标用户关注的程度的被关注评分，包括：The inputting the feature information and the user information into a pre-trained scoring model of attention to obtain a scoring of attention for characterizing the degree to which each first type of information to be displayed is focused by the target user includes:

将所述特征信息和所述用户信息输入所述至少两个评分子模型中的每个评分子模型，得到每个评分子模型对应的评分；Input the feature information and the user information into each of the at least two rating sub-models to obtain a rating corresponding to each rating sub-model;

将所得到的评分输入所述结果生成子模型，得到表征每个第一类待展示信息被所述目标用户关注的程度的被关注评分，其中，所述结果生成子模型设置为基于预设的权重，对所述至少两个评分子模型的评分进行加权求和，得到被关注评分。Input the obtained score into the result generation sub-model to obtain a focused score that characterizes the degree to which each first type of information to be displayed is focused by the target user, wherein the result generation sub-model is set based on a preset The weighting is to perform weighted summation on the scores of the at least two scoring sub-models to obtain the focused score.
根据权利要求3所述的方法，其中，所述至少两个评分子模型中的每个评分子模型，预先按照如下步骤训练得到：The method according to claim 3, wherein each of the at least two scoring sub-models is pre-trained according to the following steps:

获取每个评分子模型对应的训练样本集合，其中，所述训练样本包括预设的样本第一类待展示信息的特征信息、浏览过所述样本第一类待展示信息的样本用户的样本用户信息，以及预先标注的用于表征所述样本用户针对所述样本第一类待展示信息是否进行目标操作的标注信息；Obtain a training sample set corresponding to each scoring sub-model, where the training sample includes preset feature information of the sample first-type information to be displayed, and sample users who have browsed the sample first-type information to be displayed Information, and pre-labeled labeling information used to characterize whether the sample user performs a target operation on the first type of information to be displayed in the sample;

利用机器学习方法，将所获取的训练样本集合中的训练样本包括的特征信息和样本用户信息作为输入，将与输入的特征信息和样本用户信息对应的标注信息作为期望输出，训练得到每个评分子模型。Using machine learning methods, the training sample in the obtained training sample set includes the feature information and sample user information as input, and the label information corresponding to the input feature information and sample user information is used as the expected output, and each score is obtained by training Submodel.
根据权利要求1-4之一所述的方法，其中，生成每个第一类展示代价值所使用的补偿展示代价值，预先通过如下步骤得到：The method according to any one of claims 1 to 4, wherein the compensation display generation value used for generating each first type of display generation value is obtained in advance by the following steps:

获取预设的比例集合，其中，所述比例集合中的比例是在预设时间段内向预设的用户集合中的用户所使用的终端推送的第一类待展示信息的数量和第二类待展示信息的数量的比例；Obtain a preset ratio set, where the ratio in the ratio set is the number of first-type information to be displayed and the second-type information to be displayed to the terminal used by the user in the preset user set within a preset time period Proportion of the amount of information displayed;

对于所述比例集合中的比例，确定每个比例对应的损失用户的数量，其中，所述损失用户是在预设时间段内，按照所述每个比例推送第一类待展示信息和第二类待展示信息，且在所述预设时间段后不使用终端接收推送的第一类待展示信息和第二类待展示信息的用户；基于所述损失用户的数量，确定所述每个比例对应的补充代价值，其中，所述补充代价值用于表征增加与损失用户等量的用户所付出的代价；确定所述预设时间段内使用终端浏览按照所述每个比例推送的第一类待展示信息的用户的数量，将所确定的所述补充代价值除以所确定的用户数量，将所得到的商确定为所述每个比例对应的补偿展示代价值；For the proportions in the proportion set, determine the number of lost users corresponding to each proportion, wherein the lost users push the first type of information to be displayed and the second according to each proportion within a preset time period Users of the type to be displayed, and who do not use the terminal to receive the first type of information to be displayed and the second type of information to be displayed after the preset period of time; based on the number of lost users, determine each proportion Corresponding supplemental generation value, wherein the supplemental generation value is used to characterize the price paid by users who have increased and lost the same amount of users; it is determined that using the terminal to browse the first pushed in accordance with each proportion within the preset time period The number of users of the type of information to be displayed is divided by the determined supplemental generation value by the determined number of users, and the obtained quotient is determined as the compensation display generation value corresponding to each proportion;

从所确定的补偿展示代价值中，选择补偿展示代价值作为生成第一类展示代价值所使用的补偿展示代价值。From the determined compensation display generation value, the compensation display generation value is selected as the compensation display generation value used to generate the first type of display generation value.
根据权利要求5所述的方法，其中，所述从所确定的补偿展示代价值中，选择补偿展示代价值作为生成第一类展示代价值所使用的补偿展示代价值，包括：The method according to claim 5, wherein the selection of the compensation display generation value as the compensation display generation value used to generate the first type of display generation value from the determined compensation display generation value includes:

对于所确定的补偿展示代价值中的每个补偿展示代价值，将按照每个补偿展示代价值对应的比例推送的第二类待展示信息的第二类展示代价值的和作为每个补偿展示代价值对应的第二类总展示代价值；确定第二类总展示代价值与每个补偿展示代价值对应的补充代价值的差值；For each compensated display generation value in the determined compensation display generation value, the sum of the second type of display generation value of the second type of information to be displayed pushed according to the proportion corresponding to each compensation display generation value is taken as each compensation display The second category total display generation value corresponding to the generation value; determine the difference between the second category total display generation value and the supplementary generation value corresponding to each compensated display generation value;

从所确定的差值中选择最大值，将所述最大值对应的补偿展示代价值作为生成第一类展示代价值所使用的补偿展示代价值。A maximum value is selected from the determined difference values, and the compensation display generation value corresponding to the maximum value is used as the compensation display generation value used to generate the first type of display generation value.
一种生成信息的装置，包括：A device for generating information, including:

第一获取单元，被配置成获取所述目标用户的用户信息和至少一个第一类待展示信息，所述第一类待展示信息为待向目标用户的终端推送以在所述终端上展示的信息；The first obtaining unit is configured to obtain the user information of the target user and at least one first-type to-be-displayed information, the first-type to-be-displayed information is to be pushed to a terminal of the target user to be displayed on the terminal information;

生成单元，被配置成对于至少一个第一类待展示信息中的每个第一类待展示信息，获取每个第一类待展示信息的特征信息；将所述特征信息和所述用户信息输入预先训练的被关注评分模型，得到用于表征每个第一类待展示信息被所述目标用户关注的程度的被关注评分；获取预设的补偿展示代价值，并基于所述补偿展示代价值和所述被关注评分，生成每个第一类待展示信息的第一类展示代价值。The generating unit is configured to acquire, for each first-type to-be-displayed information in at least one first-type to-be-displayed information, characteristic information of each first-type to-be-displayed information; input the characteristic information and the user information A pre-trained attention-scoring model to obtain a concern score that characterizes the degree to which each first-type information to be displayed is focused by the target user; obtain a preset compensation display generation value, and based on the compensation display generation value And the focused score to generate the first-generation display generation value of each first-type information to be displayed.
根据权利要求7所述的装置，还包括：The device according to claim 7, further comprising:

第二获取单元，被配置成获取预设的至少一个第二类待展示信息，以及所述第二类待展示信息对应的第二类展示代价值；The second obtaining unit is configured to obtain at least one preset second-type to-be-displayed information and the second-type display substitute value corresponding to the second-type to-be-displayed information;

合并单元，被配置成将所述至少一个第二类展示代价值和至少一个第一类展示代价值合并为展示代价值集合，并将所述至少一个第一类待展示信息和所述至少一个第二类待展示信息合并为待展示信息集合；A merging unit, configured to merge the at least one second-generation display generation value and at least one first-type display generation value into a display generation value set, and merge the at least one first-type to-be-displayed information and the at least one The second type of information to be displayed is merged into a collection of information to be displayed;

排序单元，被配置成基于所述展示代价值集合中的展示代价值的大小，对所述展示代价值集合中的展示代价值进行排序；A sorting unit configured to sort the display generation value in the display generation value set based on the size of the display generation value in the display generation value set;

推送单元，被配置成从排序后的展示代价值集合中选择展示代价值，并将所选择的展示代价值对应的包含于所述待展示信息集合中的待展示信息推送至所述终端。The pushing unit is configured to select the display generation value from the sorted display generation value set, and push the to-be-displayed information included in the to-be-displayed information set corresponding to the selected display generation value to the terminal.
根据权利要求7所述的装置，其中，所述被关注评分模型包括结果生成子模型和至少两个评分子模型；The apparatus according to claim 7, wherein the focused scoring model includes a result generation sub-model and at least two scoring sub-models;

所述生成单元包括：The generating unit includes:

第一生成模块，被配置成将所述特征信息和所述用户信息输入所述至少两个评分子模型中的每个评分子模型，得到每个评分子模型对应的评分；The first generation module is configured to input the feature information and the user information into each of the at least two rating sub-models to obtain a rating corresponding to each rating sub-model;

第二生成模块，被配置成将所得到的评分输入所述结果生成子模型，得到表征每个第一类待展示信息被所述目标用户关注的程度的被关注评分，其中，所述结果生成子模型设置为基于预设的权重，对所述至少两个评分子模型的评分进行加权求和，得到被关注评分。The second generation module is configured to input the obtained score into the result generation sub-model to obtain a focused score characterizing the degree to which each first type of information to be displayed is focused by the target user, wherein the result is generated The sub-model is set to perform weighted summation of the scores of the at least two scoring sub-models based on preset weights to obtain the score of interest.
根据权利要求9所述的装置，其中，所述至少两个评分子模型中的每个评分子模型，预先按照如下步骤训练得到：The apparatus according to claim 9, wherein each of the at least two scoring sub-models is pre-trained according to the following steps:

获取每个评分子模型对应的训练样本集合，其中，所述训练样本包括预设的样本第一类待展示信息的特征信息、浏览过所述样本第一类待展示信息的样本用户的样本用户信息，以及预先标注的用于表征所述样本用户针对所述样本第一类待展示信息是否进行目标操作的标注信息；Obtain a training sample set corresponding to each scoring sub-model, where the training sample includes preset feature information of the sample first-type information to be displayed, and sample users who have browsed the sample first-type information to be displayed Information, and pre-labeled labeling information used to characterize whether the sample user performs a target operation on the first type of information to be displayed in the sample;

利用机器学习方法，将所获取的训练样本集合中的训练样本包括的特征信息和样本用户信息作为输入，将与输入的特征信息和样本用户信息对应的标注信息作为期望输出，训练得到每个评分子模型。Using machine learning methods, the training sample in the obtained training sample set includes the feature information and sample user information as input, and the label information corresponding to the input feature information and sample user information is used as the expected output, and each score is obtained by training Submodel.
根据权利要求7-10之一所述的装置，其中，生成每个第一类展示代价值所使用的补偿展示代价值，预先通过如下步骤得到：The apparatus according to any one of claims 7-10, wherein the compensation display generation value used for generating each first type of display generation value is obtained in advance by the following steps:

获取预设的比例集合，其中，所述比例集合中的比例是在预设时间段内向预设的用户集合中的用户所使用的终端推送的第一类待展示信息的数量和第二类待展示信息的数量的比例；Obtain a preset ratio set, where the ratio in the ratio set is the number of first-type information to be displayed and the second-type information to be displayed to the terminal used by the user in the preset user set within a preset time period Proportion of the amount of information displayed;

对于所述比例集合中的比例，确定每个比例对应的损失用户的数量，其中，所述损失用户是在预设时间段内，按照所述每个比例推送第一类待展示信息和第二类待展示信息，且在所述预设时间段后不使用终端接收推送的第一类待展示信息和第二类待展示信息的用户；基于所述损失用户的数量，确定所述每个比例对应的补充代价值，其中，所述补充代价值用于表征增加与损失用户等量的用户所付出的代价；确定所述预设时间段内使用终端浏览按照所述每个比例推送的第一类待展示信息的用户的数量，将所确定的所述补充代价值除以所确定的用户数量，将所得到的商确定为所述每个比例对应的补偿展示代价值；For the proportions in the proportion set, determine the number of lost users corresponding to each proportion, wherein the lost users push the first type of information to be displayed and the second according to each proportion within a preset time period Users of the type to be displayed, and who do not use the terminal to receive the first type of information to be displayed and the second type of information to be displayed after the preset period of time; based on the number of lost users, determine each proportion Corresponding supplemental generation value, wherein the supplemental generation value is used to characterize the price paid by users who have increased and lost the same amount of users; it is determined that using the terminal to browse the first pushed in accordance with each proportion within the preset time period The number of users of the type of information to be displayed is divided by the determined supplemental generation value by the determined number of users, and the obtained quotient is determined as the compensation display generation value corresponding to each proportion;

从所确定的补偿展示代价值中，选择补偿展示代价值作为生成第一类展示代价值所使用的补偿展示代价值。From the determined compensation display generation value, the compensation display generation value is selected as the compensation display generation value used to generate the first type of display generation value.
根据权利要求11所述的装置，其中，所述从所确定的补偿展示代价值中，选择补偿展示代价值作为生成第一类展示代价值所使用的补偿展示代价值，包括：The apparatus according to claim 11, wherein the selection of the compensation display generation value as the compensation display generation value used to generate the first type of display generation value from the determined compensation display generation value includes:

对于所确定的补偿展示代价值中的每个补偿展示代价值，按照每个补偿展示代价值对应的比例推送的第二类待展示信息的第二类展示代价值的和作为每个补偿展示代价值对应的第二类总展示代价值；确定第二类总展示代价值与每个补偿展示代价值对应的补充代价值的差值；For each compensation display generation value in the determined compensation display generation value, the sum of the second type of display generation value of the second type of information to be displayed pushed according to the proportion corresponding to each compensation display generation value is taken as each compensation display generation The second category total display generation value corresponding to the value; determine the difference between the second category total display generation value and the supplemental generation value corresponding to each compensated display generation value;

从所确定的差值中选择最大值，将所述最大值对应的补偿展示代价值作为生成第一类展示代价值所使用的补偿展示代价值。A maximum value is selected from the determined difference values, and the compensation display generation value corresponding to the maximum value is used as the compensation display generation value used to generate the first type of display generation value.
一种服务器，包括：A server, including:

至少一个处理器；At least one processor;

存储装置，其上存储有至少一个程序，A storage device on which at least one program is stored,

当所述至少一个程序被所述至少一个处理器执行，使得所述至少一个处理器实现如权利要求1-6中任一所述的方法。When the at least one program is executed by the at least one processor, the at least one processor implements the method according to any one of claims 1-6.
一种计算机可读介质，其上存储有计算机程序，其中，所述程序被处理器执行时实现如权利要求1-6中任一所述的方法。A computer-readable medium on which a computer program is stored, wherein the program is executed by a processor to implement the method according to any one of claims 1-6.