CN114996571A - Information pushing method and device, storage medium and electronic equipment - Google Patents
Information pushing method and device, storage medium and electronic equipment Download PDFInfo
- Publication number
- CN114996571A CN114996571A CN202210538857.3A CN202210538857A CN114996571A CN 114996571 A CN114996571 A CN 114996571A CN 202210538857 A CN202210538857 A CN 202210538857A CN 114996571 A CN114996571 A CN 114996571A
- Authority
- CN
- China
- Prior art keywords
- user
- information
- target
- characteristic
- pushed
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/335—Filtering based on additional data, e.g. user or group profiles
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses an information pushing method, an information pushing device, a storage medium and electronic equipment, wherein the method comprises the following steps: extracting user characteristics; extracting user characteristics of a target user; extracting message characteristics of target information to be pushed; determining whether to push the target information to the target user as preferred information or not based on a processing result of a double-layer network model on the user characteristic and the message characteristic under the condition that the push mode of the target information is determined to be a model push mode based on a matching strategy; and comparing the user characteristics with the message characteristics based on a preset rule to determine whether to push the target information to the target user as preferred information or not under the condition that the push mode of the target information is determined to be a matching rule push mode based on a matching strategy. The method pushes different information for different users through different characteristics of the users, achieves personalized information pushing, and improves the information utilization rate and the efficiency.
Description
Technical Field
The present invention relates to the field of information processing, and in particular, to an information pushing method and apparatus, a storage medium, and an electronic device.
Background
With the rapid development of the internet technology, the internet information is explosively increased, the time and the energy of a user are limited, and in order to improve the user stickiness, the user needs to be analyzed, and information which the user is interested in is accurately pushed to the user. Taking the example of pushing news, current news, news updates and the like are all obtained by manually screening news by operators and pushing the news to a whole number of users, and the screened news is the same for one of the users, namely, news received by all users is the same, so that the interest difference of different users is ignored, and the screened articles cannot be guaranteed to be liked by the received users. The click rate of the pushed news is low and fluctuation is large. The effectiveness of pushing news depends heavily on manual experience and the popularity of the news.
In order to solve the problem, personalized information push needs to be realized through a proper push algorithm, so that the click rate of the pushed information is improved.
Disclosure of Invention
In view of this, the present invention provides an information pushing method and apparatus, a storage medium, and an electronic device, which can solve the technical problem that it is difficult to effectively push information to a user.
The invention provides an information pushing method. The method comprises the following steps:
extracting user characteristics of a target user;
extracting message characteristics of target information to be pushed;
determining whether to push the target information to the target user as preferred information or not based on a processing result of a double-layer network model on the user characteristic and the message characteristic under the condition that a push mode of the target information is determined to be a model push mode based on a matching strategy, wherein a first layer of the double-layer network model is used for receiving input characteristic information and selecting a corresponding characteristic subset from the received input characteristic information, and a second layer of the double-layer network model is used for learning the weight of each characteristic in the first layer selected characteristic subset; and
and under the condition that the pushing mode of the target information is determined to be the matching rule pushing mode based on the matching strategy, comparing the user characteristic with the message characteristic based on a preset rule to determine whether to push the target information to the target user as preferred information.
Optionally, acquiring basic data of the target user; and
and mapping the corresponding user characteristics into numerical values based on the basic data of the target user to form the basic characteristics of the target user.
Optionally, obtaining a custom user characteristic for the target user;
and mapping the user-defined user characteristics to corresponding numerical values based on the basic characteristics of the target user.
Optionally, determining that the push mode of the target information is a regular push mode when the correlation degree between the message characteristic of the target information and the user characteristic of the target user is higher than or equal to a preset threshold;
and under the condition that the correlation degree of the message characteristics of the target information and the user characteristics of the target user is lower than the preset threshold value, determining that the pushing mode of the target information is a model pushing mode.
Optionally, the pushing based on the matching rule is performed, that is, a matching rule is set, the matching rule presets a subset of the user characteristics and a subset of the message characteristics of the target information to be pushed, and if there are characteristics that are mutually matched in the subset of the user characteristics and the subset of the message characteristics of the target information to be pushed, the target information to be pushed is used as preferred pushing information to be pushed to the target user.
Optionally, before pushing the preferred information to the target user, performing deduplication processing on the preferred pushed information.
The invention provides an information pushing device. The device comprises:
the user characteristic extraction module: configured to extract user features of a target user;
the message characteristic extraction module: the method comprises the steps of configuring to extract message characteristics of target information to be pushed;
a mode determination module: determining whether to push the target information to the target user as preferred information or not based on a processing result of a double-layer network model to the user characteristic and the message characteristic under the condition that a push mode of the target information is determined to be a model push mode based on a matching strategy, wherein a first layer of the double-layer network model is used for receiving input characteristic information and selecting a corresponding characteristic subset from the received input characteristic information, and a second layer of the double-layer network model is used for learning the weight of each characteristic in the characteristic subset selected by the first layer; and
and comparing the user characteristics with the message characteristics based on a preset rule to determine whether to push the target information to the target user as preferred information or not under the condition that the push mode of the target information is determined to be a rule push mode based on a matching strategy.
The invention provides a storage medium, wherein a plurality of instructions are stored in the storage medium; the plurality of instructions for loading and executing the method of any of the preceding claims by a processor.
The present invention provides an electronic device, including:
a processor for executing a plurality of instructions;
a memory to store a plurality of instructions;
wherein the instructions are for storage by the memory and for loading and executing by the processor the method of any preceding claim.
The method constructs a news content pool, extracts multidimensional characteristics from the user and the article content corresponding to the news, and removes the duplication of the calculated article corresponding to the news to be pushed. A large number of articles are obtained from a content pool, different articles are pushed for each user by utilizing a pushing algorithm model, personalized pushing (push) of the articles is achieved, different news flashes are pushed for each user through attributes such as behaviors of different characteristics of each user, thousands of people and thousands of faces of information pushing are achieved, the flash click rate is improved, the efficiency is improved, and fluctuation is reduced.
The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.
Drawings
The above and other objects, features and advantages of the present invention will become more apparent by describing in more detail embodiments of the present invention with reference to the attached drawings. The accompanying drawings are included to provide a further understanding of the embodiments of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings, like reference numbers generally represent like parts or steps.
Fig. 1 is a schematic flow chart of a method provided by an exemplary embodiment of the present invention.
Fig. 2 is a schematic diagram illustrating an effect of a method according to an exemplary embodiment of the present invention.
Fig. 3 is a schematic structural diagram of an apparatus according to an exemplary embodiment of the present invention.
Fig. 4 is a structure of an electronic device according to an exemplary embodiment of the present invention.
Detailed Description
Hereinafter, example embodiments according to the present invention will be described in detail with reference to the accompanying drawings. It is to be understood that the described embodiments are merely a subset of embodiments of the invention and not all embodiments of the invention, with the understanding that the invention is not limited to the example embodiments described herein.
It should be noted that: the relative arrangement of the components and steps, the numerical expressions and numerical values set forth in these embodiments do not limit the scope of the present invention unless it is specifically stated otherwise.
It will be understood by those of skill in the art that the terms "first," "second," and the like in the embodiments of the present invention are used merely to distinguish one element, step, device, module, or the like from another element, and do not denote any particular technical or logical order therebetween.
It should also be understood that in embodiments of the present invention, "a plurality" may refer to two or more and "at least one" may refer to one, two or more.
It is also to be understood that any reference to any component, data, or structure in the embodiments of the invention may be generally understood as one or more, unless explicitly defined otherwise or stated to the contrary hereinafter.
In addition, the term "and/or" in the present invention is only one kind of association relationship describing the associated object, and means that there may be three kinds of relationships, for example, a and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone. In the present invention, the character "/" generally indicates that the preceding and following related objects are in an "or" relationship.
It should also be understood that the description of the embodiments of the present invention emphasizes the differences between the embodiments, and the same or similar parts may be referred to each other, so that the descriptions thereof are omitted for brevity.
Meanwhile, it should be understood that the sizes of the respective portions shown in the drawings are not drawn in an actual proportional relationship for the convenience of description.
The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses.
Techniques, methods, and apparatus known to those of ordinary skill in the relevant art may not be discussed in detail, but are intended to be part of the specification where appropriate.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, further discussion thereof is not required in subsequent figures.
Embodiments of the invention are operational with numerous other general purpose or special purpose computing system environments or configurations, and with numerous other electronic devices, such as terminal devices, computer systems, servers, etc. Examples of well known terminal devices, computing systems, environments, and/or configurations that may be suitable for use with electronic devices, such as terminal devices, computer systems, servers, and the like, include, but are not limited to: personal computer systems, server computer systems, thin clients, thick clients, hand-held or laptop devices, microprocessor-based systems, set-top boxes, programmable consumer electronics, networked personal computers, minicomputer systems, mainframe computer systems, distributed cloud computing environments that include any of the above, and the like.
Electronic devices such as terminal devices, computer systems, servers, etc. may be described in the general context of computer system-executable instructions, such as program modules, being executed by a computer system. Generally, program modules may include routines, programs, objects, components, logic, data structures, etc. that perform particular tasks or implement particular abstract data types. The computer system/server may be practiced in distributed cloud computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed cloud computing environment, program modules may be located in both local and remote computer system storage media including memory storage devices.
Exemplary method
Fig. 1 is a flowchart illustrating an information pushing method according to an exemplary embodiment of the present invention. The embodiment can be applied to an electronic device, as shown in fig. 1, and includes the following steps:
and step S1, extracting the user characteristics of the target user.
The user characteristics can represent behavior characteristics and preference characteristics of the target user, namely characteristics capable of reflecting individuation of the target user. In this embodiment, the user features have num1 dimensions, each dimension corresponds to a feature, and features of num1 dimensions can be represented in the form of a first feature vector. The feature of num1 dimensions is divided into five major categories of marketing attribute, basic attribute, behavior attribute, transaction attribute and enterprise attribute. Taking a certain website or application as an example of a main body for collecting information of a target user, the marketing attribute is a registration channel name of the target user, that is, the target user registers the website or application through friend push, registers the website or application through preferential activities, and the like. The basic attributes are one or more of occupation, age, gender, region, city of the target user. The behavior attribute is extracted from behaviors of the target user in searching, browsing, monitoring, Push clicking, collecting, commenting and the like of the website or the application. The transaction attribute is an attribute extracted by acquiring transaction information of the target user on the website or the application, for example, VIP related data, consumption institute invoicing related data (including units, consumption dates, invoicing times and the like), VIP order payment data, enterprise service orders and behaviors of the user are acquired, and further, the transaction attribute of the target user is acquired. The enterprise attribute means that the target user is an enterprise user and has information such as the industry of the enterprise and tax payment behaviors of the enterprise, and the enterprise attribute of the target user is determined.
Further, acquiring basic data of the target user; and mapping the corresponding user characteristics into numerical values based on the basic data of the target user to form the basic characteristics of the target user. The basic data comprises behavior data and registration data of the target user.
Further, obtaining a user-defined user characteristic aiming at the target user; and mapping the user-defined user characteristics to corresponding numerical values based on the basic characteristics of the target user. The user-defined characteristics are the values of the user-defined indexes solved based on the basic data, the user-defined indexes and the formulas corresponding to the indexes of the target user, and the user-defined characteristics are the user-defined characteristics of the target user.
In this embodiment, taking the website or application setting buried point event as an example, the buried point data collected by the website or application is used as the obtained user data to form the original buried point data wide table of the target user. Since the data acquired in the buried point data is not regular and the data types are various, it is necessary to structure irregular data in the buried point data. In this embodiment, the buried point data may be used as basic data, a plurality of indexes, that is, user-defined indexes, may be defined based on the buried point data, and a value of the user-defined index obtained by calculation according to a numerical value corresponding to the buried point data is a user-defined characteristic of the target user.
Taking the first active date in the collected buried point data as an example, the data type corresponding to the first active date is date type data, and in order to facilitate subsequent use requirements, for example, calculation possibly related to a user-defined index converts a numerical value of a specific date type corresponding to the first active date into a number of days away from the current time, that is, converts the date type data into numerical data.
Taking the distribution of the active time of the target user in the collected buried point data as an example, the numerical form is as follows:
{"00":0,"01":0,"02":0,"03":0,"04":0,"05":0,"06":3,"07":0,"08":0,"09":2,"10":0,"11":3,"12":0,"13":0,"14":0,"15":1,"16":0,"17":0,"18":0,"19":0,"20":0,"21":0,"22":0,"23":0}。
in order to facilitate subsequent use, the numerical value corresponding to the active time distribution of the target user is converted into numerical data, and the numerical data is converted into the integral time with the maximum active proportion. Taking the value corresponding to the active time distribution of the target user as an example, if the target user is active at 09, 11, 15, and the most active time point is 11 points, the value corresponding to the active time distribution of the target user is converted into 11, that is, the target user is most active at 11, and the type of the complex data is converted into numerical data.
In addition, for example, the category-type characteristics such as the city where the user is located, the favorite tag for reading, and the query function preference in the collected data of the target user are taken as examples, because the computer can only recognize numbers, a part of the category-type user characteristics need to be analyzed and encoded. If the user characteristic is 'three-check preference', the three-check comprises checking companies, checking boss and checking incidence relation, and analyzing and coding the characteristic value, wherein the coding comprises the following steps: { "company checking": 1, "boss for looking up": 2, "find association": 3, if the collected data of the user is determined to be characterized as 'company searching', setting the characteristic value of the three-search preference of the user as a numerical value 1.
Step S2: and extracting message characteristics of the target information to be pushed.
Taking the target information to be pushed as news and articles as examples. In this embodiment, the message features of the target information to be pushed have num2 dimensions, each dimension corresponds to one feature, and features of num2 dimensions can be represented in the form of a second feature vector. The message features of num2 dimensions are classified into basic features, interactive features, content features, category features, emotional features and the like.
The target information to be pushed generally comprises two parts, namely a title and a content. Correspondingly, the characteristics of the target information to be pushed comprise basic characteristics of article source media, article sending time, article browsing, clicking, collecting and the like. The interactive characteristics refer to interactive behaviors between the target information to be pushed and a user, for example, browsing, clicking, collecting, and the like of the target information to be pushed. The content features are used for representing content main points of the target information to be pushed, for example, the number of title words of the target information to be pushed, the number of times of each part of speech in a title, the number of main point words of the target information to be pushed, and the like. The category feature refers to a category to which the target information to be pushed belongs, for example, a feature of the target information to be pushed is extracted through a natural language processing algorithm, and the target information to be pushed is mapped to a specific category through a classification algorithm, where the category includes, but is not limited to: 1. international 2, sports 3, entertainment 4, society 5, finance 6, current affairs 7, science 8, emotion 9, automobile 10, education 11, fashion 12, games 13, military 14, tourism 15, cate 16, culture 17, health preserving 18, fun 19, home 20, cartoon 21, pet 22, mother and infant 23, constellation and fortune 24, history 25, music 26, and synthesis. The emotional characteristics are characteristics representing the emotion of the target information to be pushed, for example, the emotional information of the text content corresponding to the target information to be pushed is analyzed through the number of words of various parts of speech and entities, the number of words of a title of an article, and the number of words of a push content in the target information to be pushed, which are identified by a jieba toolkit.
Step S3, determining whether to push the target information to the target user as preferred information based on the processing result of a double-layer network model to the user characteristic and the message characteristic under the condition that the push mode of the target information is determined to be a model push mode based on a matching strategy, wherein a first layer of the double-layer network model is used for receiving input characteristic information and selecting a corresponding characteristic subset from the received input characteristic information, and a second layer of the double-layer network model is used for learning the weight of each characteristic in the characteristic subset selected by the first layer; and
and under the condition that the pushing mode of the target information is determined to be a matching rule pushing mode based on a matching strategy, comparing the user characteristics with the message characteristics based on a preset rule to determine whether to push the target information to the target user as preferred information.
In this embodiment, when the correlation between the message characteristic of the target information and the user characteristic of the target user is higher than or equal to a preset threshold, determining that the push mode of the target information is a regular push mode; and under the condition that the correlation degree of the message characteristics of the target information and the user characteristics of the target user is lower than the preset threshold value, determining that the pushing mode of the target information is a model pushing mode.
The inputs to the two-tier network model are the user characteristics and the message characteristics. The first layer of the two-layer network model is used for receiving input features and selecting a feature subset from the input features; the second layer of the two-layer network model is used for learning the weight of each feature in the selected feature subset. The structure of the double-layer network model is the same as that of a conventional neural network model, for example, a neural network model such as CNN, RNN and the like is used. In this embodiment, the dual-layer network model is trained in a manner of combining lightgbm with lr. The first layer of the double-layer network model is used for receiving input features, and is trained by using lightgbm, so that partial features are screened from the input features to form a feature subset, and the effect of reducing dimensions is achieved. And training a second layer of the double-layer network model by using lr, wherein the second layer of the double-layer network model is used for learning the weight of each feature in the selected feature subset, and the weight of each feature in the feature subset can be adjusted. And adjusting the weight of the features in the feature subset to further influence the subsequent feature matching effect. And the output of the double-layer network model is the matching score of the information to be pushed and the user.
Further, the double-layer network model is obtained by splicing the user characteristics and the message characteristics as the input of the double-layer network model, and obtaining the matching score between the target information to be pushed and the target user.
Further, the double-layer neural network model sets a second threshold, and if the matching score between the characteristics of the target information to be pushed and the user characteristics of the target user is greater than or equal to the second threshold, the target information is used as preferred information and pushed to the target user. In this embodiment, the second threshold is set to 0.7 to ensure a matching effect, and if a matching score between the feature of the target information to be pushed and the user feature of the target user is greater than or equal to 0.7, the target information to be pushed is pushed to the target user as the preferred push information.
Further, the embodiment supports synchronous matching of the features of a plurality of pieces of target information to be pushed and the user features of the target user, and constructs a spliced vector of the user features and the features of the information to be pushed for each piece of target information to be pushed; and combining the spliced vectors of all the information to be pushed into a matrix, and inputting the matrix into the double-layer network model to realize synchronous matching of the characteristics of a plurality of pieces of target information to be pushed and the user characteristics of the target user.
The neural network model has the potential defect that the gradient is slowly reduced to cause the local optimization, so that the double-layer network model obtained by training has the potential defect that the matching precision is slightly insufficient.
Therefore, in this embodiment, a matching policy is set to determine a push mode of the target information to be pushed, and different push modes match the user characteristics with the characteristics of the target information to be pushed in different manners.
The matching strategy is that if the correlation between the message characteristics of the target information to be pushed and the user characteristics is higher than or equal to a preset threshold value, pushing is carried out based on a matching rule; otherwise, pushing based on the double-layer network model. In this embodiment, the correlation is determined based on a conventional manner such as an euclidean distance, and the preset threshold is preset based on experience.
And the pushing based on the matching rule is to set the matching rule, the matching rule selects the subset of the user characteristics and the subset of the message characteristics of the target information to be pushed in advance, and if the matched characteristics exist in the subset of the user characteristics and the subset of the message characteristics of the target information to be pushed, the target information to be pushed is used as preferred pushing information and is used for pushing the preferred pushing information to the target user.
For example, the matching rule is set as (entity-related; region; label; industry), that is, the features corresponding to the entity-related, region, label, and industry are selected from the subset of the user features and the subset of the message features of the target information to be pushed. The entity correlation is used for indicating that the target information to be pushed corresponds to the user entity, the region is used for indicating that the region related to the target information to be pushed corresponds to the user, the label indicates that the target information to be pushed corresponds to the user label, and the industry indicates that the target information to be pushed is consistent with the user industry. If the characteristics of the target information to be pushed have 15 characteristics, the user characteristics have 12 characteristics, both have entity-related, regional, tag, and industry characteristics, and the characteristic values match, all characteristics in the subset of the user characteristics and the subset of the message characteristics of the target information to be pushed are matched with each other, and the target information to be pushed is used as preferred push information to be pushed to the target user.
Further, before the step S1, extracting the user characteristics of the target user, the method includes:
and acquiring the target information to be pushed, and preprocessing the target information to be pushed.
In this embodiment, for example, news and news flashes are crawled from a collaborative media website as target information to be pushed. The preprocessing of the target information to be pushed comprises the following steps: and analyzing and cleaning the target information to be pushed, and removing redundant html labels, messy codes caused by transcoding, special symbols caused by the messy codes, invalid links, filtering sensitive words and the like in the target information to be pushed. The invalid link cannot be opened due to incompleteness, for example, the sensitive words to be filtered can compare the target information to be pushed with the sensitive words in the sensitive word blacklist in a mode of setting a sensitive word blacklist, and if the same word group exists, the information to be pushed is filtered, so that filtering of information such as bad vulgars is achieved.
Further, the preprocessed target information to be pushed is stored in the content pool. The content pool is a database for storing target information to be pushed and/or preprocessed target information to be pushed.
Further, after the step S3, a step S4 is also included.
In step S4, the preferred push information for pushing is determined, and before the preferred push information is pushed to the target user, the preferred push information is subjected to deduplication processing.
In this embodiment, the preferred push information is deduplicated to ensure that the user receives the non-repeated preferred push information. Multiple repetitions of the same preferred push information, or multiple preferred push information with highly similar content, are not received.
Further, an index is calculated for each of the preferred push information, and an index value similar _ key is assigned to each of the preferred push information. And adding an index column in the data table corresponding to the content pool for recording the similar _ key. Determining the identity or similarity of the preferred push information based on the title and/or text similarity of the preferred push information. The same or similar preferred push information is assigned with a same index value, similar _ key, and the preferred push information is deduplicated by the index value, similar _ key.
The way of calculating the index value is: and dividing sentences in the preferable push information into long sentences and short sentences, wherein the standard for dividing the long sentences and the short sentences is a general dividing standard in the field. And performing multi-round segmentation on the short sentence based on the symbol and word number threshold values to ensure that the length of the segmented clause is within a preset range. In this embodiment, the word count threshold is, for example, 20, and the symbols include punctuation marks such as a pause, a comma, a semicolon, and a period. And comparing the cut clauses with the long clauses in the information to be pushed, and returning the larger numerical values in the format tower and the editing distance. And weighting based on the returned data and the proportion of the cut clauses in the length of the short clause, and comparing the similarity of the weighted value and the weighted values of other pushed information in the content pool. If the similarity is greater than or equal to the threshold, the articles are considered to be similar, and the same similar _ key is set.
Yet another way to compute the index value is: and dividing sentences in the preferable push information into long sentences and short sentences, wherein the standard for dividing the long sentences and the short sentences is a general dividing standard in the field. And performing multi-round segmentation on the short sentence based on the symbol and word number threshold values to ensure that the length of the segmented clause is within a preset range. In this embodiment, the word count threshold is, for example, 20, and the symbols include punctuation marks such as a pause, a comma, a semicolon, and a period. And comparing the cut clauses with the long clauses in the information to be pushed, and returning the larger numerical values in the format tower and the editing distance. And weighting based on the returned data and the proportion of the clauses after cutting to the length of the short sentence, and taking the obtained numerical value as the index value of the preferred push information. And taking the preferred push information with the same index value as the same preferred push information, and taking the preferred push information with the index value deviation degree smaller than the threshold value as the similar preferred push information. And pushing the preferred push information with the characteristics of the preferred push information which are matched with the characteristics of the user with the highest matching scores to the user in a group of similar preferred push information.
Fig. 2 is a diagram illustrating a technical effect of an information pushing method according to an exemplary embodiment of the present invention.
Exemplary devices
Fig. 3 is a schematic structural diagram of an information push apparatus according to an exemplary embodiment of the present invention. As shown in fig. 3, the present embodiment includes:
the user characteristic extraction module: configured to extract user features of a target user;
the message feature extraction module: the method comprises the steps of configuring to extract message characteristics of target information to be pushed;
a mode determination module: the target information processing method comprises the steps that whether target information is pushed to a target user or not is determined based on a processing result of a double-layer network model on user characteristics and message characteristics under the condition that a matching strategy determines that the pushing mode of the target information is a model pushing mode, wherein a first layer of the double-layer network model is used for receiving input characteristic information and selecting a corresponding characteristic subset from the received input characteristic information, and a second layer of the double-layer network model is used for learning the weight of each characteristic in the characteristic subset selected by the first layer; and
and comparing the user characteristics with the message characteristics based on a preset rule to determine whether to push the target information to the target user as preferred information or not under the condition that the push mode of the target information is determined to be a rule push mode based on a matching strategy.
Exemplary electronic device
Fig. 4 is a structure of an electronic device according to an exemplary embodiment of the present invention. The electronic device may be either or both of the first device and the second device, or a stand-alone device separate from them, which stand-alone device may communicate with the first device and the second device to receive the acquired input signals therefrom. FIG. 4 illustrates a block diagram of an electronic device in accordance with an embodiment of the disclosure. As shown in fig. 4, the electronic device includes one or more processors 41 and memory 42.
The processor 41 may be a Central Processing Unit (CPU) or other form of processing unit having data processing capabilities and/or instruction execution capabilities, and may control other components in the electronic device to perform desired functions.
The input device 43 may also include, for example, a keyboard, a mouse, and the like.
The output device 44 can output various kinds of information to the outside. The output devices 34 may include, for example, a display, speakers, a printer, and a communication network and its connected remote output devices, among others.
Of course, for simplicity, only some of the components of the electronic device relevant to the present disclosure are shown in fig. 4, omitting components such as buses, input/output interfaces, and the like. In addition, the electronic device may include any other suitable components, depending on the particular application.
Exemplary computer program product and computer-readable storage Medium
In addition to the above-described methods and apparatus, embodiments of the present disclosure may also be a computer program product comprising computer program instructions that, when executed by a processor, cause the processor to perform the steps in the information push method according to various embodiments of the present disclosure described in the "exemplary methods" section above of this specification.
The computer program product may write program code for carrying out operations for embodiments of the present disclosure in any combination of one or more programming languages, including an object oriented programming language such as Java, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server.
Furthermore, embodiments of the present disclosure may also be a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, cause the processor to perform steps in an information push method according to various embodiments of the present disclosure described in the "exemplary methods" section above of this specification.
The computer-readable storage medium may take any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may include, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
The foregoing describes the general principles of the present disclosure in conjunction with specific embodiments, however, it is noted that the advantages, effects, etc. mentioned in the present disclosure are merely examples and are not limiting, and they should not be considered essential to the various embodiments of the present disclosure. Furthermore, the foregoing disclosure of specific details is for the purpose of illustration and description and is not intended to be limiting, since the disclosure will be described in detail with reference to specific details.
In the present specification, the embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same or similar parts in the embodiments are referred to each other. For the system embodiment, since it basically corresponds to the method embodiment, the description is relatively simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The block diagrams of devices, apparatuses, systems referred to in this disclosure are only given as illustrative examples and are not intended to require or imply that the connections, arrangements, configurations, etc. must be made in the manner shown in the block diagrams. These devices, apparatuses, devices, systems may be connected, arranged, configured in any manner, as will be appreciated by those skilled in the art. Words such as "including," "comprising," "having," and the like are open-ended words that mean "including, but not limited to," and are used interchangeably therewith. The words "or" and "as used herein mean, and are used interchangeably with, the word" and/or, "unless the context clearly dictates otherwise. The word "such as" is used herein to mean, and is used interchangeably with, the phrase "such as but not limited to".
The methods and apparatus of the present disclosure may be implemented in a number of ways. For example, the methods and apparatus of the present disclosure may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above-described order for the steps of the method is for illustration only, and the steps of the method of the present disclosure are not limited to the order specifically described above unless specifically stated otherwise. Further, in some embodiments, the present disclosure may also be embodied as programs recorded in a recording medium, the programs including machine-readable instructions for implementing the methods according to the present disclosure. Thus, the present disclosure also covers a recording medium storing a program for executing the method according to the present disclosure.
It is also noted that in the devices, apparatuses, and methods of the present disclosure, each component or step can be decomposed and/or recombined. Such decomposition and/or recombination should be considered as equivalents of the present disclosure. The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the disclosure. Thus, the present disclosure is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
The foregoing description has been presented for purposes of illustration and description. Furthermore, the description is not intended to limit embodiments of the disclosure to the form disclosed herein. While a number of example aspects and embodiments have been discussed above, those of skill in the art will recognize certain variations, modifications, alterations, additions and sub-combinations thereof.
Claims (9)
1. An information pushing method, comprising:
extracting user characteristics of a target user;
extracting message characteristics of target information to be pushed;
determining whether to push the target information to the target user as preferred information or not based on a processing result of a double-layer network model on the user characteristic and the message characteristic under the condition that a push mode of the target information is determined to be a model push mode based on a matching strategy, wherein a first layer of the double-layer network model is used for receiving input characteristic information and selecting a corresponding characteristic subset from the received input characteristic information, and a second layer of the double-layer network model is used for learning the weight of each characteristic in the first layer selected characteristic subset; and
and under the condition that the pushing mode of the target information is determined to be the matching rule pushing mode based on the matching strategy, comparing the user characteristic with the message characteristic based on a preset rule to determine whether to push the target information to the target user as preferred information.
2. The method of claim 1, further comprising:
acquiring basic data of the target user; and
and mapping the corresponding user characteristics into numerical values based on the basic data of the target user to form the basic characteristics of the target user.
3. The method of claim 2, further comprising:
obtaining user-defined user characteristics aiming at the target user;
and mapping the user-defined user characteristics to corresponding numerical values based on the basic characteristics of the target user.
4. The method of claim 1, wherein:
determining that the push mode of the target information is a regular push mode under the condition that the correlation degree of the message characteristics of the target information and the user characteristics of the target user is higher than or equal to a preset threshold value;
and under the condition that the correlation degree of the message characteristics of the target information and the user characteristics of the target user is lower than the preset threshold value, determining that the pushing mode of the target information is a model pushing mode.
5. The method according to claim 1, wherein the pushing based on the matching rule is to set a matching rule, the matching rule pre-selects the subset of the user features and the subset of the message features of the target information to be pushed, and if there are features matching with each other in the subset of the user features and the subset of the message features of the target information to be pushed, the target information to be pushed is used as preferred pushing information to be pushed to the target user.
6. The method of claim 1, further comprising:
before the preferred information is pushed to the target user, the preferred pushed information is subjected to deduplication processing.
7. An information pushing apparatus, comprising:
the user characteristic extraction module: configured to extract user features of a target user;
the message characteristic extraction module: the method comprises the steps of configuring to extract message characteristics of target information to be pushed;
a mode determination module: determining whether to push the target information to the target user as preferred information or not based on a processing result of a double-layer network model to the user characteristic and the message characteristic under the condition that a push mode of the target information is determined to be a model push mode based on a matching strategy, wherein a first layer of the double-layer network model is used for receiving input characteristic information and selecting a corresponding characteristic subset from the received input characteristic information, and a second layer of the double-layer network model is used for learning the weight of each characteristic in the characteristic subset selected by the first layer; and
and comparing the user characteristics with the message characteristics based on a preset rule to determine whether to push the target information to the target user as preferred information or not under the condition that the push mode of the target information is determined to be a rule push mode based on a matching strategy.
8. A computer-readable storage medium having stored therein a plurality of instructions; the plurality of instructions for being loaded by a processor and for performing the method of any one of claims 1-6.
9. An electronic device, characterized in that the electronic device comprises:
a processor for executing a plurality of instructions;
a memory to store a plurality of instructions;
wherein the plurality of instructions are to be stored by the memory and to be loaded and executed by the processor to perform the method of any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210538857.3A CN114996571A (en) | 2022-05-17 | 2022-05-17 | Information pushing method and device, storage medium and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210538857.3A CN114996571A (en) | 2022-05-17 | 2022-05-17 | Information pushing method and device, storage medium and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114996571A true CN114996571A (en) | 2022-09-02 |
Family
ID=83026394
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210538857.3A Pending CN114996571A (en) | 2022-05-17 | 2022-05-17 | Information pushing method and device, storage medium and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114996571A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116405453A (en) * | 2023-04-23 | 2023-07-07 | 中航信移动科技有限公司 | Information pushing method based on multiple features, storage medium and electronic equipment |
-
2022
- 2022-05-17 CN CN202210538857.3A patent/CN114996571A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116405453A (en) * | 2023-04-23 | 2023-07-07 | 中航信移动科技有限公司 | Information pushing method based on multiple features, storage medium and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chua et al. | Automatic summarization of events from social media | |
| US8676730B2 (en) | Sentiment classifiers based on feature extraction | |
| US10565519B2 (en) | Systems and method for performing contextual classification using supervised and unsupervised training | |
| Venugopalan et al. | Exploring sentiment analysis on twitter data | |
| US8972408B1 (en) | Methods, systems, and articles of manufacture for addressing popular topics in a social sphere | |
| US11727211B2 (en) | Systems and methods for colearning custom syntactic expression types for suggesting next best correspondence in a communication environment | |
| Alharbi et al. | Evaluation of sentiment analysis via word embedding and RNN variants for Amazon online reviews | |
| Kaur | Incorporating sentimental analysis into development of a hybrid classification model: A comprehensive study | |
| CN107832338B (en) | Method and system for recognizing core product words | |
| Yun et al. | Computationally analyzing social media text for topics: A primer for advertising researchers | |
| Yang et al. | A topic model for co-occurring normal documents and short texts | |
| CN114238573A (en) | Information pushing method and device based on text countermeasure sample | |
| Sasmita et al. | Unsupervised aspect-based sentiment analysis on Indonesian restaurant reviews | |
| Rani et al. | Study and comparision of vectorization techniques used in text classification | |
| Tang et al. | Evaluation of Chinese sentiment analysis APIs based on online reviews | |
| WO2023129339A1 (en) | Extracting and classifying entities from digital content items | |
| CN112115712A (en) | Topic-based group emotion analysis method | |
| Al-Otaibi et al. | Finding influential users in social networking using sentiment analysis | |
| US10614100B2 (en) | Semantic merge of arguments | |
| CN114996571A (en) | Information pushing method and device, storage medium and electronic equipment | |
| CN111737607B (en) | Data processing method, device, electronic equipment and storage medium | |
| CN107291686B (en) | Method and system for identifying emotion identification | |
| CN117113973A (en) | Information processing method and related device | |
| Lo et al. | An emperical study on application of big data analytics to automate service desk business process | |
| Clarizia et al. | Sentiment analysis in social networks: A methodology based on the latent dirichlet allocation approach |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |