CN115080840A

CN115080840A - Content pushing method and device and storage medium

Info

Publication number: CN115080840A
Application number: CN202110281298.8A
Authority: CN
Inventors: 王韵陶; 陈炳文
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2021-03-16
Filing date: 2021-03-16
Publication date: 2022-09-20

Abstract

The embodiment of the application discloses a content pushing method, wherein a processing device can acquire content behavior data of a target user corresponding to target content in a target time interval, and respectively counts a plurality of statistical characteristics of the target user related to the target content based on different statistical periods, the plurality of statistical characteristics form a multi-period statistical characteristic of the target user corresponding to the target content, and the multi-period statistical characteristic can embody the change situation of the intention of the target user to the target content along with time, so that the interest of the target user can be more accurately expressed. Finally, by utilizing technologies such as machine learning in the artificial intelligence technology, the association parameters of the target user and the target content can be determined according to the multi-cycle statistical characteristics, and whether the target content is pushed to the target user is determined based on the association parameters, so that the processing equipment can push the target content to a user with a high intention on the target content in the near future, and the receiving probability of the user on the target content and the content recommendation efficiency are improved.

Description

Content pushing method and device and storage medium

Technical Field

The present application relates to the field of data processing, and in particular, to a content push method, device and storage medium.

Background

The content platform can push the contents which the user may like to the user by identifying and mining the user, so that the purposes of expanding the number of users in the platform, increasing the user viscosity of the users in the platform and the like are achieved.

Currently, mining of users is mainly based on user profiles of users, and user profiles by identifying characteristics such as user preferences can be used as a basis for matching contents that users may like.

However, the applicability of the user portrait in some content recommendation scenes is not good, and the accuracy of the user portrait is difficult to guarantee, which may result in determining wrong content when the user portrait is not correct, and affect the content recommendation effect.

Disclosure of Invention

In order to solve the foregoing technical problem, an embodiment of the present application provides a content pushing method, which introduces an association between a user and content as a criterion for determining whether to push the content or not, and considers a change trend of a user intention on the one hand, thereby facilitating improvement of accuracy of content pushing.

The embodiment of the application discloses the following technical scheme:

in a first aspect, an embodiment of the present application provides a content pushing method, where the method includes:

acquiring content behavior data of a target user corresponding to target content in a target time interval;

performing data statistics processing based on different statistics periods on the content behavior data to obtain multi-period statistics characteristics related to the target content of the target user, wherein the multi-period statistics characteristics comprise a first statistics characteristic corresponding to a first statistics period and a second statistics characteristic corresponding to a second statistics period, the first statistics period and the second statistics period have different period lengths, and the period lengths of the first statistics period and the second statistics period are both smaller than or equal to the duration identified by the target time interval;

determining the association parameters of the target user and the target content according to the multi-period statistical characteristics;

determining whether to push the target content to the target user based on the association parameter.

In a second aspect, an embodiment of the present application provides a content pushing apparatus, where the apparatus includes an obtaining unit, a first determining unit, a second determining unit, and a pushing unit:

the acquisition unit is used for acquiring content behavior data of a target user corresponding to target content in a target time interval;

the first determining unit is configured to perform data statistics processing on the content behavior data based on different statistics periods to obtain multi-period statistics characteristics of the target user related to the target content, where the multi-period statistics characteristics include a first statistics characteristic corresponding to a first statistics period and a second statistics characteristic corresponding to a second statistics period, the first statistics period and the second statistics period have different period lengths, and the period lengths of the first statistics period and the second statistics period are both less than or equal to a duration identified by the target time interval;

the second determining unit is used for determining the association parameters of the target user and the target content according to the multi-cycle statistical characteristics;

the pushing unit is used for determining whether to push the target content to the target user based on the association parameters.

In a third aspect, an embodiment of the present application provides a computer device, where the device includes a processor and a memory:

the memory is used for storing program codes and transmitting the program codes to the processor;

the processor is configured to perform the content push method of the first aspect according to instructions in the program code.

In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium for storing a computer program, where the computer program is configured to execute the content push method described in the first aspect.

According to the technical scheme, in the content recommendation scene, the content behavior data of the target user corresponding to the target content in the target time interval is obtained, the content behavior data can identify the behavior of the target user related to the target content in the target time interval, and the association between the target user and the target content can be embodied through the content behavior data. And determining multi-period statistical characteristics from the content behavior data according to different statistical periods, wherein the multi-period statistical characteristics at least comprise a first statistical characteristic determined based on the first statistical period and a second statistical characteristic determined based on the second statistical period. Because the period lengths of the first statistical period and the second statistical period are different, the change of the association between the target user and the target content in a time sequence angle is reflected through the change of content behavior data in different statistical periods, namely, the multi-period statistical characteristics can express the time sequence change trend of the target user to the intention of the target content. The multicycle statistical characteristics with rich expression ability play a role in more accurately depicting the preference of the target user, so that when the relevance parameters determined by the multicycle statistical characteristics are used for measuring whether the target content is pushed to the target user, if the target content is determined to be pushed, the probability that the target user receives the target content is higher, and the content recommendation efficiency is improved.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.

Fig. 1 is a schematic diagram of a content push method in an actual application scenario provided in an embodiment of the present application;

fig. 2 is a flowchart of a content pushing method according to an embodiment of the present application;

fig. 3 is a schematic diagram of a content push method in an actual application scenario according to an embodiment of the present application;

fig. 4 is a block diagram illustrating a content pushing apparatus according to an embodiment of the present disclosure;

fig. 5 is a block diagram of a computer device according to an embodiment of the present disclosure;

fig. 6 is a block diagram of a server according to an embodiment of the present disclosure;

fig. 7 is a flowchart of a content push method in an actual application scenario according to an embodiment of the present application.

Detailed Description

Embodiments of the present application are described below with reference to the accompanying drawings.

Pushing various related contents to a user is one of common means for product promotion, and whether reasonable content pushing can be performed can directly affect the promotion effect of the product. In the related art, in order to recommend a content with a high acceptance level to a user, a user figure corresponding to the user is generally generated based on information of the user, such as height, sex, age, place of birth, occupation, and the like, and a pushed content that may be of interest to the user is identified from the user figure. However, the judgment basis of the method is too single, so that the method is difficult to meet the requirements of rich and diverse content push, and meanwhile, because the user information is relatively fixed, the change situation of the interests and hobbies of the user along with the time is difficult to embody, so that the relatively accurate content push cannot be realized.

In order to solve the foregoing technical problem, an embodiment of the present application provides a content push method, where a processing device may determine a multi-period statistical characteristic of a target content corresponding to a target user by counting content behavior data in different time periods, where the multi-period statistical characteristic can reflect a change situation of an association relationship between the target user and the target content along with time, so that on one hand, association between the user and the content is introduced as a criterion for determining whether to push the content, on the other hand, a change trend of a user intention is considered, and thus, the content push accuracy is improved.

It is understood that the method may be applied to a processing device, which is a processing device with a content push function, for example, a terminal device or a server with a content push function. The method is independently executed by the terminal equipment or the server, can also be applied to a network scene of communication between the terminal equipment and the server, and is operated by the cooperation of the terminal equipment and the server. The terminal device may be a mobile phone, a desktop computer, a Personal Digital Assistant (PDA for short), a tablet computer, or the like. The server may be understood as an application server, or may also be a Web server, and in actual deployment, the server may be an independent physical server, or may be a server cluster or a distributed system formed by multiple physical servers. The terminal and the server may be directly or indirectly connected through wired or wireless communication, and the application is not limited herein.

In addition, the present application also relates to Artificial Intelligence (AI). Artificial intelligence is a theory, method, technique and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use the knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technique of computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can react in a manner similar to human intelligence. Artificial intelligence is the research of the design principle and the realization method of various intelligent machines, so that the machines have the functions of perception, reasoning and decision making.

The artificial intelligence technology is a comprehensive subject and relates to the field of extensive technology, namely the technology of a hardware level and the technology of a software level. The artificial intelligence infrastructure generally includes technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and the like. The technical scheme mainly relates to a machine learning technology.

Machine Learning (ML) is a multi-domain cross discipline, and relates to a plurality of disciplines such as probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory and the like. The special research on how a computer simulates or realizes the learning behavior of human beings so as to acquire new knowledge or skills and reorganize the existing knowledge structure to continuously improve the performance of the computer. Machine learning is the core of artificial intelligence, is the fundamental approach for computers to have intelligence, and is applied to all fields of artificial intelligence. Machine learning and deep learning generally include techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and formal education learning. For example, the embodiment of the application can train the classifier model by using a machine learning technology, so that the classifier model can more accurately determine the associated parameters.

In order to facilitate understanding of the technical solution of the present application, a content push method provided in the embodiments of the present application will be described below with reference to an actual application scenario.

Referring to fig. 1, fig. 1 is a schematic diagram of a content push method in an actual application scenario provided by an embodiment of the present application, in the actual application scenario, a processing device is a server 101 having a content push function, content behavior data between a user and content is stored in the server 101, where the content behavior data is data capable of representing an association between the user and the content, for example, when the content is an advertisement of a certain product, the content behavior data may be the number of the advertisement pushed to the user, and the data may represent the intention of the user to the advertisement.

Before pushing a target content to a target user, the server 101 may first obtain content behavior data of the target user corresponding to the target content in a target time interval, for example, the number of target content pushed to the target user in 1 year may be used. The server 101 may perform data statistics processing on the content behavior data based on different statistics periods to obtain a multi-period statistical characteristic of the target user related to the target content, where, to reflect a change of an association relationship between the target user and the target content over time, the server 101 may perform statistics on the content behavior data in a plurality of different statistics periods, for example, a first statistical characteristic may be obtained by statistics in a first statistics period, and a second statistical characteristic may be obtained by statistics in a second statistics period, where the first statistics period and the second statistics period have different period lengths. In this practical application scenario, when the target time interval is 1 year, the first statistical period may be 1 month, and the second statistical period may be 3 months, as long as the target time interval is less than or equal to the target time.

Therefore, through the first statistical feature and the second statistical feature in the multi-period statistical feature, the server 101 can obtain the change condition of the association relationship between the target user and the target content between about 3 months and about 1 month, and further through the multi-period statistical feature, the server 101 can accurately depict the preference of the target user. The server 101 may determine, according to the multi-period statistical characteristic, a correlation parameter between the target user and the target content, where the correlation parameter can reflect, to a certain extent, the acceptance of the target user for the target content at the current time. For example, if the content behavior data is the push quantity, and the push quantity in the first statistical characteristic is higher in the push quantity corresponding to the second statistical characteristic, it indicates that the push quantity in the first two months in the last 3 months is smaller than the push quantity in the last 1 month, that is, the push quantity is increasing continuously, so that it can be indicated that the target user may have a high current acceptance degree of the target content, and the determined association parameter may be high to some extent.

The server 101 may determine whether to push the target content to the target user based on the association parameter, for example, if the association parameter is higher than a certain threshold, push the target content to the target user; if the target content is not higher than the preset target content, pushing is not performed, so that the server 101 can push the target content to a user with higher acceptance based on the angle of the target content, and the pushing efficiency and the effectiveness are improved; based on the perspective of the target user, the server 101 may push the content with a higher acceptance to the target user, so as to improve the pushed experience of the target user, and finally improve the rationality of the whole content pushing process.

Next, a content push method provided in an embodiment of the present application will be described with reference to the accompanying drawings.

Referring to fig. 2, fig. 2 is a flowchart of a content pushing method provided in an embodiment of the present application, where the method includes:

s201: and acquiring content behavior data of the target user corresponding to the target content in the target time interval.

The content behavior data is data capable of representing the related behaviors of the target user and the target content in the target time interval, and the association relationship between the target user and the target content can be represented through the content behavior data. The target time interval can be a time interval with any length, the target user can be any user, and the target content can be content related to any product. For example, the content behavior data may be the number of times an advertisement for a product is pushed to a user within a target time interval.

S202: and performing data statistics processing on the content behavior data based on different statistical periods to obtain multi-period statistical characteristics related to the target content and the target user.

In order to push contents more accurately and effectively, the processing device needs to judge the user's taste accurately to push the interested and intended contents to the user. It will be appreciated that the degree of user intent on the content is often not fixed, but may vary over the time series. For example, a user may need to purchase product a three months ago, so the intention of the advertisement related to product a is high, and the number of advertisements pushed by the advertisement is large; in the last three months, since the user has purchased the product a, the intention of the advertisement related to the product a is reduced, and at this time, if the advertisement of the product is pushed to the user, the acceptance of the advertisement by the user may be low.

Meanwhile, the content behavior data corresponding to the content of the product generally has the characteristic of high sparseness, and the interaction time between the target user and the target content may be separated by several months, so that if the content is directly pushed according to the content behavior data in the target time interval, the processing device may not accurately judge the intention degree of the target user for the target content.

Based on this, after the content behavior data of the target content corresponding to the target time interval is acquired, in order to more accurately depict the preference of the target user for the target content, the processing device may analyze the association relationship between the target user and the target content based on the time series, so as to determine the change condition of the intention of the target user for the target content through the association relationship changing along with the time series. In this embodiment, the processing device may first set a first statistical period and a second statistical period, where the first statistical period and the second statistical period have different period lengths, and in order to improve the effectiveness of statistics, the period lengths of the first statistical period and the second statistical period are both less than or equal to the duration identified by the target time interval.

The processing device may perform data statistics on the content behavior data in combination with the plurality of statistical cycles to obtain a multi-cycle statistical characteristic of the target user related to the target content, where the data statistics refers to performing statistics on the data according to a certain statistical cycle, and the multi-cycle statistical characteristic is used to identify a characteristic of the content behavior data in the plurality of statistical cycles. In an embodiment of the present application, the multi-cycle statistical characteristic includes a first statistical characteristic corresponding to a first statistical cycle and a second statistical characteristic corresponding to a second statistical cycle. Because the first statistical period and the second statistical period have different period lengths, the incidence relation between the target user and the target content in different time periods can be embodied through the first statistical characteristic and the second statistical characteristic, so that the change condition of the incidence relation on the time sequence can be embodied through the multi-period statistical characteristic, and further the intention change of the target user to the target content can be embodied. Due to the fact that the change condition of the intention of the target user to the target content along with the time sequence is combined, the processing equipment can depict the user preference more vividly and accurately through the multi-period statistical characteristics, and content pushing of the target user is facilitated subsequently.

For example, in one possible implementation, the target time interval may be 1 year, the first statistical period may be within 3 months, the second statistical period may be within 2 weeks, and the content behavior data may be the number of times the target content is pushed to the target user. The first statistical characteristic can be pushed 20 times in 3 months, and the second statistical characteristic can be pushed 0 times in 2 weeks, so that the target user can be determined to have lower and lower intention on the target content in the near future through the multi-period statistical characteristic.

S203: and determining the associated parameters of the target user and the target content according to the multi-period statistical characteristics.

In order to reasonably judge to which users the content should be pushed, the processing device may determine, based on a specific judgment criterion, association parameters corresponding to the multiple users, where the association parameters may embody the receptivity of the users to the content, so that the processing device may select the user with the higher receptivity to push the content.

As mentioned above, through the multi-period statistical characteristic, the processing device can determine a change trend of the intention of the target user to the target content along with time, and the intention of the target user to the target content can embody the receptivity of the target user to the target content to a certain extent. For example, in connection with the above example, if the number of times that the target content is pushed to the target user within 2 weeks is much less than the number of times that the target content is pushed within 3 months, it indicates that the intention degree of the target user to the target content is low in the recent period, and the determined association parameter may be low; if the number of times that the target content is pushed to the target user within 2 weeks is close to the number of times that the target content is pushed within 3 months, it is indicated that the pushing of the target content to the target user mainly hits the target content within 2 weeks, at this time, the intention degree of the target user to the target content is high, and the determined association parameter may be high.

S204: determining whether to push the target content to the target user based on the association parameter.

Through the association parameters, the processing equipment can determine the acceptance of the target content by the target user, so that based on the association parameters, the processing equipment can push the content to the target user when the acceptance of the target content by the target user is high, and from the perspective of the user, the processing equipment can push more content which is likely to be interested to the user, so that the content pushing experience of the user is improved; from the content perspective, the processing device can push the content to the user with higher acceptance, so that the content pushing efficiency and the content pushing effect are improved, and the two-way win-win of the content and the user is realized.

According to the technical scheme, in a content recommendation scene, due to the fact that the period lengths of the first statistical period and the second statistical period are different, through the change of content behavior data in different statistical periods, the change of the association between the target user and the target content in a time sequence angle is reflected, namely, the multi-period statistical characteristics can express the time sequence change trend of the target user on the intention of the target content. The multicycle statistical characteristics with rich expression capability play a role in more accurately depicting the preference of the target user, so that when the target content is pushed to the target user or not is measured through the association parameters determined by the multicycle statistical characteristics, if the target content is pushed, the possibility that the target user receives the target content is higher, and the content recommendation efficiency is improved.

In order to further improve the accuracy of the associated parameter, the processing device may further perform a more detailed analysis on the information used for determining the associated parameter. Next, a more detailed description will be made in terms of both the target content and the content behavior data.

(1) Target content angle

It is understood that in real life, the content pushed to the user generally includes a plurality of push types, and the push types can be divided according to the push purpose of the content pushed to the user. When the content is pushed to the user with different pushing purposes, the specific content information and the content pushing stage where the user is located may be different, and the content pushing stage can embody the degree of interaction between the user and the product corresponding to the content.

For example, when the product is an insurance product, the content may be an advertisement related to the insurance product, where the push type may include an insurance type, a verification type and a member interaction type, the push purpose of the marketing advertisement is to make the insurance product known to more users, the verification type advertisement may be an advertisement pushed when the user needs to verify the insurance product, and the push purpose may be to introduce more information related to the insurance product to the user if the user is already interested in the insurance product; the member interaction advertisement refers to an advertisement which is pushed to a user when the user performs related interaction after becoming a member of the insurance product, and the pushing purpose of the advertisement can introduce updated contents related to the insurance product to the user. Therefore, in the content pushing stage corresponding to the marketing advertisement, the verification advertisement and the member interaction advertisement respectively, the interaction degree of the user and the insurance product is continuously improved.

Because the interaction degrees between the user and the product are possibly different, if different types of target contents can be classified and analyzed, the processing equipment can obtain the association relationship between the target content and the target user at different content pushing stages to a certain extent, so that the information content included in the multi-period statistical data can be enriched, and the determination accuracy of the association parameters is further improved. In one possible implementation, the processing device may perform type division on the target content based on the push type to obtain multiple push type dimensions of the target content, where the push type dimensions may be used to perform classification statistics on content behavior data, for example, the multiple push type dimensions may include a marketing class dimension, a verification class dimension, a member interaction class dimension, and the like. When the multi-period statistical characteristics are determined, the processing equipment can perform data statistical processing based on different statistical periods on the content behavior data to obtain the multi-period statistical characteristics of the target user and the target content which are respectively related under multiple push type dimensions, so that the association relation between different types of target content and the target user for the same product can be embodied, and further the intention degree of the target user to the target content when the target user is in different content push stages can be embodied.

It is understood that, in addition to dividing the push type for the purpose of pushing, the processing device may determine the push type based on other manners, so that by determining the association relationship between different types of target content and the target user, other information related to the degree of intention of the target user can be obtained, which is not limited herein. Based on the association relationship between the target content of different push types and the target user, the processing device can more carefully depict the preference of the target user, so that the content can be more accurately pushed.

(2) Content behavior data angle

Since the association manner between the user and the content may include various ways, the content included in the content behavior data also has a certain diversity. For example, in one possible implementation, the content behavior data may include first data of the pushed target content of the target user and/or second data of the target user interacting with the pushed target content. The first data can embody the association between the target user and the target content from the content push level, and the second data can embody the association between the target user and the target content from the user interaction level, for example, the second data may be the click rate, the conversion rate, and the like of the target user on the target content. Therefore, through the first data and the second data, the processing device can determine the association relationship between the target user and the target content from different levels, further enrich the content which can be expressed by the content behavior data, enable the multi-period statistical characteristics determined based on the content behavior data to reflect the intention change of the target user to the target content from multiple levels, and finally determine more accurate association parameters.

In addition to improving the level of detail in analyzing data, expanding the amount of data used for analysis is one way to improve the accuracy of the final results. It is understood that if the user has a high degree of intent on content corresponding to a product, the user may have a high degree of intent on content corresponding to a product related to the product. For example, if the user has a high degree of intention for the related content of the insurance product, the financial product having a high degree of relation with the insurance product may also be interested. Based on the method, the processing equipment can further analyze the intention degree of the target content corresponding to the target user by combining the content corresponding to the related product of the product corresponding to the target content, and the richness of data expression is improved.

In a possible implementation manner, the processing device may first determine a first product field of a product corresponding to the target content, and then acquire, from the first product field, first behavior data of the target user corresponding to the target content in the target time interval, where the first behavior data is used to reflect an association between the target user and the target content in the target time interval. The processing device may then determine an associated product having an association with the product, the associated product belonging to a second product domain. The processing device may obtain, from the second product field, second behavior data of the target user corresponding to the associated content in the target time interval, where the second behavior data is used to represent an association between the target user and the associated content in the target time interval, and the associated content is content related to the associated product. The processing device can determine content behavior data according to the first behavior data and the second behavior data, so that the content behavior data can embody not only the association between the target user and the target content, but also the association between the target user and the related content in the related field, and further embody the intention of the target user to the target content and enrich the expression form and content of the data based on the association with the related content.

It can be understood that, in order to implement the above technical content, the associated product only needs to be associated with the product corresponding to the target content, and the product field to which the product belongs is not specifically limited, and the second product field and the first product field may be the same product field or different product fields. The product with the association refers to a plurality of products that may be interested simultaneously from the perspective of the user, and the specific form of the product is not limited. For example, in order to protect the income of financial products, more users may choose to purchase an insurance for the financial products, and thus the financial products and the insurance products belong to products that may be simultaneously interested by the users, i.e. belong to products with related relationships.

As mentioned above, obtaining the content data of the related product can expand the data volume, enrich the data expression, and make the processing device judge the user's intention more accurately. In some possible cases, the first behavior data acquired by the processing device from the first product field already has sufficient data volume, and at this time, the processing device may determine a more accurate multi-cycle statistical characteristic based on the first behavior data only, and does not need to acquire the second behavior data, so that the data volume required to be processed by the processing device is controlled within a reasonable range, and the processing efficiency is improved; in other possible cases, the processing device may obtain lower first behavior data from the first product domain, and the processing device may combine with second behavior data obtained from the second product domain to accurately determine the intent of the target user with a sufficient amount of data.

As can be seen, in order to be able to analyze the intention of the target user with a reasonable amount of data, the processing device may determine whether to acquire the second behavior data according to the amount of data of the acquired first behavior data. In one possible implementation, the processing device may determine a preset data amount, which is a data amount required to enable accurate analysis of the intention of the target user. The processing device may determine whether the data volume of the first behavior data satisfies a preset data volume, and if so, determine the first behavior data as content behavior data without acquiring second behavior data; if not, a step of determining associated products having an association with the product is performed, thereby enabling acquisition of second behavioural data from a second product domain to augment the data volume.

Meanwhile, the method and the device can determine the associated parameters by adopting the behavior data of the associated products and the associated contents, so that for some contents which are not associated by the user, even if the processing equipment cannot acquire the content behavior data corresponding to the content and the user, other related behavior data can be acquired to accurately judge the intention of the user, thereby properly solving the problem of 'cold start', and avoiding the condition that the user cannot be accurately pushed due to lack of data.

In addition, in order to more accurately determine the association parameters of the target user and the target content, in addition to the multi-period statistical features, the processing device can incorporate more features to enrich the data dimensionality. For example, in a possible implementation manner, the processing device may embody the association between the user and the product through the multi-period statistical feature, and may also introduce a user attribute feature and a product attribute feature, where the user attribute feature may embody an attribute feature of the user, and the product data feature may embody an attribute feature of the product. Therefore, based on the information, the processing device can judge the intention degree of the user more accurately from three layers of the user, the product and the association between the user and the content.

First, the processing device may determine product attribute characteristics of a product corresponding to the target content. The product attributes can be divided into numeric attributes and non-numeric attributes, the numeric attributes can include price, weight, quantity, and the like, and the non-numeric attributes can include product category, origin, and the like. The processing device may convert the non-numerical attributes into calculable numerical features for determination of the product attribute features, for example, the product category may be represented by a specific number, and when there are 10 categories of the product, the processing device may number the 10 categories as 1-10 and convert the 10 categories into one-hot wash vectors for feature calculation. If a product belongs to category 3 of 10 categories, the product attribute feature corresponding to the product may be [0,0,1,0,0,0,0,0 ].

Then, the processing device may determine the user attribute characteristics of the target user according to the user information of the target user, where the user data characteristics may include gender, age, region, and the like, the processing device may indicate the gender of the user by 0 and 1, and indicate the age of the user by using the age group, for example, the processing device may be divided into 7 age groups, each age group is respectively identified by an integer of 1 to 7, and the division method is 0 to 18, 18 to 25, 25 to 30, 30 to 40, 50 to 60, and 60 or more. For the region information, the processing device may number all regions, and use one-hot sparse vector to represent the region information.

The processing equipment can determine a classification fusion characteristic according to the product attribute characteristic, the user attribute characteristic and the multi-period statistical characteristic, and the classification fusion characteristic can embody the intention degree of the target user to the target product from the three layers. The processing equipment can determine the associated parameters of the target user and the target content according to the classification fusion characteristics, so that the associated parameters can more accurately depict the preference of the target user through richer data dimensions. Of course, the processing device may also merge features of more dimensions in addition to the above three features, and is not limited herein.

As mentioned above, the variation of the intention degree of the target user to the target content along with the time sequence can be reflected through the multi-cycle statistical characteristics of the target user related to the target content, and the user characteristics of one user can be analyzed from the things in which the user is interested, for example, if one user is interested in the related content of fitness equipment and outdoor sports equipment, the user is most likely to be a user fond of sports, and the processing device can push more related content of some sports products to the user in the subsequent process. Therefore, through the multi-period statistical characteristics, whether content is pushed to the user or not can be judged, and the favorite characteristics of the user can be depicted.

In a possible implementation manner, in order to obtain a richer and more stereoscopic user representation, the processing device may obtain total content behavior data of the target user corresponding to different contents in the target time interval, respectively, where the different contents include the target content, and the total content behavior data includes content behavior data corresponding to each content. The processing device can perform data statistics processing based on different statistical periods on the total content behavior data to obtain multi-period statistical characteristics of the target user respectively related to different contents, and determine a user interest portrait of the target user according to the multi-period statistical characteristics related to the different contents, wherein the user interest portrait is used for reflecting the interest orientation of the user to the contents. The multi-period statistical characteristics can reflect the change condition of the user intention degree along with the time sequence, so that the user interest portrait determined based on the multi-period statistical characteristics not only can reflect the interest of the user in multiple dimensions, but also can dynamically analyze the interest of the user, and has more flexible and accurate information reflecting effect relative to certain fixed user attributes such as age, gender and the like. Thus, based on the user interest representation, the processing device may mine users from the user population that are more interested in the related content when analysis of the users is needed.

When determining the association parameters in the above process, the processing device may adopt various modes. In one possible implementation, the processing device may determine the association parameters between the target user and the target content through a classifier model, which may be, for example, an XGBoost model, according to the multi-cycle statistical characteristics. Wherein the classifier model may be trained according to the following:

first, the processing device may determine a training sample according to the target content, where the training sample includes historical content behavior data of a sample user and a sample label for identifying whether the historical user has an association with the target content, and the historical content behavior data refers to content behavior data that has been used for determining an association with the target content in a historical time. The processing device may train an initial classifier based on the training samples to obtain a classifier model. In the training process, the processing device can determine the multi-period statistical characteristics of the corresponding target content based on the historical content behavior data through various statistical periods, and then learn the relationship between the multi-period statistical characteristics and the associated parameters, so that how to accurately determine the corresponding associated parameters through the multi-period statistical characteristics can be learned.

Besides the classifier model, the processing device may also determine the correlation parameter by using a logistic regression model, a decision tree model, a neural network model, a naive bayes model, or the like, which is not limited herein.

In order to obtain more accurate associated parameters through the classifier models, in a possible implementation manner, the number of the classifier models may be multiple, when the associated parameters are determined through the classifier models, the processing device may respectively determine the to-be-determined associated parameters of the target user and the target content through the multiple classifier models according to the multi-cycle statistical characteristics, and then determine the associated parameters of the target user and the target content according to the respectively-determined to-be-determined associated parameters, so that the processing device may synthesize the output results of the multiple classifier models to comprehensively judge the associated parameters, and avoid the problem that the output result is inaccurate due to the small probability abnormality of a few classifier models. For example, the processing device may take an average of a plurality of pending correlation parameters as the finally determined correlation parameter.

For example, the XGBoost model may include a plurality of base classifiers, and each base classifier corresponds to one sub-model. During training, the processing device may fuse the results of multiple base classifiers in the XGBoost, consider the complexity of each sub-model, and optimize the model complexity and the output result at the same time, where the objective function is:

wherein the content of the first and second substances,

for the loss function between the desired output and the predicted output, Ω (f) _k ) The model complexity of the kth sub-classifier, and w are the sub-classifier parameters to be learned. The processing device may perform multiple rounds of training optimization on the objective function, where the loss function expression at the t-th time is:

the processing device may perform a second order taylor expansion on the above equation, g being the first derivative, h being the 2 nd derivative:

when the XGboost is used, the processing equipment can determine i pending correlation parameters through the i submodels in the XGboost, and then determine the correlation parameters based on the i pending correlation parameters through the XGboost.

In addition to the above situation, the multiple classifier models may also be different models, for example, the XGBoost classifier model, the naive bayes classifier, and the neural network classifier may be simultaneously adopted to perform training respectively, when in use, the processing device may obtain three to-be-determined correlation parameters through the three different classifiers respectively, and then synthesize the three to-be-determined correlation parameters, for example, obtain final correlation parameters through weighting fusion or the like.

In order to more clearly understand the technical solution of the present application, a content push method provided in the embodiments of the present application will be introduced in combination with an actual application scenario.

Referring to fig. 3, fig. 3 is a schematic diagram of a content push method in an actual application scenario provided by the embodiment of the present application, in the actual application scenario, a processing device may be a content recommendation server, and content behavior data may be advertisement push data for pushing financial and insurance-type advertisements to a user. Referring to fig. 7, fig. 7 is a flowchart of a content push method in an actual application scenario provided in an embodiment of the present application, where the method includes:

s701: and acquiring total content behavior data of the target user corresponding to different contents in the target time interval.

The content push server may first obtain the advertisement push data, where the advertisement push data may include not only advertisement push data related to a target product pushed this time, but also advertisement push data corresponding to a related product, for example, the target content may be advertisement push data corresponding to a commodity a, a commodity B is a related product of a commodity a, the total content behavior data includes advertisement push data of a commodity a and advertisement push data of a commodity B, and the target time interval is 12 months.

S702: and determining the user attribute characteristics of the target user according to the user attributes of the target user.

S703: and carrying out data preprocessing.

The content push server may perform data preprocessing operations on the advertisement push data and the user attribute characteristics. In the data preprocessing process, the content push server can remove dirty data and invalid data, and selects advertisement data related to products from the advertisement push data according to the product information targeted by the advertisement push data. In addition, in order to make the data more reasonable, when the advertisement push data of the associated product is obtained, if the number of the advertisement push data corresponding to a certain associated product is less than N, the advertisement push data corresponding to the associated product can be removed, and the number of N can be determined based on historical data processing experience, so that the problems of too large error and the like caused by too small data amount can be avoided, and the accuracy of content push is further improved.

S704: and performing type division on different contents based on the push type to obtain a plurality of push type dimensions of the different contents.

After the advertisement push data is preprocessed, the content push server can classify the push data according to the push type corresponding to the advertisement push data, and perform statistical period division according to the classified advertisement push data.

S705: and performing data statistics processing based on different statistics periods on the total content behavior data to obtain multi-period statistics characteristics of the target user related to different contents respectively under multiple push type dimensions.

The statistical period may be 1 week, 2 weeks, 1 month, 3 months, 6 months and 12 months, and the processed advertisement push data is shown in the following table:

TABLE 1 advertisement push data for target user and A Commodity

TABLE 2 advertisement push data for target users and B commodities

The content push server may determine a multi-cycle statistical feature according to the advertisement push data after the classification statistics, and if the actual application scene has 6 statistical cycles, 10 commodities, and 3 push categories, a feature vector of the multi-cycle statistical feature with dimensions 6 × 10 × 3 — 180 may be generated.

S706: and determining the user interest portrait of the target user according to the multi-period statistical characteristics related to different contents.

In this embodiment, the advertisement push data may include advertisement push data of a product a corresponding to the target user and a product B corresponding to the target user, and the user interest image corresponding to the target user may be formed through a multi-period statistical characteristic determined based on these data.

S707: and determining the product attribute characteristics of the product corresponding to the target content.

In the actual application scenario, the product corresponding to the target content may be an article a, and the content push server may determine the product attribute feature corresponding to the article a.

S708: and determining classification fusion characteristics according to the product attribute characteristics, the user attribute characteristics and the multi-period statistical characteristics.

The content push server can perform feature fusion on the multi-period statistical features, the product attribute features and the user attribute features to obtain classified fusion features.

S709: and determining the associated parameters of the target user and the target content according to the classification fusion characteristics.

The content push server can determine the corresponding associated parameters of the target user through the XGboost classifier according to the classification fusion characteristics, and the associated parameters can reflect the intention degree of the target user for the related advertisement of the A commodity.

S710: determining whether to push the target content to the target user based on the association parameter.

The content push server can select the intention crowd based on the associated parameters, so that the content push of the user with higher intention degree of the related advertisement of the commodity A is determined, the push success rate and efficiency of the related advertisement of the commodity A are improved, and the popularization of the commodity A is facilitated.

Based on the content push method provided in the foregoing embodiment, an embodiment of the present application further provides a content push apparatus, referring to fig. 4, and fig. 4 is a block diagram of a structure of a content push apparatus 400 provided in the embodiment of the present application, where the apparatus 400 includes an obtaining unit 401, a first determining unit 402, a second determining unit 403, and a pushing unit 404:

an obtaining unit 401, configured to obtain content behavior data of a target content corresponding to a target user in a target time interval;

a first determining unit 402, configured to determine, according to the content behavior data, a multi-period statistical feature of the target user related to the target content through data periodic statistics, where the multi-period statistical feature includes a first statistical feature corresponding to a first statistical period and a second statistical feature corresponding to a second statistical period, the first statistical period and the second statistical period have different period lengths, and the period lengths of the first statistical period and the second statistical period are both less than or equal to a duration identified by the target time interval;

a second determining unit 403, configured to determine, according to the multi-cycle statistical characteristic, an association parameter between the target user and the target content;

a pushing unit 404, configured to determine whether to push the target content to the target user based on the association parameter.

In one possible implementation, the apparatus 400 further includes a classification unit:

the classification unit is used for carrying out type division on the target content based on the push type to obtain a plurality of push type dimensions of the target content;

the first determining unit 402 is specifically configured to:

and according to the content behavior data, determining multi-period statistical characteristics respectively related to the target content and the target user in the plurality of push type dimensions through data periodic statistics.

In a possible implementation manner, the content behavior data includes first data of the target user pushed the target content and/or second data of the target user interacting with the pushed target content.

In a possible implementation manner, the obtaining unit 401 is specifically configured to:

determining a first product field of a product corresponding to the target content;

acquiring first behavior data of the target user corresponding to the target content in the target time interval from the first product field;

determining an associated product having an association with the product, the associated product belonging to a second product domain;

acquiring second behavior data of the target user corresponding to the associated content in the target time interval from the second product field, wherein the associated content is related to the associated product;

determining the content behavior data from the first behavior data and the second behavior data.

In one possible implementation, the apparatus 400 further includes a third determining unit:

a third determination unit configured to determine whether a data amount of the first behavior data satisfies a preset data amount;

if so, determining the first behavior data as the content behavior data;

if not, the step of determining the associated product associated with the product is performed.

In one possible implementation, the apparatus 400 further includes a fourth determining unit and a fifth determining unit:

a fourth determining unit, configured to determine a product attribute feature of a product corresponding to the target content;

a fifth determining unit, configured to determine a user attribute feature of the target user according to the user information of the target user;

the second determination unit 403 is specifically used for

Determining classification fusion characteristics according to the product attribute characteristics, the user attribute characteristics and the multi-period statistical characteristics;

and determining the association parameters of the target user and the target content according to the classification fusion characteristics.

acquiring total content behavior data of the target user respectively corresponding to different contents in the target time interval, wherein the different contents comprise the target contents;

the first determination unit 401 is specifically configured to

According to the total content behavior data, determining multi-period statistical characteristics of the target user respectively related to the different contents through data periodic statistics;

and determining the user interest portrait of the target user according to the multi-period statistical characteristics related to the different contents.

In a possible implementation manner, the second determining unit 403 is specifically configured to:

determining the associated parameters of the target user and the target content through a classifier model according to the multi-period statistical characteristics;

wherein the classifier model is trained according to the following:

determining a training sample according to the target content, wherein the training sample comprises historical content behavior data of a sample user and a sample label for identifying whether the historical user has an association with the target content;

and training an initial classifier according to the training sample to obtain the classifier model.

In a possible implementation manner, the number of the classifier models is multiple, and the second determining unit 403 is specifically configured to:

respectively determining undetermined association parameters of the target user and the target content through a plurality of classifier models according to the multi-period statistical characteristics;

and determining the association parameters of the target user and the target content according to the respectively determined pending association parameters.

The embodiment of the present application further provides a computer device, which is described below with reference to the accompanying drawings. Referring to fig. 5, an embodiment of the present application provides a device, which may also be a terminal device, where the terminal device may be any intelligent terminal including a mobile phone, a tablet computer, a Personal Digital Assistant (PDA), a Point of Sales (POS), a vehicle-mounted computer, and the terminal device is taken as the mobile phone as an example:

fig. 5 is a block diagram illustrating a partial structure of a mobile phone related to a terminal device provided in an embodiment of the present application. Referring to fig. 5, the handset includes: a Radio Frequency (RF) circuit 510, a memory 520, an input unit 530, a display unit 540, a sensor 550, an audio circuit 560, a wireless fidelity (WiFi) module 570, a processor 580, and a power supply 590. Those skilled in the art will appreciate that the handset configuration shown in fig. 5 is not intended to be limiting and may include more or fewer components than those shown, or some components may be combined, or a different arrangement of components.

The following describes each component of the mobile phone in detail with reference to fig. 5:

RF circuit 510 may be used for receiving and transmitting signals during information transmission and reception or during a call, and in particular, for processing downlink information of a base station after receiving the downlink information to processor 580; in addition, the data for designing uplink is transmitted to the base station. In general, RF circuit 510 includes, but is not limited to, an antenna, at least one Amplifier, a transceiver, a coupler, a Low Noise Amplifier (LNA), a duplexer, and the like. In addition, RF circuit 510 may also communicate with networks and other devices via wireless communications. The wireless communication may use any communication standard or protocol, including but not limited to Global System for Mobile communication (GSM), General Packet Radio Service (GPRS), Code Division Multiple Access (CDMA), Wideband Code Division Multiple Access (WCDMA), Long Term Evolution (LTE), email, Short Message Service (SMS), and the like.

The memory 520 may be used to store software programs and modules, and the processor 580 executes various functional applications and data processing of the mobile phone by operating the software programs and modules stored in the memory 520. The memory 520 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required by at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data (such as audio data, a phonebook, etc.) created according to the use of the cellular phone, and the like. Further, memory 520 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid state storage device.

The input unit 530 may be used to receive input numeric or character information and generate key signal inputs related to user settings and function control of the cellular phone. Specifically, the input unit 530 may include a touch panel 531 and other input devices 532. The touch panel 531, also called a touch screen, can collect touch operations of a user (such as operations of the user on the touch panel 531 or near the touch panel 531 by using a finger, a stylus pen or any other suitable object or accessory) thereon or nearby, and drive the corresponding connection device according to a preset program. Alternatively, the touch panel 531 may include two parts, a touch detection device and a touch controller. The touch detection device detects the touch direction of a user, detects a signal brought by touch operation and transmits the signal to the touch controller; the touch controller receives touch information from the touch sensing device, converts the touch information into touch point coordinates, and sends the touch point coordinates to the processor 580, and can receive and execute commands sent by the processor 580. In addition, the touch panel 531 may be implemented by various types such as a resistive type, a capacitive type, an infrared ray, and a surface acoustic wave. The input unit 530 may include other input devices 532 in addition to the touch panel 531. In particular, other input devices 532 may include, but are not limited to, one or more of a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a trackball, a mouse, a joystick, and the like.

The display unit 540 may be used to display information input by the user or information provided to the user and various menus of the mobile phone. The Display unit 540 may include a Display panel 541, and optionally, the Display panel 541 may be configured in the form of a Liquid Crystal Display (LCD), an Organic Light-Emitting Diode (OLED), or the like. Further, the touch panel 531 may cover the display panel 541, and when the touch panel 531 detects a touch operation on or near the touch panel 531, the touch panel is transmitted to the processor 580 to determine the type of the touch event, and then the processor 580 provides a corresponding visual output on the display panel 541 according to the type of the touch event. Although in fig. 5, the touch panel 531 and the display panel 541 are two independent components to implement the input and output functions of the mobile phone, in some embodiments, the touch panel 531 and the display panel 541 may be integrated to implement the input and output functions of the mobile phone.

The handset may also include at least one sensor 550, such as a light sensor, motion sensor, and other sensors. Specifically, the light sensor may include an ambient light sensor and a proximity sensor, wherein the ambient light sensor may adjust the brightness of the display panel 541 according to the brightness of ambient light, and the proximity sensor may turn off the display panel 541 and/or the backlight when the mobile phone moves to the ear. As one of the motion sensors, the accelerometer sensor can detect the magnitude of acceleration in each direction (generally, three axes), can detect the magnitude and direction of gravity when stationary, and can be used for applications of recognizing the posture of a mobile phone (such as horizontal and vertical screen switching, related games, magnetometer posture calibration), vibration recognition related functions (such as pedometer and tapping), and the like; as for other sensors such as a gyroscope, a barometer, a hygrometer, a thermometer, and an infrared sensor, which can be configured on the mobile phone, further description is omitted here.

Audio circuitry 560, speaker 561, and microphone 562 may provide an audio interface between a user and a cell phone. The audio circuit 560 may transmit the electrical signal converted from the received audio data to the speaker 561, and the electrical signal is converted into an audio signal by the speaker 561 and output; on the other hand, the microphone 562 converts the collected sound signals into electrical signals, which are received by the audio circuit 560 and converted into audio data, which are then processed by the audio data output processor 580, and then passed through the RF circuit 510 to be sent to, for example, another cellular phone, or output to the memory 520 for further processing.

WiFi belongs to short distance wireless transmission technology, and the mobile phone can help the user to send and receive e-mail, browse web pages, access streaming media, etc. through the WiFi module 570, which provides wireless broadband internet access for the user. Although fig. 5 shows the WiFi module 570, it is understood that it does not belong to the essential constitution of the handset, and can be omitted entirely as needed within the scope not changing the essence of the invention.

The processor 580 is a control center of the mobile phone, connects various parts of the entire mobile phone by using various interfaces and lines, and performs various functions of the mobile phone and processes data by operating or executing software programs and/or modules stored in the memory 520 and calling data stored in the memory 520, thereby performing overall monitoring of the mobile phone. Alternatively, processor 580 may include one or more processing units; preferably, the processor 580 may integrate an application processor, which mainly handles operating systems, user interfaces, application programs, etc., and a modem processor, which mainly handles wireless communications. It will be appreciated that the modem processor described above may not be integrated into processor 580.

The handset also includes a power supply 590 (e.g., a battery) for powering the various components, which may preferably be logically coupled to the processor 580 via a power management system, such that the power management system may be used to manage charging, discharging, and power consumption.

Although not shown, the mobile phone may further include a camera, a bluetooth module, etc., which are not described herein.

In this embodiment, the processor 580 included in the terminal device further has the following functions:

determining multi-period statistical characteristics related to the target content of the target user through data periodic statistics according to the content behavior data, wherein the multi-period statistical characteristics comprise first statistical characteristics corresponding to a first statistical period and second statistical characteristics corresponding to a second statistical period, the first statistical period and the second statistical period have different period lengths, and the period lengths of the first statistical period and the second statistical period are both smaller than or equal to the duration identified by the target time interval;

Referring to fig. 6, fig. 6 is a block diagram of a server 600 provided in this embodiment, and the server 600 may have a relatively large difference due to different configurations or performances, and may include one or more Central Processing Units (CPUs) 622 (e.g., one or more processors) and a memory 632, and one or more storage media 630 (e.g., one or more mass storage devices) for storing applications 642 or data 644. Memory 632 and storage medium 630 may be, among other things, transient or persistent storage. The program stored in the storage medium 630 may include one or more modules (not shown), each of which may include a series of instruction operations for the server. Still further, the central processor 622 may be configured to communicate with the storage medium 630 to execute a series of instruction operations in the storage medium 630 on the server 600.

The server 600 may also include one or more power supplies 626, one or more wired or wireless network interfaces 650, one or more input-output interfaces 658, and/or one or more operating systems 641, such as Windows Server, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, and so forth.

The steps performed by the server in the above embodiments may be based on the server structure shown in fig. 6.

The embodiment of the present application further provides a computer-readable storage medium for storing a computer program, where the computer program is used to execute any implementation manner of the content push method described in the foregoing embodiments.

Those of ordinary skill in the art will understand that: all or part of the steps for realizing the method embodiments can be completed by hardware related to program instructions, the program can be stored in a computer readable storage medium, and the program executes the steps comprising the method embodiments when executed; and the aforementioned storage medium may be at least one of the following media: various media that can store program codes, such as read-only memory (ROM), RAM, magnetic disk, or optical disk.

It should be noted that, in the present specification, all the embodiments are described in a progressive manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the apparatus and system embodiments, since they are substantially similar to the method embodiments, they are described in a relatively simple manner, and reference may be made to some of the descriptions of the method embodiments for related points. The above-described embodiments of the apparatus and system are merely illustrative, and units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one position, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.

The above description is only one specific embodiment of the present application, but the scope of the present application is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present application should be covered within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A method for pushing content, the method comprising:

2. The method of claim 1, further comprising:

performing type division on the target content based on a push type to obtain a plurality of push type dimensions of the target content;

the performing data statistics processing based on different statistics periods on the content behavior data to obtain multi-period statistics characteristics related to the target content and the target user includes:

and performing data statistics processing based on different statistics periods on the content behavior data to obtain multi-period statistics characteristics respectively related to the target user and the target content under the multiple push type dimensions.

3. The method of claim 1, wherein the content behavior data comprises first data of the target user pushed the target content and/or second data of the target user interacting with the pushed target content.

4. The method of claim 1, wherein the obtaining content behavior data of the target user corresponding to the target content in the target time interval comprises:

5. The method of claim 4, further comprising:

determining whether the data volume of the first behavior data meets a preset data volume;

if so, determining the first behavior data as the content behavior data;

6. The method of claim 1, further comprising:

determining product attribute characteristics of a product corresponding to the target content;

determining the user attribute characteristics of the target user according to the user information of the target user;

the determining the association parameters of the target user and the target content according to the multi-period statistical characteristics comprises:

7. The method according to any one of claims 1 to 6, wherein the obtaining content behavior data of the target user corresponding to the target content in the target time interval comprises:

acquiring total content behavior data of the target user corresponding to different contents respectively in the target time interval, wherein the different contents comprise the target content;

the performing data statistics processing based on different statistics periods on the content behavior data to obtain multi-period statistical characteristics of the target user related to the target content includes:

performing data statistics processing based on different statistics periods on the total content behavior data to obtain multi-period statistical characteristics of the target user respectively related to the different contents;

8. The method according to any one of claims 1 to 6, wherein the determining the association parameter of the target user with the target content according to the multi-period statistical characteristics comprises:

wherein the classifier model is trained according to the following:

9. The method according to claim 8, wherein the number of the classifier models is plural, and the determining the association parameter of the target user and the target content through the classifier models according to the multi-cycle statistical features comprises:

respectively determining undetermined associated parameters of the target user and the target content through a plurality of classifier models according to the multi-period statistical characteristics;

10. A content pushing apparatus, characterized in that the apparatus comprises an acquisition unit, a first determination unit, a second determination unit, and a pushing unit:

11. The apparatus according to claim 10, characterized in that the apparatus further comprises a classification unit:

the classification unit is used for performing type division on the target content based on the push type to obtain a plurality of push type dimensions of the target content;

the first determining unit is specifically configured to:

12. The apparatus of claim 10, wherein the content behavior data comprises first data of the target user pushed the target content and/or second data of the target user interacting with the pushed target content.

13. The apparatus according to claim 10, wherein the obtaining unit is specifically configured to:

14. A computer device, the device comprising a processor and a memory:

the processor is configured to execute the content push method according to any one of claims 1 to 9 according to instructions in the program code.

15. A computer-readable storage medium for storing a computer program for executing the content push method according to any one of claims 1 to 9.