Information push method, device, server and storage medium

By analyzing the user's stable and recent portrait characteristics and combining scene strategies to push personalized information, the problem of low targeting of the existing intelligent recommendation system is solved and the user experience is improved.

CN112417301BActive Publication Date: 2024-12-03CHINA CONSTRUCTION BANK
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202011437267.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-12-07
Publication Date
2024-12-03
Estimated Expiration
2040-12-07

AI Technical Summary

Technical Problem

The existing intelligent recommendation system has low targeted problems in user personalized recommendations, resulting in limited user experience.

Method used

By obtaining the user's stable image feature set and recent image feature set, analyzing recent feature changes, determining the target object, and using a preset push mode to push personalized information according to the current scene, including integrating scene-based, stack-pressing and general message modes.

Benefits of technology

It improves the initiative, pertinence and accuracy of information push, and improves user preference and usage satisfaction.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN112417301B_ABST
    Figure CN112417301B_ABST
Patent Text Reader

Abstract

The embodiments of the present application provide an information push method, device, server and storage medium. The method obtains a stable portrait feature set and a recent portrait feature set of a user, and determines whether a feature change has occurred in the recent portrait feature set based on the stable portrait feature set. When it is determined that a feature change has occurred in the recent portrait feature set, the target object recommended to the user is determined based on the target portrait feature that caused the feature change. According to the current scenario, the information of the target object is pushed to the user using a preset push mode, thereby realizing personalized information push, improving the initiative, pertinence and accuracy of information push, and improving the user experience.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The embodiments of the present application relate to the field of Internet technology, and in particular, to an information push method, device, server and storage medium. Background Art

[0002] The rapid development of Internet technology and the rapid popularization of intelligent terminals have provided favorable infrastructure conditions for the development of e-commerce, enabling the rapid rise and vigorous development of e-commerce. As one of the important links in e-commerce, intelligent recommendation (recommending products and information to users) is the key to business expansion and customer stickiness for e-commerce companies. Therefore, it has become an important task for many e-commerce companies.

[0003] At present, intelligent recommendations are mostly based on offline data of a large number of users to analyze the content that should be recommended to users. However, this solution often recommends the same content or random content to all users at the same time, which is not very targeted and affects the user experience. Summary of the invention

[0004] The embodiments of the present application provide an information push method, device, server and storage medium to solve the problem of low pertinence in the prior art.

[0005] In a first aspect, an embodiment of the present application provides an information push method, including:

[0006] Obtain the user's stable portrait feature set and recent portrait feature set;

[0007] Determining whether a feature change occurs in the recent portrait feature set according to the stable portrait feature set;

[0008] When it is determined that the recent portrait feature set has undergone a feature change, determining a target object to be recommended to the user based on the target portrait feature causing the feature change;

[0009] According to the current scenario, the information of the target object is pushed to the user using a preset push mode.

[0010] Optionally, determining whether a feature change occurs in the recent portrait feature set according to the stable portrait feature set includes:

[0011] Determining whether there is a difference between the recent portrait feature set and the stable portrait feature set;

[0012] If there is a difference between the recent portrait feature set and the stable portrait feature set, determining whether the difference is a level difference;

[0013] If the difference is a level difference, it is determined that a feature change has occurred in the recent portrait feature set.

[0014] Optionally, determining the target object to be recommended to the user according to the target portrait feature causing the feature change includes:

[0015] Extracting the target image feature causing the feature change from the recent image feature set;

[0016] The target image feature is matched with the candidate object, and the candidate object that successfully matches the target image feature is determined as the target object.

[0017] Optionally, the using a preset push mode to push the information of the target object to the user according to the current scenario includes:

[0018] Determining whether the user is online;

[0019] If the user is online, determining whether there is an association relationship between the target object and the page currently visited by the user;

[0020] If there is an association relationship between the target object and the user's currently visited page, the information of the target object is immediately pushed to the user by integrating into the scenario mode.

[0021] Optionally, the method further comprises:

[0022] If the user is not online or there is no association relationship between the target object and the page currently visited by the user, determining whether the user is a high-frequency user recently;

[0023] If the user is a high-frequency user recently, the push strategy is adopted to push the information of the target object to the user by integrating the scenario mode.

[0024] Optionally, the method further comprises:

[0025] If the user is not a high-frequency user recently, the information of the target object is pushed to the user through a general message mode.

[0026] Optionally, the step of obtaining a stable portrait feature set and a recent portrait feature set of a user includes:

[0027] Obtaining historical behavior data and recent behavior data of the user;

[0028] Analyze the historical behavior data according to data mining technology to obtain a stable profile feature set of the user;

[0029] The recent behavior data is analyzed according to data mining technology to obtain a recent portrait feature set of the user.

[0030] Optionally, before pushing the information of the target object to the user in a preset push mode according to the current scenario, the method further includes:

[0031] Determining whether the target object has been recommended to the user recently;

[0032] If the target object has not been recommended to the user recently, construct push information according to the target object;

[0033] The push information is privacy filtered to obtain the information of the target object.

[0034] Optionally, performing privacy filtering on the push information to obtain the information of the target object includes:

[0035] Determining whether the push information contains sensitive information;

[0036] If the push information contains sensitive information, determining whether the user has authorized the push of sensitive information;

[0037] If the user has authorized the push of sensitive information, the push information is used as the information of the target object;

[0038] If the user does not authorize the push of sensitive information, the sensitive information is processed to obtain the information of the target object.

[0039] In a second aspect, an embodiment of the present application provides an information push device, including:

[0040] An acquisition module is used to obtain a user's stable portrait feature set and a recent portrait feature set;

[0041] a processing module, configured to determine whether a feature change occurs in the recent portrait feature set according to the stable portrait feature set; when it is determined that a feature change occurs in the recent portrait feature set, determine a target object to be recommended to the user according to the target portrait feature causing the feature change;

[0042] The push module is used to push the information of the target object to the user in a preset push mode according to the current scenario.

[0043] In a third aspect, an embodiment of the present application provides a server, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the information push method as described in the first aspect above is implemented.

[0044] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the information push method as described in the first aspect above.

[0045] The information push method, device, server and storage medium provided in the embodiments of the present application obtain a stable portrait feature set and a recent portrait feature set of a user, determine whether a feature change has occurred in the recent portrait feature set based on the stable portrait feature set, and when it is determined that a feature change has occurred in the recent portrait feature set, determine the target object to be recommended to the user based on the target portrait feature that caused the feature change, and according to the current scenario, use a preset push mode to push the information of the target object to the user, determine the recommended object based on the mutation features in the user's recent portrait feature set, and select an appropriate push mode to push the information of the target object based on the current scenario, thereby realizing personalized information push, which not only improves the initiative, pertinence and accuracy of information push, but also improves the user's favorability and usage satisfaction, thereby improving the user's usage experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A schematic diagram of an application scenario of an embodiment of the present application;

[0047] Figure 2 A flowchart of the information push method provided in Example 1 of the present application;

[0048] Figure 3 A schematic diagram of an execution logic of S104 in the first implementation of this application;

[0049] Figure 4 A flowchart of the information push method provided in Example 2 of the present application;

[0050] Figure 5 A schematic diagram of the structure of the information push device provided in Example 3 of the present application;

[0051] Figure 6 A schematic diagram of the structure of a server provided in Example 4 of the present application. DETAILED DESCRIPTION

[0052] The present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for ease of description, only the parts related to the present application, rather than all structures, are shown in the accompanying drawings.

[0053] The main idea of ​​the technical solution of the present application: Based on the technical problems existing in the prior art, the embodiment of the present application provides a technical solution for information push, which finds the "mutation" points of user portrait features based on data mining, behavior analysis, etc., determines the recommended objects, and pushes personalized products and information to the user's device based on scenario strategies (such as whether the user is online, the relevance of the pages visited by the user and the determined recommended objects, etc.). It comprehensively adopts information collection, data mining, recommendation object adaptation, push message filtering, push strategy / mode adaptation and other technologies to realize personalized information push, improve the initiative, pertinence and accuracy of information push, and improve user favorability and usage satisfaction.

[0054] For example, Figure 1 Schematic diagram of application scenarios of the embodiments of the present application, such as Figure 1 As shown, this embodiment provides an information push system including: a server and multiple user devices, the server and the multiple user devices are communicatively connected, the user devices can be electronic devices that directly serve users, such as smart speakers, mobile phones, tablet computers, etc., each user device is installed with an application client of the target application, and the server is installed with an application server of the target application. The user accesses the target application through the application client installed on the user device used by the user, uses the services provided by the target application and obtains related products and information of the target application pushed by the server. The server is used to manage the target application and execute the information push methods in the following embodiments to push products, information, etc. related to the target application to the user.

[0055] Embodiment 1

[0056] Figure 2 This is a flow chart of the information push method provided in the first embodiment of the present application. The method of this embodiment can be executed by the information push device provided in the embodiment of the present application. The device can be implemented by software and / or hardware and can be integrated in the server. Figure 2 As shown, the information push method of this embodiment includes:

[0057] S101. Obtain a stable profile feature set and a recent profile feature set of the user.

[0058] In this embodiment, the user's behavior data is analyzed by data mining technology to obtain a portrait describing the user, and then the user features in the portrait are extracted to obtain the user's portrait feature set. In this step, the user's stable portrait feature set and recent portrait feature set can be periodically and actively obtained, and the user's role, identity, etc. are located and analyzed based on the stable portrait feature set and the recent portrait feature set, so as to timely discover some changes in the user, analyze these changes, and push information based on the analysis results, which helps to improve the initiative and accuracy of information push.

[0059] Among them, the stable portrait feature set is a set composed of a series of stable portrait features. The stable portrait features are mainly used to describe users' long-term and stable preferences, habits, ability levels, etc.

[0060] The recent portrait feature set is a set consisting of a series of recent portrait features. The recent portrait features are mainly used to describe the user's preferences, habits, ability level, etc. in the recent period of time.

[0061] In a possible implementation, in this step, by obtaining the user's historical behavior data and recent behavior data, the historical behavior data is analyzed according to data mining technology to obtain the user's stable portrait feature set, and the recent behavior data is analyzed according to data mining technology to obtain the user's recent portrait feature set.

[0062] In this embodiment, the server can accumulate user behavior data through the data warehouse thereon. Optionally, the data in the data warehouse includes user device-side data, server-side data, and third-party data. The user device-side data can be obtained by the server through the embedding points set on the user device and obtained from the user device through the time sliding window. The third-party data can be obtained by the server through the embedding points set on other devices and obtained from other devices (see Figure 1 ) obtained.

[0063] Among them, historical behavior data refers to the user behavior data accumulated in the data warehouse over a relatively long period of time (such as the past year) from the current date to a point in the past; recent behavior data refers to the user behavior data accumulated in the data warehouse over a relatively short period of time (such as the past week) from the current date to a point in the past. Accordingly, in this embodiment, historical behavior data and recent behavior data can be filtered out from the data warehouse by means of time query.

[0064] Optionally, in this embodiment, the time length of historical behavior data and the time length of recent behavior data can be pre-set, and the period for obtaining the user portrait feature set can be set, so that the server can periodically and automatically complete the acquisition of the user's stable portrait feature set and recent portrait feature set.

[0065] Optionally, in this embodiment, a data analysis model may be pre-trained, and the data analysis model obtains and analyzes the user's historical behavior data and recent behavior data, and ultimately outputs the user's stable portrait feature set and recent portrait feature set.

[0066] Optionally, in this embodiment, categories of portrait features of interest can be defined in advance, such as occupation, income level, consumption characteristics, interests and hobbies, etc., and stable portrait feature sets and recent portrait feature sets can be acquired based on the defined portrait feature categories of interest. Since only stable portrait features and recent portrait features corresponding to the portrait feature categories of interest need to be acquired, data processing efficiency is improved and server load is reduced.

[0067] S102. Determine whether a recent portrait feature set undergoes feature changes based on a stable portrait feature set.

[0068] In this step, after S101, by comparing the recent portrait feature set with the stable portrait feature set, it is determined whether the user's recent portrait feature set has undergone feature changes, wherein feature change refers to the level change of portrait features of the same category. For example, the income level is divided into three levels: level one (less than 10,000), level two (10,000 to 30,000), and level three (greater than 30,000). The actual value of the user's income level obtained is a portrait feature of the user. If the recent portrait feature of a user's income level obtained is different from the level of the stable portrait feature, it can be determined that the recent portrait feature corresponding to the user's income level has undergone a level change relative to the stable portrait feature.

[0069] In a possible implementation, in this embodiment, it is determined whether there is a difference between the recent portrait feature set and the stable portrait feature set; if there is a difference between the recent portrait feature set and the stable portrait feature set, it is determined whether the difference is a level difference; if the difference is a level difference, it is determined that the recent portrait feature set has undergone feature changes.

[0070] Optionally, in this embodiment, by comparing the recent portrait features in the recent portrait feature set with the stable portrait features of the same category in the stable portrait feature set, it is determined whether there is a difference between the two. It can be understood that as long as there is a difference between a set of recent portrait features and the stable portrait features, it can be determined that there is a difference between the recent portrait feature set and the stable portrait feature set.

[0071] Optionally, in this embodiment, the correspondence between the portrait features and the levels can be pre-set and stored in the data warehouse of the server, by respectively determining the levels of the recent portrait features and the stable portrait features in each group of recent portrait features and stable portrait features that have differences, and determining whether the levels of the two are the same, if not, then determining that the difference between the two is a level difference. It can be understood that as long as the difference between a group of recent portrait features and stable portrait features is a level difference, it can be determined that the difference between the recent portrait feature set and the stable portrait feature set is a level difference, and then determining that the recent portrait feature set has undergone a feature change. On the contrary, if it is determined that there is no difference between the recent portrait feature set and the stable portrait feature set or the difference that exists is not a level difference, then it is determined that the recent portrait feature set has not undergone a feature change, and the process ends, and information is not pushed to the user for the time being.

[0072] S103. When it is determined that a recent portrait feature set has undergone a feature change, a target object to be recommended to the user is determined based on the target portrait feature causing the feature change.

[0073] In this step, if it is determined according to S102 that a feature change has occurred in the recent portrait feature set, the object recommended to the user is determined based on the target portrait feature that caused the feature change in the recent portrait feature set, thereby ensuring that the object recommended to the user is closer to the user's current preferences, habits, and ability level, etc., and can better meet the user's needs. Among them, the target portrait feature is one or more recent portrait features in the recent portrait feature set that have a level difference with the stable portrait feature set. The object recommended to the user can be a product or information, etc., and accordingly, the target object can be a specific product among the candidate products or a specific information among the candidate information, etc.

[0074] In a possible implementation, in this step, the target portrait features that cause feature changes are extracted from the recent portrait feature set, and the target portrait features are matched with candidate objects, and the candidate object that successfully matches the target portrait features is determined as the target object.

[0075] Among them, the candidate objects are some pre-customized products or information related to the target application. For example, if the target application is mobile banking, the candidate objects can be a series of financial products customized for the user. If the target application is an entertainment application, the candidate objects can be a series of information related to movies, books, and music.

[0076] Optionally, in this embodiment, a corresponding user feature portrait can be set in advance for each candidate object, and the number of target portrait features included in the user feature portrait of each candidate object can be determined by matching the target portrait features with the user feature portrait corresponding to each candidate object. The determined number is compared with a preset number threshold, and the candidate object containing the number of target portrait features greater than the number threshold is determined as the target object.

[0077] Optionally, in this embodiment, the target object can be obtained by matching the target portrait features with the candidate objects through a corresponding matching algorithm and / or a pre-trained matching model. For example, when there are multiple target portrait features, a score for each candidate object is obtained based on the weight of each target portrait feature for different candidate objects, and the candidate object whose score exceeds a preset score threshold is determined as the target object.

[0078] It is understandable that when there are multiple target objects, the multiple target objects can be recommended in batches or all at once. When all candidate objects fail to match the target portrait features, the process ends and no information is recommended to the user.

[0079] S104: According to the current scenario, the information of the target object is pushed to the user using a preset push mode.

[0080] In this step, in order to enhance the recommendation effect and user targeting of the target object, a push mode suitable for the current scenario is selected from the preset push modes, and the information of the target object is pushed to the user using the push mode.

[0081] The target object information is information generated based on the target object. Optionally, the target object information includes at least one of the name of the target object, the reason for recommending the target object, the introduction of the target object, or the benefits of using the target object.

[0082] The current scenario refers to some situations in which the user currently uses the target application, such as whether the user is online, the page the user is currently visiting, etc.

[0083] Push mode refers to the way of pushing information to the target object. Optionally, the preset push modes include scenario-based mode and general message mode. Among them, scenario-based mode refers to the way of embedding the recommended products and information into the relevant pages for one-time display. It is mainly achieved by reserving a display window in the application client page design, adapting the product or information binding on the application server, and checking whether there are adaptable products or information that can be pushed during real-time access. When there are products or information that can be pushed, a one-time display is performed through the reserved display window in the access page.

[0084] General message mode: refers to a push mode in which information is pushed through the message center of the application client in the form of a dedicated message presentation window. Information pushed using this push mode can only be seen by users who are connected to the Internet and online and have logged into the message center.

[0085] It can be understood that in this step, pushing the target object information to the user means displaying the target object information through the application client on the user device to achieve the purpose of pushing the target object information to the user.

[0086] In one possible implementation, Figure 3 A schematic diagram of an execution logic of S104 in one implementation of this application is shown as follows: Figure 3 As shown, in this implementation, according to the current scenario, the information of the target object is pushed to the user using a preset push mode, including:

[0087] (1) Determine whether the user is online.

[0088] In this step, you can check the user status recorded by the application server to determine whether the user is online, that is, whether the user has successfully logged in to the application client.

[0089] (2) If the user is online, determine whether there is an association between the target object and the page the user is currently visiting.

[0090] In this step, if it is determined according to step (1) that the user is online, it is further determined whether there is an association relationship between the target object and the page currently being visited by the user. For example, the target object is a certain financial product. If the page currently being visited by the user is also the page of the financial product, it is determined that there is an association relationship between the target object and the page currently being visited by the user. If the page currently being visited by the user is the user's basic information page, it is determined that there is no association relationship between the target object and the page currently being visited by the user.

[0091] (3) If there is an association between the target object and the user's current visited page, the target object's information is immediately pushed to the user by integrating into the scenario-based mode.

[0092] In this step, if it is determined according to step (2) that there is an association between the target object and the user's current visited page, the information of the target object is immediately pushed to the user in a scenario-based mode, and the target object information is loaded and displayed through the reserved display window in the currently visited page, thereby improving the pertinence and accuracy of the push. In addition, the target object information can be pushed to the user without the user having to jump out of the current page and without affecting the display effect of the user's current page, thereby improving the user experience.

[0093] (4) If the user is not online or there is no association between the target object and the user's currently visited page, determine whether the user is a high-frequency user recently.

[0094] In this step, if it is determined according to step (1) that the user is not online or it is determined according to step (2) that there is no association between the target object and the user's currently visited page, it is further determined whether the user is a high-frequency user of the target application in the recent period.

[0095] Among them, recent time refers to a period of time not far from the current time. It can be understood that the time length for determining whether the user is a high-frequency user in the recent time in this step can be the same as the time length for obtaining the user's recent behavior data, or it can be different. It can be set according to actual conditions and is not limited here.

[0096] Optionally, in this step, the user's recent access records can be obtained, and based on the number of visits and visit duration of the user to the target application, it can be determined whether the user is a high-frequency user in the recent period. Specifically, a threshold for the number of visits or the visit duration can be set, and based on whether the number of visits or the visit duration of the user to the target application in the recent period exceeds the corresponding threshold, it can be determined whether the user is a high-frequency user in the recent period.

[0097] (5) If the user is a high-frequency user recently, the push strategy is adopted to push the target object's information to the user by integrating the scenario model.

[0098] In this step, if it is determined according to step (4) that the user is a high-frequency user recently, the difference from step (3) is that the information of the target object is pushed to the user by adopting the push strategy through the integration of the scenario mode. The push strategy refers to a strategy of temporarily storing the information without pushing it, waiting for external factors to trigger the push, that is, waiting for the opportunity, when the user visits a page associated with the target object, the information of the target object is pushed to the user by adopting the same scenario mode as in step (3), which improves the user's reading probability of the target object information and the user's usage experience.

[0099] (6) If the user is not a high-frequency user recently, the target object's information is pushed to the user through the general message mode.

[0100] In this step, if the user is determined to be a low-frequency user according to step (4), the information of the target object is pushed to the user through the general message mode. Since the information pushed to the user in the general message mode can only be seen when the user enters the message center in the logged-in state, that is, the user can only see it when he wants to see it, so it will not cause information interference to the user, which is conducive to improving the user's experience.

[0101] In this embodiment, by selecting the appropriate push mode and the appropriate time to push the target object information to the user according to the current scenario, not only the pertinence and accuracy of the push are further improved, but also the personalized needs of the user are met, and the user's favorability and usage satisfaction are improved.

[0102] In this embodiment, by obtaining the user's stable portrait feature set and recent portrait feature set, it is determined whether the recent portrait feature set has undergone feature changes based on the stable portrait feature set. When it is determined that the recent portrait feature set has undergone feature changes, the target object to be recommended to the user is determined based on the target portrait features that caused the feature changes. According to the current scenario, the information of the target object is pushed to the user using a preset push mode. The recommended object is determined based on the mutation features in the user's recent portrait feature set, and based on the current scenario, an appropriate push mode is selected to push the target object information. This not only improves the initiative, pertinence and accuracy of information push and meets the user's personalized needs, but also improves the user's favorability and usage satisfaction, and enhances the user's usage experience.

[0103] Embodiment 2

[0104] Figure 4 This is a flow chart of the information push method provided in the second embodiment of the present application, based on the above-mentioned first embodiment, such as Figure 4 As shown, in this embodiment, before the information of the target object is pushed to the user in a preset push mode according to the current scenario, the method of this embodiment further includes:

[0105] S201: Determine whether a target object has been recommended to a user recently.

[0106] In this step, after determining the target object, it is determined whether the target object has been recommended to the user recently (such as today), thereby controlling the recommendation frequency of the same target object and further improving the user's favorability and usage satisfaction.

[0107] Optionally, in this step, by obtaining records of products or information recently recommended to the user, using the name of the target object as a keyword, a keyword query method can be used to determine whether the target object is recently recommended to the user.

[0108] Optionally, in this embodiment, an anti-duplicate filter may be provided. Before the target object's information is pushed to the user, the target object is identified through the anti-duplicate filter to avoid frequent recommendations of the same target object.

[0109] It is understandable that if it is determined through this step that this target object has not been recommended to the user recently, the subsequent steps will be executed. If it is determined through this step that this target object has been recommended to the user recently, the subsequent steps will not be executed and this operation process ends.

[0110] S202: If the target object has not been recommended to the user recently, construct push information based on the target object.

[0111] In this step, if it is determined according to S201 that the target object has not been recommended to the user recently, push information is constructed based on the target object, such as constructing push information including the target object according to a preset information format. Optionally, the push information includes at least one of the name of the target object, the reason for recommending the target object, a brief introduction of the target object, or the benefits of using the target object.

[0112] S203: Perform privacy filtering on the pushed information to obtain information of the target object.

[0113] After S202, in order to improve the security of the pushed information and protect the privacy of the user, in this step, the pushed information is further subjected to privacy filtering to obtain the information of the target object.

[0114] In a possible implementation, this step can be performed by determining whether the push information contains sensitive information. If the push information contains sensitive information, determining whether the user has authorized the push of sensitive information. If the user has authorized the push of sensitive information, the push information is used as the information of the target object. If the user has not authorized the push of sensitive information, the sensitive information is processed to obtain the information of the target object.

[0115] Optionally, in this step, the text features of sensitive information can be defined in advance, such as whether it contains numbers and the length of the numbers contained, etc., to check the content of the pushed information to determine whether it contains sensitive information. Similarly, in this step, a sensitive information filter can also be set to determine whether the pushed information contains sensitive information through the sensitive information filter, thereby improving the accuracy and efficiency of the screening.

[0116] Optionally, the processing performed on the sensitive information may include deleting the sensitive information, blurring the sensitive information, or changing the expression of the sensitive information.

[0117] It is understandable that whether the user has authorized the push of sensitive information can be determined based on the permissions of the target application activated by the user.

[0118] In this embodiment, by determining whether the target object has been recommended to the user recently, if the target object has not been recommended to the user recently, push information is constructed according to the target object, and the push information is privacy filtered to obtain the information of the target object. This not only reduces the probability of frequently recommending the same target object to the same user, but also protects the user's privacy to a certain extent and improves information security, thereby further improving the user's experience.

[0119] Embodiment 3

[0120] Figure 5 This is a schematic diagram of the structure of the information push device provided in Example 3 of the present application, such as Figure 5 As shown, the information push device 10 in this embodiment includes:

[0121] An acquisition module 11 , a processing module 12 and a push module 13 .

[0122] An acquisition module 11 is used to acquire a stable portrait feature set and a recent portrait feature set of a user;

[0123] The processing module 12 is used to determine whether the recent portrait feature set has undergone feature change based on the stable portrait feature set; when it is determined that the recent portrait feature set has undergone feature change, determine the target object recommended to the user based on the target portrait feature causing the feature change;

[0124] The push module 13 is used to push the information of the target object to the user in a preset push mode according to the current scenario.

[0125] Optionally, the processing module 12 is specifically used for:

[0126] Determining whether there is a difference between the recent portrait feature set and the stable portrait feature set;

[0127] If there is a difference between the recent portrait feature set and the stable portrait feature set, determining whether the difference is a level difference;

[0128] If the difference is a level difference, it is determined that a feature change has occurred in the recent portrait feature set.

[0129] Optionally, the processing module 12 is specifically used for:

[0130] Extracting the target image feature causing the feature change from the recent image feature set;

[0131] The target image feature is matched with the candidate object, and the candidate object that successfully matches the target image feature is determined as the target object.

[0132] Optionally, the push module 13 is specifically used for:

[0133] Determining whether the user is online;

[0134] If the user is online, determining whether there is an association relationship between the target object and the page currently visited by the user;

[0135] If there is an association relationship between the target object and the user's currently visited page, the information of the target object is immediately pushed to the user by integrating into the scenario mode.

[0136] Optionally, the push module 13 is further used for:

[0137] If the user is not online or there is no association relationship between the target object and the page currently visited by the user, determining whether the user is a high-frequency user recently;

[0138] If the user is a high-frequency user recently, the push strategy is adopted to push the information of the target object to the user by integrating the scenario mode.

[0139] Optionally, the push module 13 is further used for:

[0140] If the user is not a high-frequency user recently, the information of the target object is pushed to the user through a general message mode.

[0141] Optionally, the acquisition module 11 is specifically used for:

[0142] Obtaining historical behavior data and recent behavior data of the user;

[0143] Analyze the historical behavior data according to data mining technology to obtain a stable profile feature set of the user;

[0144] The recent behavior data is analyzed according to data mining technology to obtain a recent portrait feature set of the user.

[0145] Optionally, the processing module 12 is further used for:

[0146] Determining whether the target object has been recommended to the user recently;

[0147] If the target object has not been recommended to the user recently, construct push information according to the target object;

[0148] The push information is privacy filtered to obtain the information of the target object.

[0149] Optionally, the processing module 12 is specifically used for:

[0150] Determining whether the push information contains sensitive information;

[0151] If the push information contains sensitive information, determining whether the user has authorized the push of sensitive information;

[0152] If the user has authorized the push of sensitive information, the push information is used as the information of the target object;

[0153] If the user does not authorize the push of sensitive information, the sensitive information is processed to obtain the information of the target object.

[0154] The information push device provided in this embodiment can execute the information push method provided in the above method embodiment, and has the corresponding functional modules and beneficial effects of the execution method. The implementation principle and technical effects of this embodiment are similar to those of the above method embodiment, and will not be repeated here.

[0155] Embodiment 4

[0156] Figure 6 A schematic diagram of the structure of a server provided in Embodiment 4 of the present application, such as Figure 6 As shown, the server 20 includes a memory 21, a processor 22, and a computer program stored in the memory and executable on the processor; the number of the processor 22 in the server 20 may be one or more. Figure 6 A processor 22 is taken as an example; the processor 22 and the memory 21 in the server 20 may be connected via a bus or other means. Figure 6 The example of connecting through bus is taken in the following.

[0157] The memory 21, as a computer-readable storage medium, can be used to store software programs, computer executable programs and modules, such as program instructions / modules corresponding to the acquisition module 11, the processing module 12 and the push module 13 in the embodiment of the present application. The processor 22 runs the software programs, instructions and modules stored in the memory 21, thereby implementing various functional applications and data processing of the server, that is, realizing the above-mentioned information push method.

[0158] The memory 21 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and at least one application required for a function; the data storage area may store data created according to the use of the terminal, etc. In addition, the memory 21 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some instances, the memory 21 may further include a memory remotely arranged relative to the processor 22, and these remote memories may be connected to the server via a grid. Examples of the above-mentioned grid include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0159] Embodiment 5

[0160] Embodiment 5 of the present application further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a computer processor, the computer program is used to execute an information push method, the method comprising:

[0161] Obtain the user's stable portrait feature set and recent portrait feature set;

[0162] Determining whether a feature change occurs in the recent portrait feature set according to the stable portrait feature set;

[0163] When it is determined that the recent portrait feature set has undergone a feature change, determining a target object to be recommended to the user based on the target portrait feature causing the feature change;

[0164] According to the current scenario, the information of the target object is pushed to the user using a preset push mode.

[0165] Of course, the computer-readable storage medium provided in the embodiment of the present application, and its computer program is not limited to the method operations described above, but can also execute related operations in the information push method provided in any embodiment of the present application.

[0166] Through the above description of the implementation method, the technicians in the relevant field can clearly understand that the present application can be implemented with the help of software and necessary general hardware, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application can be essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, server, or grid device, etc.) to execute the methods described in each embodiment of the present application.

[0167] It is worth noting that in the embodiment of the above-mentioned information push device, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application.

[0168] Note that the above are only preferred embodiments of the present application and the technical principles used. Those skilled in the art will understand that the present application is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present application. Therefore, although the present application is described in more detail through the above embodiments, the present application is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present application, and the scope of the present application is determined by the scope of the appended claims.

Claims

1. An information push method, characterized in that: include: Obtain the user's stable portrait feature set and recent portrait feature set; Determining whether a feature change occurs in the recent portrait feature set according to the stable portrait feature set; When it is determined that the recent portrait feature set has undergone a feature change, determining a target object to be recommended to the user based on the target portrait feature causing the feature change; According to the current scenario, the information of the target object is pushed to the user using a preset push mode; Wherein, determining whether the recent portrait feature set undergoes feature change according to the stable portrait feature set includes: Determining whether there is a difference between the recent portrait feature set and the stable portrait feature set; If there is a difference between the recent portrait feature set and the stable portrait feature set, determining whether the difference is a level difference; If the difference is a level difference, it is determined that a feature change has occurred in the recent portrait feature set; Wherein, determining whether the difference is a level difference includes: Pre-setting the correspondence between the image features and the levels, respectively determining the level of the recent image features and the level of the stable image features in each group of recent image features and stable image features that have differences, and determining whether the levels of the two are the same, if not the same, determining that the difference is a level difference; Wherein, determining the target object recommended to the user according to the target portrait feature causing the feature change includes: Extracting the target image feature causing the feature change from the recent image feature set; Matching the target image feature with the candidate object, and determining the candidate object that successfully matches the target image feature as the target object; The step of matching the target portrait feature with a candidate object and determining the candidate object that successfully matches the target portrait feature as the target object includes: Pre-setting a corresponding user feature portrait for each of the candidate objects, matching the target portrait features with the user feature portraits corresponding to each of the candidate objects, and determining the number of the target portrait features included in the user feature portraits of each of the candidate objects; The determined number is compared with a preset number threshold, and the candidate object including the number of the target portrait features greater than the number threshold is determined as the target object.

2. The method according to claim 1, characterized in that The method of pushing the information of the target object to the user in a preset push mode according to the current scenario includes: Determining whether the user is online; If the user is online, determining whether there is an association relationship between the target object and the page currently visited by the user; If there is an association relationship between the target object and the user's currently visited page, the information of the target object is immediately pushed to the user by integrating into the scenario mode.

3. The method according to claim 2, characterized in that The method further comprises: If the user is not online or there is no association relationship between the target object and the page currently visited by the user, determining whether the user is a high-frequency user recently; If the user is a high-frequency user recently, the push strategy is adopted to push the information of the target object to the user by integrating the scenario mode.

4. The method according to claim 3, characterized in that The method further comprises: If the user is not a high-frequency user recently, the information of the target object is pushed to the user through a general message mode.

5. The method according to claim 1, characterized in that: The step of obtaining a stable profile feature set and a recent profile feature set of a user includes: Obtaining historical behavior data and recent behavior data of the user; Analyze the historical behavior data according to data mining technology to obtain a stable profile feature set of the user; The recent behavior data is analyzed according to data mining technology to obtain a recent portrait feature set of the user.

6. The method according to any one of claims 1 to 5, characterized in that: Before pushing the information of the target object to the user in a preset push mode according to the current scenario, the method further includes: Determining whether the target object has been recommended to the user recently; If the target object has not been recommended to the user recently, construct push information according to the target object; The push information is privacy filtered to obtain the information of the target object.

7. The method according to claim 6, characterized in that The performing privacy filtering on the push information to obtain the information of the target object includes: Determining whether the push information contains sensitive information; If the push information contains sensitive information, determining whether the user has authorized the push of sensitive information; If the user has authorized the push of sensitive information, the push information is used as the information of the target object; If the user does not authorize the push of sensitive information, the sensitive information is processed to obtain the information of the target object.

8. An information push device, characterized in that: include: An acquisition module is used to obtain a user's stable portrait feature set and a recent portrait feature set; a processing module, configured to determine whether a feature change occurs in the recent portrait feature set according to the stable portrait feature set; when it is determined that a feature change occurs in the recent portrait feature set, determine a target object to be recommended to the user according to the target portrait feature causing the feature change; A push module, used to push the information of the target object to the user in a preset push mode according to the current scenario; The processing module is specifically used to determine whether there is a difference between the recent portrait feature set and the stable portrait feature set; if there is a difference between the recent portrait feature set and the stable portrait feature set, determine whether the difference is a level difference; if the difference is a level difference, determine that a feature change has occurred in the recent portrait feature set; The processing module is specifically used to pre-set the correspondence between the image features and the levels, respectively determine the level of the recent image features and the level of the stable image features in each group of recent image features and stable image features that have differences, and determine whether the levels of the two are the same. If they are not the same, determine that the difference is a level difference; The processing module is specifically used to extract the target image feature that causes the feature change from the recent image feature set; match the target image feature with the candidate object, and determine the candidate object that successfully matches the target image feature as the target object; Among them, the processing module is specifically used to pre-set a corresponding user feature portrait for each of the candidate objects, match the target portrait features with the user feature portrait corresponding to each of the candidate objects, and determine the number of target portrait features included in the user feature portrait of each of the candidate objects; compare the determined number with a pre-set number threshold, and determine the candidate object whose number of target portrait features is greater than the number threshold as the target object.

9. A server comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the information push method as described in any one of claims 1-7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the information push method as described in any one of claims 1 to 7 is implemented.

Citation Information

Patent Citations

  • Voice information pushing method and device, electronic equipment and storage medium

    CN109347986A