Personalized Customized Push System and Method for User Portrait Technology

Abstract

The present invention discloses a personalized customized push system and method for user portrait technology, which relates to the field of personalized push technology. The terminal obtains the user portrait through the authorization of the target user, uploads the non-private information in the target user portrait to the server, and the server generates a general recommendation result for the target user; the server sends the general recommendation result to the terminal, and the terminal obtains a special recommendation result based on the general recommendation result and the target user portrait, and the terminal obtains a personalized customized push method according to the general recommendation result and the special recommendation result; the transmission of the user portrait containing privacy information is not involved, reducing the risk of leakage of user portrait information during the transmission process; the transmission of the personalized customized push result is not involved, reducing the risk of exposing user privacy information in the personalized customized push content; the privacy information of the user is retained in the terminal device, improving the security of user information.

Description

Technical Field

[0001] The present invention relates to the technical field of personalized push, and specifically to a personalized customization push system and method for user portrait technology. Background Art

[0002] User portrait is a commonly used concept in fields such as data analysis, marketing, and product design. By collecting and analyzing various information such as users' characteristics, behaviors, and preferences, a virtual and representative user model is constructed to better understand user needs, optimize products and services, formulate marketing strategies, etc.; by constructing user portraits, enterprises can better understand user needs, preferences, and behaviors, so as to achieve precise advertising placement, personalized recommendations, etc.; user portraits rely on a large amount of user data. During the data transmission process, if this data is improperly collected, stored, or shared, it may lead to the leakage of user privacy; in addition, there are also security and privacy issues with the personalized recommendation content provided to users based on user portraits; therefore, how to apply user portrait technology for personalized customization push while protecting user privacy and security has become an urgent problem to be solved. Summary of the Invention

[0003] The purpose of the present invention is to provide a personalized customization push system and method for user portrait technology to solve the problems raised in the prior art.

[0004] To achieve the above purpose, the present invention provides the following technical solution: A personalized customization push system for user portrait technology, including a terminal and a server; the terminal is connected to the server, obtains a target user portrait through authorization, and transmits the non-privacy information in the target user portrait to the server; the server determines the group information of the target user based on the non-privacy information in the target user portrait, determines the general recommendation result of the target user based on the group information of the target user and sends it to the terminal; the terminal obtains the special recommendation result of the target user based on the general recommendation result of the target user and the target user portrait, and obtains personalized customization push based on the special recommendation result and the general recommendation result of the target user.

[0005] Specifically, the terminal further includes an information collection module, a first communication module, a first data storage module, a special recommendation module, a personalized customization push module, and a deviation analysis module; the information collection module is used to obtain a target user profile; the first communication module is used to establish a connection with the server; the first data storage module is used to store the user profile of the target user; the special recommendation module obtains the special recommendation result of the target user based on the general recommendation result of the target user and the target user profile; the personalized customization push module obtains the personalized customization push based on the special recommendation result and the general recommendation result of the target user; the deviation analysis module is used to determine the deviation between the personalized customization push and the general recommendation result. The server further includes a second communication module, a second data storage module, a group analysis module, a general recommendation module, a group update module, and a model generation module; the second communication module is used to realize data interaction between the server and the terminal; the second data storage module is used to store historical user profiles; the group analysis module determines the group information of the target user based on the necessary features of the target user; the general recommendation module obtains the general recommendation result of the target user based on the group information of the target user; the group update module is used to determine the necessary features of the target user and update the necessary features of the target user according to the deviation between the personalized customization push and the general recommendation result; the model generation module is used to train the first recommendation model and the second recommendation model; the first recommendation model obtains a recommendation result based on the user profile, and the second recommendation model corrects the recommendation result of the first recommendation model through additional privacy information on the basis of the recommendation result of the first recommendation model.

[0006] Specifically, the server trains the second recommendation model on the basis of the first recommendation model, uses the historical user profile containing privacy information as the advanced profile, and the historical user profile without privacy information as the basic profile. The first recommendation model is used to recommend the basic profile and the advanced profile to obtain the basic recommendation result and the advanced recommendation result. According to the deviation between the basic recommendation result and the advanced recommendation result, the influence of privacy information on the recommendation is obtained; the server uses the basic recommendation result and the influence of privacy information on the recommendation as the input, and the advanced recommendation result as the output for training to obtain the second recommendation model; the server sends the general recommendation result and the second recommendation model to the terminal, and the terminal uses the privacy information in the locally stored user profile to obtain the special recommendation result by using the second recommendation model and the general recommendation result; the personalized customization push is obtained according to the general recommendation result and the special recommendation result.

[0007] To achieve the above object, the present invention provides the following technical solution: A personalized customization push method for user profile technology, including the following steps:

[0008] The terminal obtains the user portrait through the authorization of the target user, uploads the non-privacy information in the target user portrait to the server, and the server generates the general recommendation result of the target user;

[0009] The server sends the general recommendation result to the terminal. The terminal obtains the special recommendation result based on the general recommendation result and the target user portrait, and the terminal obtains the personalized customized push method according to the general recommendation result and the special recommendation result.

[0010] Specifically, the server generating the general recommendation result of the user further includes the following steps:

[0011] The terminal obtains the user portrait through authorization. The server obtains the non-privacy information of the target user from the terminal, and determines the necessary features from the non-privacy information of the target user based on the first recommendation model and the historical user portrait in the server;

[0012] Let x i represent the i-th feature in the non-privacy information of the target user. The server screens out the historical user portraits in which the difference only exists in the feature x i from the historical user portraits, and divides the screened historical user portraits into the first type of user portrait and the second type of user portrait. The first type of user portrait includes the feature x i , and the second type of user portrait does not include the feature x i ; The server uses the first recommendation model to recommend the first type of user portrait and the second type of user portrait respectively to obtain the first type of recommendation result and the second type of recommendation result, calculates the root mean square error of the first type of recommendation result and the second type of recommendation result. If the root mean square error is greater than the error threshold, the i-th feature in the non-privacy information of the target user is used as a necessary feature; if the error is not greater than the error threshold, the i-th feature in the non-privacy information of the target user is used as an optional feature;

[0013] Among the historical user portraits, there are both user portraits with privacy information and user portraits without privacy information. The privacy information in the historical user portraits can be extended by the server on the basis of the historical user portraits without privacy information to generate user portraits with privacy information; or it can be uploaded by the user from the terminal under the condition of historical user authorization; with the user's knowledge and consent, the user can upload the user portrait with privacy information to the server to obtain better personalized push results;

[0014] Necessary features have a greater impact on the recommendation results, while optional features have a lower impact on the recommendation results. The group information of the user is determined through the necessary features. The first recommendation model obtains the recommendation results based on the user profile. The second recommendation model corrects the recommendation results of the first recommendation model through additional privacy information on the basis of the first recommendation model. The first recommendation model may also include privacy information, but the target user does not need to upload their own privacy information to the server. It does not involve the transmission of the target user's privacy information and reduces the risk of data leakage.

[0015] The server recommends the stored historical user profiles through the first recommendation model to obtain the recommendation results. The input of the first recommendation model is the features in the user profile, and the output is the recommendation scores for all services to be recommended. The recommendation scores for all services to be recommended are used as the input vector, and clustering is performed through the DBSCAN algorithm to obtain clustering clusters. There are several historical user profiles in each clustering cluster.

[0016] Obtain all the necessary features in the non-privacy information of the target user, and determine the clustering cluster that includes all the necessary features in the non-privacy information of the target user. As long as there is a historical user profile in the clustering cluster that includes all the necessary features in the non-privacy information of the target user, the clustering cluster also includes all the necessary features in the non-privacy information of the target user. According to the determined clustering cluster, obtain the group information that the target user conforms to.

[0017] Specifically, the steps for the server to generate the general recommendation results for the user also include the following:

[0018] The server recommends each group that the target user conforms to according to the group information of the target user to obtain the recommendation result T of the group j , where T j represents the recommendation result of the jth group; sum up the recommendation results of all groups that the target user conforms to to obtain the general recommendation result T of the target user, T = ∑(W j × T j ), where W j is the weight of the jth group that the target user conforms to and is calculated through the following formula, , where n j represents the number of historical user profiles that include all the necessary features in the non-privacy information of the target user in the jth group, and N j represents the number of historical user profiles in the jth group.

[0019] Specifically, the steps for obtaining the personalized customized push method also include the following:

[0020] The server trains a second recommendation model based on the first recommendation model. It uses the historical user portrait containing privacy information as the advanced portrait and the historical user portrait without privacy information as the basic portrait. The first recommendation model is used to make recommendations for the basic portrait and the advanced portrait, obtaining the basic recommendation result and the advanced recommendation result. Based on the deviation between the basic recommendation result and the advanced recommendation result, the impact of privacy information on the recommendation is obtained. The server uses the basic recommendation result and the impact of privacy information on the recommendation as inputs and the advanced recommendation result as the output for training to obtain the second recommendation model.

[0021] The impact of privacy information on the recommendation can be determined by controlling variables, that is, controlling the advanced portrait to contain all non-privacy features of the basic portrait, with the difference only in privacy features. The impact of privacy information on the recommendation is obtained based on the recommendation result after the presence of privacy features.

[0022] Specifically, the method for obtaining personalized customized push further includes the following steps:

[0023] The server sends the general recommendation result and the second recommendation model to the terminal. The terminal, based on the privacy information in the locally stored user portrait, uses the second recommendation model and the general recommendation result to obtain the special recommendation result. The personalized customized push G is obtained based on the general recommendation result and the special recommendation result, where G = U×T + V×E. Here, U and V are the weights of the general recommendation result and the special recommendation result respectively, and E is the special recommendation result of the target user.

[0024] Specifically, the method for obtaining personalized customized push further includes the following steps:

[0025] According to the group information of the target user, the features included in the group information but not in the target user portrait and the desirable features of the target user are obtained as identification features, and the impact of the identification features on the recommendation is determined. The matching degree between the impact of each identification feature on the recommendation and the deviation is calculated to obtain the identification feature with the highest matching degree. If the matching degree is greater than the threshold, the identification feature with the highest matching degree is added to the main features of the target user. Based on the main features of the target user, the group information of the target user is updated. The server updates the general recommendation result of the target user according to the updated group information of the target user and sends it to the terminal.

[0026] The impact of the identification features on the recommendation can be determined by referring to the impact of privacy information on the recommendation.

[0027] Compared with the prior art, the beneficial effects of the present invention are as follows: It does not involve the transmission of user portraits containing privacy information, reducing the risk of leakage of user portrait information during the transmission process; it does not involve the transmission of personalized customized push results, reducing the risk of exposing user privacy information in the personalized customized push content; it retains the user's privacy information in the terminal device, improving the security of user information. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a schematic structural diagram of a personalized customized push system for the user portrait technology of the present invention;

[0029] Figure 2 It is a flowchart of a personalized customized push method for the user portrait technology of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0030] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0031] Embodiment: As Figure 1 shown, the present invention provides a technical solution, a personalized customized push system for user portrait technology, including a terminal and a server; the terminal is connected to the server, obtains a target user portrait through authorization, and transmits non-privacy information in the target user portrait to the server; the server determines group information of the target user based on the non-privacy information in the target user portrait, determines a general recommendation result of the target user based on the group information of the target user and sends it to the terminal; the terminal obtains a special recommendation result of the target user based on the general recommendation result of the target user and the target user portrait, and obtains a personalized customized push based on the special recommendation result and the general recommendation result of the target user. The terminal includes but is not limited to mobile phones, tablets, computers, etc., and obtains user tags during the use by the user.

[0032] Specifically, the terminal further includes an information collection module, a first communication module, a first data storage module, a special recommendation module, a personalized customization push module, and a deviation analysis module; the information collection module is used to obtain a target user profile; the first communication module is used to establish a connection with the server; the first data storage module is used to store the user profile of the target user; the special recommendation module obtains the special recommendation result of the target user based on the general recommendation result of the target user and the target user profile; the personalized customization push module obtains the personalized customization push based on the special recommendation result and the general recommendation result of the target user; the deviation analysis module is used to determine the deviation between the personalized customization push and the general recommendation result. The server further includes a second communication module, a second data storage module, a group analysis module, a general recommendation module, a group update module, and a model generation module; the second communication module is used to implement data interaction between the server and the terminal; the second data storage module is used to store historical user profiles; the group analysis module determines the group information of the target user based on the necessary features of the target user; the general recommendation module obtains the general recommendation result of the target user based on the group information of the target user; the group update module is used to determine the necessary features of the target user and update the necessary features of the target user according to the deviation between the personalized customization push and the general recommendation result; the model generation module is used to train the first recommendation model and the second recommendation model; the first recommendation model obtains a recommendation result based on the user profile, and the second recommendation model corrects the recommendation result of the first recommendation model through additional privacy information on the basis of the recommendation result of the first recommendation model.

[0033] Specifically, the server trains the second recommendation model on the basis of the first recommendation model, uses the historical user profile containing privacy information as the advanced profile, and the historical user profile without privacy information as the basic profile. The first recommendation model is used to recommend the basic profile and the advanced profile to obtain the basic recommendation result and the advanced recommendation result. According to the deviation between the basic recommendation result and the advanced recommendation result, the influence of privacy information on the recommendation is obtained; the server uses the basic recommendation result and the influence of privacy information on the recommendation as the input, and the advanced recommendation result as the output for training to obtain the second recommendation model; the server sends the general recommendation result and the second recommendation model to the terminal, and the terminal uses the privacy information in the locally stored user profile, and uses the second recommendation model and the general recommendation result to obtain the special recommendation result; the personalized customization push is obtained according to the general recommendation result and the special recommendation result.

[0034] Embodiment: As Figure 2 shown, the present invention provides a technical solution, a personalized customization push method for user profile technology, including the following steps:

[0035] The terminal obtains the user portrait through the authorization of the target user, uploads the non-private information in the target user portrait to the server, and the server generates the general recommendation result of the target user:

[0036] Let x i represent the i-th feature in the non-private information of the target user. The server filters out the historical user portraits from the historical user portraits where the difference only exists in the feature x i of the historical user portraits, and divides the filtered historical user portraits into the first type of user portraits and the second type of user portraits, where the first type of user portraits includes the feature x i , and the second type of user portraits does not include the feature x i ; The server uses the first recommendation model to recommend the first type of user portraits and the second type of user portraits respectively, obtains the first type of recommendation results and the second type of recommendation results, calculates the root mean square error of the first type of recommendation results and the second type of recommendation results. If the root mean square error is greater than the error threshold, the i-th feature in the non-private information of the target user is used as a necessary feature; if the error is not greater than the error threshold, the i-th feature in the non-private information of the target user is used as an optional feature;

[0037] The output results of the first recommendation model and the second recommendation model are the recommendation scores for the recommended services. For example, if the enterprise is involved in a total of 5 services from A1 to A5, the first recommendation model and the second recommendation model will obtain the recommendation scores for these 5 services, and obtain the weights of each service according to the recommendation scores, and display the services of the enterprise to the user according to the weights.

[0038] The server recommends the stored historical user portraits through the first recommendation model to obtain the recommendation results; the input of the first recommendation model is the features in the user portrait, and the output is the recommendation scores for all services to be recommended; the recommendation scores for all services to be recommended are used as the input vector, and clustering is performed through the DBSCAN algorithm to obtain the clustering clusters; several historical user portraits in each clustering cluster;

[0039] The specific steps of clustering are as follows:

[0040] Obtain the recommendation results of the first recommendation model for the historical user portraits, that is, the recommendation scores for 5 services, and perform DBSCAN clustering with the recommendation scores as the input vector. First, set the neighborhood radius and the minimum number of points; the neighborhood radius and the minimum number of points can be tested through different combinations;

[0041] Step 1, select an input vector as the starting point;

[0042] Step 2, with the selected input vector as the center, find other input vectors within the neighborhood radius;

[0043] Step 3, if the number of other input vectors found within the neighborhood radius is not less than the minimum number of points, mark the selected input vector as a core point and form a clustering cluster with this core point; if the number of other input vectors found within the neighborhood radius is less than the minimum number of points, mark this input vector as a noise point;

[0044] Step 4, for the core points, add all the input vectors within the neighborhood radius centered on the core points to the clustering cluster formed by the core points; for the other input vectors added to the clustering cluster of the core points, determine whether they are core points according to Step 3. If they are core points, continue to expand the clustering cluster; if they are not core points, do not expand;

[0045] Repeat Steps 1 to 4 until all input vectors have been visited.

[0046] Obtain all the necessary features in the non-private information of the target user, determine the clustering cluster including all the necessary features in the non-private information of the target user. As long as there is a historical user portrait in the clustering cluster that includes all the necessary features in the non-private information of the target user, the clustering cluster also includes all the necessary features in the non-private information of the target user; according to the determined clustering cluster, obtain the group information that the target user conforms to.

[0047] The server makes recommendations for each group that the target user conforms to according to the group information of the target user, and obtains the recommendation result T of the group j , where T j represents the recommendation result of the jth group; sum up the recommendation results of all the groups that the target user conforms to to obtain the general recommendation result T of the target user, T = ∑(W j ×T j ), where W j is the weight of the jth group that the target user conforms to and is calculated by the following formula , where n j represents the number of historical user portraits that include all the necessary features in the non-private information of the target user in the jth group, and N j represents the number of historical user portraits in the jth group.

[0048] The server sends the general recommendation result to the terminal. The terminal obtains the special recommendation result based on the general recommendation result and the target user portrait, and the terminal obtains the personalized customized push method according to the general recommendation result and the special recommendation result.

[0049] The server trains a second recommendation model based on the first recommendation model. It uses the historical user profile containing privacy information as the advanced profile and the historical user profile without privacy information as the basic profile. Through the first recommendation model, it makes recommendations for the basic profile and the advanced profile, obtaining the basic recommendation result and the advanced recommendation result. Based on the deviation between the basic recommendation result and the advanced recommendation result, it gets the impact of privacy information on the recommendation. The server uses the basic recommendation result and the impact of privacy information on the recommendation as the input and the advanced recommendation result as the output for training to obtain the second recommendation model.

[0050] The server sends the general recommendation result and the second recommendation model to the terminal. The terminal, based on the privacy information in the locally stored user profile, uses the second recommendation model and the general recommendation result to obtain the special recommendation result. According to the general recommendation result and the special recommendation result, it gets the personalized customized push G, where G = U×T + V×E. Here, U and V are the weights of the general recommendation result and the special recommendation result respectively, and E is the special recommendation result of the target user.

[0051] The terminal gets the deviation based on the personalized customized push G of the target user and the general recommendation result T, and feeds the deviation back to the server.

[0052] According to the group information of the target user, it obtains the features included in the group information but not in the target user profile and the desirable features of the target user as the identification features, determines the impact of the identification features on the recommendation. It calculates the matching degree between the impact of each identification feature on the recommendation and the deviation, and gets the identification feature with the highest matching degree. If the matching degree is greater than the threshold, it adds the identification feature with the highest matching degree to the main features of the target user, and updates the group information of the target user according to the main features of the target user. The server updates the general recommendation result of the target user according to the updated group information of the target user and sends it to the terminal.

[0053] The deviation between the personalized customized push G and the general recommendation result T is the deviation between the recommendation scores of 5 services, which can be reflected by the root mean square error. According to the impact of each identification feature on the recommendation, for example, when the identification feature is B1, it will cause the recommendation score of service A1 to increase, and the necessary feature of the user does not include the identification feature B1 yet. When the recommendation score of service A1 increases and makes the deviation between the personalized customized push G and the general recommendation result T decrease, it can be obtained that the identification feature B1 contributes to reducing the deviation. According to the contribution, the matching degree between B1 and the deviation is obtained. When the matching degree is high enough, B1 is added to the necessary features.

[0054] It is obvious to those skilled in the art that the present invention is not limited to the details of the above-described exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention. Therefore, in all respects, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes that fall within the meaning and scope of the equivalent elements of the claims are intended to be embraced within the present invention. Any reference signs in the claims should not be construed as limiting the claimed invention.

Claims

1. A personalized customized push method based on user portrait technology, characterized in that: The following steps are involved: The terminal obtains the user portrait through authorization of the target user, uploads the non-privacy information in the target user portrait to the server, and the server generates a universal recommendation result for the target user; The server generates a universal recommendation result for the target user and further comprises the following steps: The terminal obtains the target user portrait through authorization, the server obtains the target user's non-private information from the terminal, and determines necessary features from the target user's non-private information based on the first recommendation model and the historical user portrait in the server; Let x i represents the i-th feature in the target user’s non-private information. The server selects the features whose differences only exist in feature x from the historical user profiles. i The historical user portraits are divided into the first category of user portraits and the second category of user portraits, where the first category of user portraits includes the feature x i , the second type of user portrait does not include feature x i The server uses the first recommendation model to recommend the first type of user portrait and the second type of user portrait respectively, obtains the first type of recommendation results and the second type of recommendation results, calculates the root mean square error between the first type of recommendation results and the second type of recommendation results, and if the root mean square error is greater than the error threshold, the i-th feature in the target user's non-private information is taken as a necessary feature; if the error is not greater than the error threshold, the i-th feature in the target user's non-private information is taken as a required feature; The server recommends the stored historical user portraits through the first recommendation model to obtain a recommendation result; the input of the first recommendation model is the features in the user portrait, and the output is the recommendation score of all the services to be recommended; the recommendation scores of all the services to be recommended are used as input vectors, and clustering is performed through the DBSCAN algorithm to obtain clusters; a number of historical user portraits in each cluster; Obtain all necessary features in the target user's non-private information, and determine a cluster that includes all necessary features in the target user's non-private information. As long as there is a historical user profile in the cluster that includes all necessary features in the target user's non-private information, the cluster also includes all necessary features in the target user's non-private information. According to the determined cluster, obtain the group information that the target user meets; The server sends the universal recommendation results to the terminal, and the terminal obtains the specific recommendation results based on the universal recommendation results and the target user portrait. The terminal obtains a personalized push method based on the universal recommendation results and the specific recommendation results.

2. The personalized customized push method based on user portrait technology according to claim 1 is characterized in that: The server generates a universal recommendation result for the user, and further comprises the following steps: The server recommends each group that the target user matches based on the target user’s group information, and obtains the group recommendation result T j , where T j represents the recommendation result of the jth group; sum the recommendation results of all groups that the target user meets, and get the universal recommendation result T of the target user, T=∑(W j ×T j ), where W j The weight of the jth group that the target user meets is calculated using the following formula: , where n j represents the number of historical user profiles in the jth group that contain all necessary features of the target user’s non-private information, N j Represents the number of historical user portraits in the jth group.

3. The personalized customized push method based on user portrait technology according to claim 2 is characterized in that: The method for obtaining personalized push notifications also includes the following steps: The server trains a second recommendation model based on the first recommendation model, takes the historical user portrait containing privacy information as the advanced portrait, and the historical user portrait not containing privacy information as the basic portrait, and recommends the basic portrait and the advanced portrait through the first recommendation model to obtain basic recommendation results and advanced recommendation results. According to the deviation between the basic recommendation results and the advanced recommendation results, the influence of privacy information on the recommendation is obtained; the server takes the basic recommendation results and the influence of privacy information on the recommendation as input, and trains the advanced recommendation results as output to obtain the second recommendation model.

4. The personalized customized push method based on user portrait technology according to claim 3 is characterized in that: The method for obtaining personalized push notifications also includes the following steps: The server sends the universal recommendation result and the second recommendation model to the terminal. The terminal obtains the specific recommendation result based on the privacy information in the user portrait stored locally by using the second recommendation model and the universal recommendation result. The personalized push G is obtained according to the universal recommendation result and the specific recommendation result, G=U×T+V×E, where U and V are the weights of the universal recommendation result and the specific recommendation result, and E is the specific recommendation result for the target user.

5. The personalized customized push method based on user portrait technology according to claim 4 is characterized in that: The method for obtaining personalized push notifications also includes the following steps: The terminal obtains the deviation based on the personalized customized push G of the target user and the universal recommendation result T, and feeds the deviation back to the server; According to the group information of the target user, the features included in the group information but not included in the target user portrait and the required features of the target user are obtained as identification features, and the influence of the identification features on the recommendation is determined; the matching degree between the influence and deviation of each identification feature on the recommendation is calculated to obtain the identification feature with the highest matching degree. If the matching degree is greater than a threshold, the identification feature with the highest matching degree is added to the main feature of the target user, and the group information of the target user is updated according to the main feature of the target user; the server updates the universal recommendation result of the target user according to the updated group information of the target user and sends it to the terminal.

6. A personalized customized push system based on user portrait technology, using a personalized customized push method based on user portrait technology as claimed in any one of claims 1 to 5, characterized in that: It includes a terminal and a server; the terminal is connected to the server, obtains a target user portrait through authorization, and transmits non-private information in the target user portrait to the server; the server determines the group information of the target user based on the non-private information in the target user portrait, determines the universal recommendation result of the target user based on the group information of the target user, and sends it to the terminal; The terminal obtains a specific recommendation result for the target user based on the general recommendation result for the target user and the target user portrait, and obtains a personalized customized push based on the specific recommendation result and the general recommendation result for the target user.

7. The personalized customized push system based on user portrait technology according to claim 6 is characterized in that: The terminal further includes an information collection module, a first communication module, a first data storage module, a particularity recommendation module, a personalized customization push module and a deviation analysis module; the information collection module is used to obtain a target user portrait; the first communication module is used to establish a connection with a server; the first data storage module is used to store a user portrait of the target user; The specific recommendation module obtains the specific recommendation result of the target user based on the universal recommendation result of the target user and the target user portrait; The personalized push module obtains personalized push based on the specific recommendation results and universal recommendation results of the target user; The deviation analysis module is used to determine the deviation between the personalized customized push and the universal recommendation results.

8. The personalized customized push system based on user portrait technology according to claim 7 is characterized in that: The server further includes a second communication module, a second data storage module, a group analysis module, a universal recommendation module, a group update module and a model generation module; the second communication module is used to realize data interaction between the server and the terminal; the second data storage module is used to store historical user portraits; the group analysis module determines the group information of the target user based on the necessary characteristics of the target user; The universal recommendation module obtains universal recommendation results for target users based on group information of target users; The group update module is used to determine the necessary characteristics of the target user and update the necessary characteristics of the target user according to the deviation between the personalized customized push and the universal recommendation results; The model generation module is used to train a first recommendation model and a second recommendation model; the first recommendation model obtains a recommendation result based on a user portrait, and the second recommendation model corrects the recommendation result of the first recommendation model based on the recommendation result of the first recommendation model by using additional privacy information.

9. The personalized customized push system based on user portrait technology according to claim 8, characterized in that: The server trains a second recommendation model based on the first recommendation model, uses the historical user portrait containing privacy information as the advanced portrait, and the historical user portrait not containing privacy information as the basic portrait, recommends the basic portrait and the advanced portrait through the first recommendation model, obtains the basic recommendation result and the advanced recommendation result, and obtains the influence of privacy information on the recommendation according to the deviation between the basic recommendation result and the advanced recommendation result; the server uses the basic recommendation result and the influence of privacy information on the recommendation as input, and uses the advanced recommendation result as output for training, to obtain the second recommendation model; The server sends the universal recommendation result and the second recommendation model to the terminal, and the terminal obtains the specific recommendation result by using the second recommendation model and the universal recommendation result based on the privacy information in the user portrait stored locally; Receive personalized push notifications based on general and specific recommendation results.

Images