User recommendation method, device, equipment and storage medium
By pulling and updating user recommendation data from preset connection recommendation pools, dynamically adjusting the recommendation pools based on user operation information and type activity, the problem of low recommendation effect in the existing technology is solved, and the user attention rate and recommendation effect are improved. .
Patent Information
- Application Number
- CN202110144196.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-02-02
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2041-02-02
AI Technical Summary
The existing user recommendation methods are mainly based on second-degree friend recommendations, and do not involve multi-dimensional calculations, resulting in low recommendation effects. If the recommended user data cannot arouse user interest, it will cause interference to the user and cannot increase the attention rate.
By pulling the connection recommendation data for recommended users from the preset connection recommendation pool, and updating the recommendation level information based on the user's operation information, including viewing, following, blocking and other operations, dynamically adjusting the user sorting and weight value in the recommendation pool in the recommendation pool , generate recommendation level information based on user type and activity.
The effect of user recommendations has been improved, and the user's connection data has been pushed in multiple dimensions has been enhanced, and the user's attention rate has been increased by 5% through experimental statistics.
Smart Images

Figure CN112800331B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of data mining, and in particular to a user recommendation method, device, equipment and storage medium. Background Art
[0002] User network recommendation is based on user data, recommending more users of interest to users, so as to guide users to pay attention, increase user attention rate, and expand the overall relationship chain data.
[0003] However, if the recommended network data does not arouse the user's interest, not only will it fail to increase the attention rate, but the recommended network data will also cause certain interference to the user. The current user recommendation method is generally based on second-degree friends and other recommendations, and does not involve multi-dimensional calculations, resulting in low recommendation effects. Summary of the invention
[0004] In view of the above problems, embodiments of the present invention are proposed to provide a user recommendation method, apparatus, device and storage medium that overcome the above problems or at least partially solve the above problems.
[0005] In order to solve the above problem, an embodiment of the present invention discloses a user recommendation method, which includes:
[0006] Pulling the recommended network data for the recommended user from a preset recommended network pool, and returning the recommended network data to the terminal; wherein the recommended network data is pulled according to the recommended degree information of the recommended user;
[0007] receiving operation information for the recommended user sent by the terminal; the operation information is generated by a feedback operation of a terminal user on the recommended user;
[0008] The recommendation degree information of the recommended user is updated according to the operation information.
[0009] Optionally, updating the recommendation level information of the recommended user according to the operation information includes:
[0010] Updating the recommended score of the recommended user according to the operation information;
[0011] The updating of the recommended score of the recommended user according to the operation information includes:
[0012] Obtain the recommended score of the recommended user before updating;
[0013] If the operation information is a viewing operation for the recommended user, the recommended score of the recommended user is updated to be the recommended score before the update minus a preset score threshold.
[0014] Optionally, it also includes:
[0015] If the operation information is attention operation information for the recommended user, deleting the recommended user from the preset network recommendation pool;
[0016] If the operation information is blocking operation information for the recommended user, the recommended user is deleted from the preset network recommendation pool.
[0017] Optionally, before extracting the network recommendation data for the recommended user from the preset network recommendation pool, the method further includes:
[0018] Acquire user information of the recommended user, and generate recommendation degree information according to the user information;
[0019] The recommended user is updated to a preset network recommendation pool according to the recommendation degree information.
[0020] Optionally, the user information of the recommended user includes user type and user activity; and generating recommendation degree information according to the user information includes:
[0021] Generate interpersonal relationship information according to the user type and the user activity, and obtain the weight value of the recommended user in a preset interpersonal recommendation pool;
[0022] The weight value and the personal connection degree information are used to generate the recommendation degree information of the recommended user.
[0023] Optionally, the user type includes at least one of the following: user attention type, user blocking type, user friend type, user friend's common attention type, user friend's friend type and user attention type; wherein, the recommended user whose user type is user attention type or user blocking type is not in the preset network recommendation pool.
[0024] Optionally, the personal connection degree information includes first personal connection degree information and second personal connection degree information; and generating the personal connection degree information according to the user type and the user activity includes:
[0025] The first personal connection degree information is generated according to the user type of the recommended user, and the second personal connection degree information is generated according to the user activity.
[0026] Optionally, the first personal connection degree information is determined by a first personal connection score; and the generating the first personal connection degree information according to the user type of the recommended user includes:
[0027] If the user type of the recommended user is a user friend type, determining a current first personal connection score according to a preset first score;
[0028] and / or, if the user type of the recommended user is a common type of interest of the user's friends, determining the current first personal connection score according to the number of the user's friends who commonly interest the recommended user and a preset second score;
[0029] and / or, if the user type of the recommended user is the friend type of the user's friend, determining the current first personal connection score according to the number of common friends and a preset third score;
[0030] And / or, if the user type of the recommended user is a focus type that the user focuses on, the current first personal connection score is determined according to the number of common focuses and a preset fourth score.
[0031] Optionally, the second personal connection degree information is determined by a second personal connection score; and generating the second personal connection degree information according to the user activity includes:
[0032] If the recommended user has posted a dynamic within the preset time period, the time of posting the dynamic and the interaction within the preset time period are obtained, and the second personal connection score is calculated according to the time of posting the dynamic and the interaction within the preset time period;
[0033] Also includes:
[0034] Filter the recommended users who have not posted any updates within a preset time period.
[0035] Optionally, obtaining a weight value of the recommended user in a preset network recommendation pool includes:
[0036] Determine whether the recommended user exists in a preset network recommendation pool;
[0037] If the recommended user exists in the preset personal connection recommendation pool, obtaining a weight value of the recommended user; the weight value is determined by the recommendation pool score of the recommended user in the preset personal connection recommendation pool;
[0038] If the recommended user does not exist in the network recommendation pool, the weight value of the recommended user is set to a preset initial threshold.
[0039] Optionally, the recommendation degree information is determined by a recommended score; and updating the recommended user to a preset network recommendation pool according to the recommendation degree information includes:
[0040] The recommended score is used to update the ranking of the recommended user in the preset personal connection recommendation pool, and the recommended user is updated or inserted into the preset personal connection recommendation pool according to the updated ranking.
[0041] The embodiment of the present invention further discloses a user recommendation device, the device comprising:
[0042] A network recommendation data extraction module is used to extract network recommendation data for the recommended user from a preset network recommendation pool and return the network recommendation data to the terminal; wherein the network recommendation data is extracted according to the recommendation degree information of the recommended user;
[0043] An operation information receiving module, used for receiving operation information sent by the terminal for the recommended user; the operation information is generated by a feedback operation of the terminal user on the recommended user;
[0044] The recommendation degree information updating module is used to update the recommendation degree information of the recommended user according to the operation information.
[0045] Optionally, the recommendation degree information is determined by a recommended score; and the recommendation degree information updating module includes:
[0046] A recommended score updating submodule, used for updating the recommended score of the recommended user according to the operation information;
[0047] Optionally, the recommended score updating submodule includes:
[0048] A recommended score acquisition unit, used to acquire the recommended score of the recommended user before updating;
[0049] The recommended score updating unit is configured to update the recommended score of the recommended user to the recommended score before the update minus a preset score threshold if the operation information is a viewing operation for the recommended user.
[0050] Optionally, the recommendation degree information updating module further includes:
[0051] A first recommended user deletion submodule is used to delete the recommended user from a preset network recommendation pool if the operation information is a follow-up operation information for the recommended user;
[0052] The second recommended user deletion submodule is used to delete the recommended user from the preset network recommendation pool if the operation information is blocking operation information for the recommended user.
[0053] Optionally, before extracting the network recommendation data for the recommended user from the preset network recommendation pool, the method further includes:
[0054] A recommendation degree information generating module, used for acquiring user information of the recommended user and generating recommendation degree information according to the user information;
[0055] The recommended user updating module is used to update the recommended user to a preset network recommendation pool according to the recommendation degree information.
[0056] Optionally, the user information of the recommended user includes user type and user activity; and the recommendation degree information generating module includes:
[0057] A personal connection degree information generation submodule, used to generate personal connection degree information according to the user type and the user activity;
[0058] A weight value acquisition submodule is used to obtain the weight value of the recommended user in a preset network recommendation pool;
[0059] The recommendation degree information generation submodule is used to generate the recommendation degree information of the recommended user by using the weight value and the personal connection degree information.
[0060] Optionally, the user type includes at least one of the following: user attention type, user blocking type, user friend type, user friend's common attention type, user friend's friend type and user attention type; wherein, the recommended user whose user type is user attention type or user blocking type is not in the preset network recommendation pool.
[0061] Optionally, the personal connection degree information includes first personal connection degree information and second personal connection degree information; and the personal connection degree information generating submodule includes:
[0062] A first personal connection degree information generating unit, configured to generate first personal connection degree information according to the user type of the recommended user;
[0063] The second personal connection degree information generating unit is used to generate the second personal connection degree information according to the user activity.
[0064] Optionally, the first personal connection degree information is determined by a first personal connection score; and the first personal connection degree information generating unit includes:
[0065] A first personal connection score determination subunit, configured to determine a current first personal connection score according to a preset first score if the user type of the recommended user is a user friend type;
[0066] The first interpersonal relationship score determination subunit is further configured to determine the current first interpersonal relationship score according to the number of user friends who commonly follow the recommended user and a preset second score if the user type of the recommended user is a common type of interest of the user's friends;
[0067] The first personal connection score determination subunit is further configured to determine the current first personal connection score according to the number of common friends and a preset third score if the user type of the recommended user is the friend type of the user's friend;
[0068] The first personal connection score determination subunit is further configured to determine the current first personal connection score according to the number of common concerns and a preset fourth score if the user type of the recommended user is a concern type that the user concerns.
[0069] Optionally, the second personal connection degree information is determined by a second personal connection score; and the second personal connection degree information generating unit includes:
[0070] A second interpersonal relationship score determination subunit is used to obtain the time of posting the dynamics and the interaction status within the preset time period if the recommended user has posted dynamics within the preset time period, and calculate the second interpersonal relationship score according to the time of posting the dynamics and the interaction status within the preset time period;
[0071] Optionally, the second personal connection degree information generating unit further includes:
[0072] The recommended user filtering subunit is used to filter the recommended users who have not posted any dynamics within a preset time period.
[0073] Optionally, the weight value acquisition submodule includes:
[0074] A recommended user determination unit, used to determine whether the recommended user exists in a preset network recommendation pool;
[0075] A weight value acquisition unit, configured to acquire a weight value of the recommended user if the recommended user exists in a preset personal connection recommendation pool; the weight value is determined by the recommendation pool score of the recommended user in the preset personal connection recommendation pool;
[0076] The weight value setting unit is used to set the weight value of the recommended user to a preset initial threshold if the recommended user does not exist in the network recommendation pool.
[0077] Optionally, the recommended user update module includes:
[0078] The recommended user updating submodule is used to update the ranking of the recommended user in the preset personal connection recommendation pool using the recommended score, and update or insert the recommended user into the preset personal connection recommendation pool according to the updated ranking.
[0079] An embodiment of the present invention further discloses a device, comprising: a processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the computer program implements any step of the user recommendation method when executed by the processor.
[0080] The embodiment of the present invention further discloses a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of any one of the user recommendation methods are implemented.
[0081] The embodiments of the present invention include the following advantages:
[0082] In the embodiment of the present invention, the server can pull the recommended data from the recommended pool and return it to the user terminal. When the user terminal detects a feedback operation for the returned recommended user, the recommended degree information of the recommended user can be updated according to the operation information sent by the user terminal to further update the recommended pool. By combining user behavior and user feedback on the basis of the existing user relationship chain of the existing user recommendation method, the user's network data is pushed from multiple dimensions to ensure the effect of user recommendation. BRIEF DESCRIPTION OF THE DRAWINGS
[0083] Figure 1 is a flowchart of steps of an embodiment of a user recommendation method of the present invention;
[0084] Figure 2 is a flowchart of another user recommendation method embodiment of the present invention;
[0085] Figure 3 is an application scenario diagram of the user recommendation method in an embodiment of the present invention;
[0086] Figure 4 It is a structural block diagram of an embodiment of a user recommendation device of the present invention. DETAILED DESCRIPTION
[0087] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.
[0088] One of the core ideas of the embodiments of the present invention is to calculate the first-degree / second-degree connections of the recommended user based on the user's friend relationship, follow-up relationship, etc., and to calculate and rank the recommended scores of the recommended users in combination with the posting situation in the user terminal and the client behavior, and to dynamically adjust the user's follow-up strategy for the recommended user in an experimental manner.
[0089] Reference Figure 1 , shows a flowchart of a user recommendation method embodiment of the present invention, which may specifically include the following steps:
[0090] Step 101, extracting the recommended network data for the recommended user from a preset network recommendation pool, and returning the recommended network data to the terminal;
[0091] Currently, many social applications can be applied to various mobile terminals such as mobile phones and tablet computers or PC clients, so that users can communicate with other users through social applications. In actual applications, when the terminal user uses the registered social application, the server can recommend more users to the terminal user to improve the user stickiness and extensiveness of the social application.
[0092] In one embodiment of the present invention, the server can pull the recommended user and return it to the terminal. In one case, the server can passively pull the recommended user and passively return the network recommendation data to the terminal. Specifically, the user terminal can pull the recommended user from the server, the user terminal can generate a recommended user acquisition request and send the generated recommended user acquisition request to the server, and the server can respond to the recommended user acquisition request after receiving the recommended user acquisition request, and pull the recommended user's network recommendation data from a preset network recommendation pool; in another case, the server can actively pull the recommended user and return the network recommendation data to the active terminal.
[0093] The pulled recommended personal data may be a preset number of user data pulled according to the recommended information, and specifically may be a limited number of recommended user data pulled in descending order of the recommended scores. It should be noted that the value of the preset number may be determined according to actual conditions, and the embodiment of the present invention does not impose any limitation on this.
[0094] Step 102, receiving operation information for the recommended user sent by the terminal; the operation information is generated by a feedback operation of the terminal user on the recommended user;
[0095] In response to the recommended user acquisition request, the server pulls the network recommendation data from the preset network recommendation pool, and after the user terminal receives the network recommendation data returned by the server, the terminal user can operate the recommended user in the returned network recommendation data so as to update the network recommendation pool according to the terminal user's operation.
[0096] Specifically, the user terminal detects the operation of the terminal user on the recommended user, and the feedback information detected by the user terminal may include operation information corresponding to the feedback operation of the recommended user, so as to update the recommendation degree information of the recommended user through the operation information.
[0097] Step 103: Update the recommendation level information of the recommended user according to the operation information.
[0098] The recommendation degree information may be used to indicate the push degree of the recommended user in the preset network recommendation pool, which may be specifically determined by the recommendation score of the recommended user.
[0099] In actual applications, the server updates the recommendation degree information of the recommended user, which can be essentially manifested as processing the recommended score of the recommended user according to the operation information. If the operation information is viewing operation information for the recommended user, indicating that the terminal user has viewed the recommended user but has not followed or blocked the recommended user, the recommended score of the recommended user before the update can be obtained, and the recommended score of the recommended user is updated to be the recommended score before the update minus the preset score threshold.
[0100] Among them, the recommended score for determining the recommendation degree information can be obtained by summarizing the personal connection score and the weight value, and the weight value can be determined by the recommendation pool score of the recommended user in the personal connection recommendation pool. The recommended score of the recommended user is essentially processed by subtracting the recommendation pool score of the recommended user from the preset score threshold.
[0101] In a specific implementation, the preset score threshold may be the initial recommendation pool score (eg, 1000) when the recommended user is not in the network recommendation pool, so as to achieve the purpose of ranking the recommended user that has been pushed to the terminal user in the network recommendation pool at the back.
[0102] In an optional embodiment, if the updated recommendation score is a negative number, it means that the terminal user is not interested in the recommended user, and the recommended user will no longer be recommended, which can be mainly manifested in deleting the recommended user from the network recommendation pool.
[0103] In an embodiment of the present invention, the operation information corresponding to the feedback operation of the recommended user may include not only the above-mentioned viewing operation information, but also attention operation information, blocking operation information, etc.
[0104] If the operation information is the follow operation information for the recommended user, it means that the terminal user has followed the recommended user, and the recommended user can be deleted from the preset network recommendation pool at this time; if the operation information is the block operation information for the recommended user, it means that the terminal user has blocked the recommended user, that is, the terminal user does not want to follow the recommended user, and the recommended user can be deleted from the preset network recommendation pool at this time.
[0105] In the embodiment of the present invention, the server can pull the recommended data from the recommended pool and return it to the user terminal. When the user terminal detects a feedback operation for the returned recommended user, the recommended degree information of the recommended user can be updated according to the operation information sent by the user terminal to further update the recommended pool. By combining user behavior and user feedback on the basis of the existing user relationship chain of the existing user recommendation method, the user's network data is pushed from multiple dimensions to ensure the effect of user recommendation.
[0106] Reference Figure 2 , shows a flowchart of another user recommendation method embodiment of the present invention, which may specifically include the following steps:
[0107] Step 201, obtaining user information of the recommended user, and generating recommendation degree information according to the user information;
[0108] In one embodiment of the present invention, the rule for pulling the personal connection recommendation data is to pull the data of the recommended user according to the recommendation degree information (i.e., the order of the recommended scores, generally from high to low). Before the server pulls the personal connection recommendation data from the preset personal connection recommendation pool, it is also necessary to determine the recommendation degree information of the recommended user, i.e., the recommended score. Specifically, it can be manifested as determining the recommended score based on the user information of the recommended user.
[0109] In one embodiment of the present invention, step 201 may include the following sub-steps:
[0110] Sub-step S11, generating personal connection information according to the user type and the user activity;
[0111] Specifically, the personal connection degree information may be generated according to the user information of the recommended user, so as to be used to determine the recommendation degree information of the recommended user.
[0112] In one embodiment of the present invention, the user information of the recommended user may include the user type and user activity of the recommended user, and the first personal connection degree information and the second personal connection degree information may be obtained according to the user type and user activity of the recommended user, respectively. The first personal connection degree information may be determined by the first personal connection score, and the second personal connection degree information may be determined by the second personal connection score.
[0113] In a specific implementation, the user type of the recommended user may include at least one of the following: user attention type, user blocking type, user friend type, common attention type of user friend, friend type of user friend, and attention type of the user attention.
[0114] Among them, the user's attention type may refer to the users that the user actively follows, the user's blocking type may refer to the users that the user has blocked, the user's friend type may refer to the friends in the user's address book, the user's friend's common attention type may refer to the users that are followed by more than two friends, the user's friend's friend type may refer to the users that have more than two common friends, and the user's attention type may refer to the users that are followed by more than two people.
[0115] It should be noted that the recommended users whose user type is the user-followed type or the user-blocked type are not in the preset network recommendation pool.
[0116] In an embodiment of the present invention, the first personal connection degree is determined according to the user type of the recommended user. The specific steps may include: if the user type of the recommended user is a user friend type, the current first personal connection score may be determined according to a preset first score; if the user type of the recommended user is a common concern type of the user's friends, the current first personal connection score may be determined according to the number of user friends who commonly concern the recommended user and a preset second score; if the user type of the recommended user is a friend type of the user's friends, the current first personal connection score may be determined according to the number of common friends and a preset third score; if the user type of the recommended user is a user friend type of the user's friends, the current first personal connection score may be determined according to the number of common friends and a preset fourth score; if the user type of the recommended user is a user concern type, the current first personal connection score may be determined according to the number of common concerns and a preset fourth score.
[0117] In actual application, if the recommended user is a friend of the user, the current first personal connection score can be recorded as a preset first score (for example, 50 points); if the recommended user is a common follow of the user's friends, assuming that N is the number of friends who follow the recommended user (where N>=2), the current first personal connection score can be recorded as N*10 (i.e., the preset second score); if the recommended user is a friend of the user's friend, assuming that N is the number of common friends (where N>=2), that is, the number of friends among the terminal user's friends who are friends with the recommended user, the current first personal connection score can be recorded as N*8 (i.e., the preset third score); if the recommended user is a follow of the user, assuming that N is the number of followers (where N>=2), that is, the number of users among the users followed by the terminal user who follow the recommended user, the current first personal connection score can be recorded as N*5 (i.e., the preset fourth score).
[0118] It should be noted that the user type of the recommended user can be not only one of the above user types, but also any combination of the above user types. When the user type of the recommended user is any combination type, the current first personal connection degree of the recommended user can be accumulated according to the scores that meet the above types, and the total first personal connection degree obtained by the accumulation does not exceed 100. If the score exceeds 100, 100 is taken as the total first personal connection degree; and the specific values of the above scores can be dynamically adjusted, so the embodiment of the present invention does not limit the specific values of the arbitrary combination form of user types and the scores.
[0119] In one embodiment of the present invention, after calculating the relevant personal connections of the recommended user by type (which can be updated regularly), the first personal connection score can also be calculated by combining the user activity of the recommended user. The user activity can refer to the situation of the recommended user posting dynamics, such as the time interval between posting dynamics, the number of posted dynamics, the frequency of posting dynamics, and the situation of dynamic interaction.
[0120] In a specific implementation, the second personal connection degree is obtained according to the user activity, and the second personal connection degree can be determined by a second personal connection score, and the second personal connection score can be expressed as a posting score and a posting interaction score calculated in combination with the posting situation of the recommended user.
[0121] In actual applications, if the recommended user has posted dynamics within a preset time period, the time of posting the dynamics and the interaction within the preset time period can be obtained, and the second personal connection score can be calculated according to the time of posting the dynamics and the interaction within the preset time period. That is, the posting score can be calculated according to the posting time, and the posting interaction score can be calculated according to the interaction within the preset time period, so as to determine the second personal connection score according to the posting score and the posting interaction score.
[0122] As an example, the rules for post scores can be as follows: For a recommended user who has posted a post within a preset time period (e.g., the past 10 days), the post score can be calculated based on the last post time (assumed to be N days ago, where 0 <= N <= 9), and the specific formula can be (10 - N) * 10; the rules for post interaction scores can be as follows: Assume N is the total number of likes / comments obtained by the recommended user for the posts within a preset time period (e.g., the past 10 days), then the score for this item can be N, and if N is greater than 100, then 100 is taken.
[0123] In a preferred embodiment, filtering can also be performed according to the user activity of the recommended user, and the main manifestation can be to determine whether the recommended user has posted a dynamic within a preset time period (e.g., the past 10 days). If no dynamic has been posted in the past 10 days, then the recommended user who has not posted a dynamic within the preset time period is filtered, that is, the recommended user is deleted and not written into the contact recommendation pool.
[0124] Sub-step S12, obtain the weight value of the recommended user in the preset contact recommendation pool, and generate the recommendation degree information of the recommended user by using the weight value and the degree of contact relationship information.
[0125] The weight value can refer to the recommended degree of the recommended user in the contact recommendation pool. At this time, the recommended sorting of the recommended user in the contact recommendation pool is updated jointly according to the weight value of the recommended user and the degree of contact relationship, so as to calculate the user contact data from multiple dimensions.
[0126] Specifically, for the determination of the weight value, first, it can be determined whether the recommended user exists in the preset contact recommendation pool; in one case, if the recommended user exists in the preset contact recommendation pool, then the weight value of the recommended user can be obtained, and the weight value can be determined by the recommendation pool score of the recommended user in the preset contact recommendation pool; in another case, if the recommended user does not exist in the contact recommendation pool, then the weight value of the recommended user can be set to a preset initial threshold (e.g., 1000, and the value of the preset initial threshold can be determined according to actual needs). Among them, the recommendation pool score is the same as the weight value, and both can be used to define the recommended degree of the recommended user in the contact recommendation pool, and it can also represent the degree of being ranked靠前 in the contact recommendation pool.
[0127] In practical applications, if the recommended user exists in the contact recommendation pool, then the weight value of the recommended user can be obtained. Among them, assume the recommendation pool score is X, then its weight value can be the integer part of X / 1000 multiplied by 1000, otherwise, the initial weight value of the user can be set to 10000.
[0128] Step 202, updating the recommended user to a preset network recommendation pool according to the recommended score;
[0129] In one embodiment of the present invention, the weight value and the personal connection score can be aggregated to obtain the recommended score of the recommended user (used to indicate the recommendation degree of the recommended user), and then the recommended score is used to update the ranking of the recommended user in the preset personal connection recommendation pool, and the recommended user is updated or inserted into the preset personal connection recommendation pool according to the updated ranking.
[0130] Step 203: Pull out the personal connection recommendation data from the preset personal connection recommendation pool according to the recommendation score of the recommended user.
[0131] In one embodiment of the present invention, the server may actively or passively pull the recommended network data and return it to the terminal, wherein the pulled recommended network data may be a preset number of user data pulled according to the recommended information, which may be specifically manifested as pulling a limited number of recommended user data in descending order of the recommended scores.
[0132] In a preferred embodiment, after returning the recommended user data to the terminal, the recommendation degree information of the recommended user can be updated according to the user's feedback, that is, the recommended score of the recommended user can be updated. The specific implementation steps can be as shown in step 103, which will not be elaborated here to avoid redundant content.
[0133] In the embodiment of the present invention, the first-degree / second-degree connections of the recommended user are calculated based on the user's friend relationships, follow-up relationships, etc., and the recommended scores of the recommended users are calculated and ranked in combination with the posting situation in the user terminal and the client behavior, and the user's follow-up strategy for the recommended user is dynamically adjusted in an experimental manner.
[0134] Reference Figure 3 , shows an application scenario diagram of the user recommendation method in the implementation of the present invention. Assuming that A is one of the recommended users pulled from the preset network recommendation pool by the server, and B is the terminal user, if A is to be recommended to B, the recommendation process can be as follows:
[0135] In the embodiment of the present invention, assuming that the recommended user A is a friend of the terminal user B, and the recommended user A and the terminal user B have two common friends, and the terminal user B has three friends who follow the recommended user A, then the recommended user A's personal connection information (which can be determined by the first personal connection score) can be 50+2*10+3*8=94; assuming that the recommended user A updated his last post 2 days ago, the posting score can be (10-2)*10=80; assuming that the total number of likes / comments on the recommended user A's posts in the past 10 days is 50, then the posting interaction score can be 50;
[0136] In summary, the network score of the recommended user A can be 94+80+50=224. Assuming that the recommended user A does not exist in the network recommendation pool, its initial weight value is 10000, and the final recommendation score can be 10224.
[0137] Also, if after the recommended user A is pushed to the terminal user B, the terminal user B views the recommended user A (that is, the terminal user B does not block or follow the recommended user A), the recommended score of the recommended user A is updated to 10224-1000=9224 (which is essentially processing the recommendation pool score for determining the weight value in the recommended score) to achieve the effect of subsequently pushing other recommended users in the network recommendation pool who have not been pushed to the terminal user B first.
[0138] It should be noted that the user recommendation method according to the embodiment of the present invention can improve the attention rate of the recommended users, and the attention rate of the recommended users is increased by 5% through experiments and statistics.
[0139] It should be noted that, for the sake of simplicity, the method embodiments are described as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present invention are not limited by the order of the actions described, because according to the embodiments of the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.
[0140] Reference Figure 4 , shows a structural block diagram of an embodiment of a user recommendation device of the present invention, which may specifically include the following modules:
[0141] The network recommendation data extraction module 401 is used to extract the network recommendation data for the recommended user from the preset network recommendation pool, and return the network recommendation data to the terminal; wherein the network recommendation data is extracted according to the recommendation degree information of the recommended user;
[0142] The operation information receiving module 402 is used to receive the operation information sent by the terminal for the recommended user; the operation information is generated by the feedback operation of the terminal user on the recommended user;
[0143] The recommendation level information updating module 403 is used to update the recommendation level information of the recommended user according to the operation information.
[0144] In an embodiment of the present invention, before extracting the network recommendation data for the recommended user from the preset network recommendation pool, the following modules may also be included:
[0145] A recommendation degree information generating module, used for acquiring user information of the recommended user and generating recommendation degree information according to the user information;
[0146] The recommended user updating module is used to update the recommended user to a preset network recommendation pool according to the recommendation degree information.
[0147] In an embodiment of the present invention, the user information of the recommended user includes the user type and the user activity; the recommendation degree information generation module may include the following submodules:
[0148] A personal connection degree information generation submodule, used to generate personal connection degree information according to the user type and the user activity;
[0149] A weight value acquisition submodule is used to obtain the weight value of the recommended user in a preset network recommendation pool;
[0150] The recommendation degree information generation submodule is used to generate the recommendation degree information of the recommended user by using the weight value and the personal connection degree information.
[0151] In one embodiment of the present invention, the user type includes at least one of the following: user attention type, user blocking type, user friend type, common attention type of user friend, friend type of user friend and attention type of user attention; wherein, the recommended user whose user type is user attention type or user blocking type is not in the preset network recommendation pool.
[0152] In an embodiment of the present invention, the personal connection degree information includes first personal connection degree information and second personal connection degree information; the personal connection degree information generation submodule may include the following units:
[0153] A first personal connection degree information generating unit, configured to generate first personal connection degree information according to the user type of the recommended user;
[0154] The second personal connection degree information generating unit is used to generate the second personal connection degree information according to the user activity.
[0155] In an embodiment of the present invention, the first personal connection degree information is determined by the first personal connection score; the first personal connection degree information generating unit may include the following subunits:
[0156] A first personal connection score determination subunit, configured to determine a current first personal connection score according to a preset first score if the user type of the recommended user is a user friend type;
[0157] The first interpersonal relationship score determination subunit is further configured to determine the current first interpersonal relationship score according to the number of user friends who commonly follow the recommended user and a preset second score if the user type of the recommended user is a common type of interest of the user's friends;
[0158] The first personal connection score determination subunit is further configured to determine the current first personal connection score according to the number of common friends and a preset third score if the user type of the recommended user is the friend type of the user's friend;
[0159] The first personal connection score determination subunit is further configured to determine the current first personal connection score according to the number of common concerns and a preset fourth score if the user type of the recommended user is a concern type that the user concerns.
[0160] In an embodiment of the present invention, the second personal connection degree information is determined by the second personal connection score; and the second personal connection degree information generating unit may include the following subunits:
[0161] A second interpersonal relationship score determination subunit is used to obtain the time of posting the dynamics and the interaction status within the preset time period if the recommended user has posted dynamics within the preset time period, and calculate the second interpersonal relationship score according to the time of posting the dynamics and the interaction status within the preset time period;
[0162] In an embodiment of the present invention, the second personal connection degree information generating unit may further include the following subunits:
[0163] The recommended user filtering subunit is used to filter the recommended users who have not posted any dynamics within a preset time period.
[0164] In one embodiment of the present invention, the weight value acquisition submodule may include the following units:
[0165] A recommended user determination unit, used to determine whether the recommended user exists in a preset network recommendation pool;
[0166] A weight value acquisition unit, configured to acquire a weight value of the recommended user if the recommended user exists in a preset personal connection recommendation pool; the weight value is determined by the recommendation pool score of the recommended user in the preset personal connection recommendation pool;
[0167] The weight value setting unit is used to set the weight value of the recommended user to a preset initial threshold if the recommended user does not exist in the network recommendation pool.
[0168] In one embodiment of the present invention, the recommended user update module may include the following submodules:
[0169] The recommended user updating submodule is used to update the ranking of the recommended user in the preset personal connection recommendation pool using the recommended score, and update or insert the recommended user into the preset personal connection recommendation pool according to the updated ranking.
[0170] In an embodiment of the present invention, the recommendation degree information is determined by the recommended score; the recommendation degree information updating module 403 may include the following submodules:
[0171] A recommended score updating submodule, used for updating the recommended score of the recommended user according to the operation information;
[0172] In one embodiment of the present invention, the recommended score updating submodule may include the following units:
[0173] A recommended score acquisition unit, used to acquire the recommended score of the recommended user before updating;
[0174] The recommended score updating unit is configured to update the recommended score of the recommended user to the recommended score before the update minus a preset score threshold if the operation information is a viewing operation for the recommended user.
[0175] In an embodiment of the present invention, the recommendation degree information updating module 403 may further include the following submodules:
[0176] A first recommended user deletion submodule is used to delete the recommended user from a preset network recommendation pool if the operation information is a follow-up operation information for the recommended user;
[0177] The second recommended user deletion submodule is used to delete the recommended user from the preset network recommendation pool if the operation information is blocking operation information for the recommended user.
[0178] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.
[0179] An embodiment of the present invention further provides a device, including:
[0180] It includes a processor, a memory, and a computer program stored in the memory and capable of running on the processor. When the computer program is executed by the processor, the various processes of the above-mentioned user recommendation method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.
[0181] An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the various processes of the above-mentioned user recommendation method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.
[0182] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.
[0183] It will be appreciated by those skilled in the art that the embodiments of the present invention may be provided as methods, devices, or computer program products. Therefore, the embodiments of the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the embodiments of the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.
[0184] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0185] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0186] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable terminal device. Figure 1A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0187] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention.
[0188] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.
[0189] The user recommendation method, device, equipment and storage medium provided by the present invention are introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for general technical personnel in this field, according to the idea of the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.
Claims
1. A user recommendation method, characterized in that: The method comprises: Pulling the recommended network data for the recommended user from a preset recommended network pool, and returning the recommended network data to the terminal; wherein the recommended network data is pulled according to the recommended degree information of the recommended user; receiving operation information for the recommended user sent by the terminal; the operation information is generated by a feedback operation of a terminal user on the recommended user; Update the recommendation level information of the recommended user according to the operation information; the recommendation level information is used to indicate the push level of the recommended user in the preset network recommendation pool; Before pulling the network recommendation data for the recommended user from the preset network recommendation pool, it also includes: Acquire user information of the recommended user; the user information of the recommended user includes user type and user activity; Generate personal connection degree information according to the user type and the user activity; Obtaining a weight value of the recommended user in a preset interpersonal relationship recommendation pool, and using the weight value and the interpersonal relationship degree information to generate recommendation degree information of the recommended user; Update the recommended user to a preset network recommendation pool according to the recommendation degree information; Among them, the recommendation degree information is generated based on the superposition of the personal connection degree information and the weight value of the recommended user in the preset personal connection recommendation pool, and the weight value includes a weight value determined by the recommendation pool score of the recommended user in the preset personal connection recommendation pool when the recommended user exists in the preset personal connection recommendation pool, and / or a weight value set by a preset initial threshold when the recommended user does not exist in the preset personal connection recommendation pool, wherein the preset initial threshold is greater than the weight value determined when the recommended user exists in the preset personal connection recommendation pool, and the recommended user whose user type is a user attention type or a user blocking type does not exist in the preset personal connection recommendation pool.
2. The method according to claim 1, characterized in that The user type includes at least one of the following: a user attention type, a user blocking type, a user friend type, a common attention type of a user's friend, a friend type of a user's friend, and an attention type that the user attentions.
3. The method according to claim 2, characterized in that The interpersonal relationship information includes first interpersonal relationship information and second interpersonal relationship information; and generating the interpersonal relationship information according to the user type and the user activity includes: The first personal connection degree information is generated according to the user type of the recommended user, and the second personal connection degree information is generated according to the user activity.
4. The method according to claim 3, characterized in that The first personal connection degree information is determined by the first personal connection score; and the generating the first personal connection degree information according to the user type of the recommended user includes: If the user type of the recommended user is a user friend type, determining a current first personal connection score according to a preset first score; and / or, if the user type of the recommended user is a common type of interest of the user's friends, determining the current first personal connection score according to the number of the user's friends who commonly interest the recommended user and a preset second score; and / or, if the user type of the recommended user is the friend type of the user's friend, determining the current first personal connection score according to the number of common friends and a preset third score; And / or, if the user type of the recommended user is a focus type that the user focuses on, the current first personal connection score is determined according to the number of common focuses and a preset fourth score.
5. The method according to claim 3, characterized in that: The second personal connection degree information is determined by a second personal connection score; and the generating the second personal connection degree information according to the user activity includes: If the recommended user has posted a dynamic within the preset time period, the time of posting the dynamic and the interaction within the preset time period are obtained, and the second personal connection score is calculated according to the time of posting the dynamic and the interaction within the preset time period; Also includes: Filter the recommended users who have not posted any updates within a preset time period.
6. The method according to claim 1, characterized in that The recommendation degree information is determined by the recommended score; and updating the recommended user to a preset network recommendation pool according to the recommendation degree information includes: The recommended score is used to update the ranking of the recommended user in the preset personal connection recommendation pool, and the recommended user is updated or inserted into the preset personal connection recommendation pool according to the updated ranking.
7. The method according to claim 1, characterized in that The updating of the recommendation degree information of the recommended user according to the operation information includes: Updating the recommended score of the recommended user according to the operation information; The updating of the recommended score of the recommended user according to the operation information includes: Obtain the recommended score of the recommended user before updating; If the operation information is a viewing operation for the recommended user, the recommended score of the recommended user is updated to be the recommended score before the update minus a preset score threshold.
8. The method according to claim 7, characterized in that Also includes: If the operation information is attention operation information for the recommended user, deleting the recommended user from the preset network recommendation pool; If the operation information is blocking operation information for the recommended user, the recommended user is deleted from the preset network recommendation pool.
9. A user recommendation device, characterized in that: The device comprises: A network recommendation data extraction module is used to extract network recommendation data for the recommended user from a preset network recommendation pool and return the network recommendation data to the terminal; wherein the network recommendation data is extracted according to the recommendation degree information of the recommended user; An operation information receiving module, used for receiving operation information sent by the terminal for the recommended user; the operation information is generated by a feedback operation of the terminal user on the recommended user; A recommendation degree information updating module, used to update the recommendation degree information of the recommended user according to the operation information; the recommendation degree information is used to indicate the push degree of the recommended user in the preset network recommendation pool; The steps before pulling the recommended network data for the recommended user from the preset network recommendation pool also include: obtaining user information of the recommended user; the user information of the recommended user includes user type and user activity; generating network relationship degree information according to the user type and the user activity; obtaining the weight value of the recommended user in the preset network recommendation pool, and generating the recommendation degree information of the recommended user by using the weight value and the network relationship degree information; updating the recommended user to the preset network recommendation pool according to the recommendation degree information; Among them, the recommendation degree information is generated based on the superposition of the personal connection degree information and the weight value of the recommended user in the preset personal connection recommendation pool, and the weight value includes a weight value determined by the recommendation pool score of the recommended user in the preset personal connection recommendation pool when the recommended user exists in the preset personal connection recommendation pool, and / or a weight value set by a preset initial threshold when the recommended user does not exist in the preset personal connection recommendation pool, wherein the preset initial threshold is greater than the weight value determined when the recommended user exists in the preset personal connection recommendation pool, and the recommended user whose user type is a user attention type or a user blocking type does not exist in the preset personal connection recommendation pool.
10. A device, characterized in that: include: A processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein when the computer program is executed by the processor, the steps of the user recommendation method according to any one of claims 1 to 7 are implemented.
11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the user recommendation method according to any one of claims 1 to 7 are implemented.
