A friend-making app intelligent user automatic distribution system

Abstract

The application discloses a kind of friend-making APP intelligent user automatic distribution system, it is applied to the matching and interaction scene between first user and second user, the system includes: index module, for setting multiple index data to first user;Multiple index data include: message reply rate, message reply speed, average paid message number, video conversion rate and video average duration;Grouping module is used to group processing second user according to registration account length;Second user in each group is configured to follow the index data of the first user;Recommendation module is used to configure the index data of the first user corresponding to the second user in each group, weight calculation is carried out to obtain weighted recommendation score;The first user with the highest weighted recommendation score is recommended to the corresponding second user.The system can solve the quality problem of inaccurate matching when social products recommend users, and help improve user experience.

Description

Technical Field

[0001] This invention relates to the field of mobile internet social networking APP technology, and more specifically to an intelligent user automatic distribution system for dating APPs. Background Technology

[0002] Currently, with the rapid development of mobile internet technology, social products are playing an increasingly important role in people's lives. However, most social products still suffer from problems such as inaccurate system matching and poor user experience.

[0003] These issues significantly impact the development of social networking products. Therefore, they must be given full attention during development, design, and promotion, with continuous optimization and improvement to enhance user experience and refine recommendation strategies. Only in this way can we win over more users and encourage long-term use.

[0004] Therefore, how to solve the problems of inaccurate user matching and poor user experience in social products is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] In view of this, the present invention provides an intelligent user automatic distribution system for dating apps, which can solve the quality problem of inaccurate matching when recommending users in social products and help improve the user experience.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] A smart user automatic distribution system for dating apps, applied to matching and interaction scenarios between first and second users, includes:

[0008] The metrics module is used to set various metrics for the first user; these metrics include: message reply rate, message reply speed, average number of paid messages, video conversion rate, and average video duration.

[0009] The grouping module is used to group the second user according to the length of time the account has been registered; the second user in each group is configured with corresponding indicator data that focuses on the first user;

[0010] The recommendation module is used to calculate a weighted recommendation score by assigning the corresponding indicator data of the second user in each group to the first user and performing weighted recommendation data on the second user. The first user with the highest weighted recommendation score is then recommended to the corresponding second user.

[0011] In one embodiment, it also includes:

[0012] The allocation module is used to assign the second user in each group to the new user pool and the ruminant pool according to the allocation strategy, and wait for the distribution recommendation.

[0013] In one embodiment, it also includes:

[0014] The configuration module is used to configure the validity period and capacity for the first user; the validity period is the interaction duration between the first user and any second user; the capacity is the number of second users that the first user can connect to simultaneously.

[0015] In one embodiment, it also includes:

[0016] The adjustment module is used to adjust the time-limit and capacity of each first user according to the ratio of the weighted recommendation score to the base score; wherein, the base score is the average of the indicators of each first user as the base score for different group distribution configurations of the first user; the weighted recommendation score is the weighted recommendation score of the recommendation module.

[0017] In one embodiment, it also includes:

[0018] The distribution module is used to encapsulate the second user who has successfully logged in as a distribution task object according to the judgment rules and add it to the new user pool or the rumination pool, waiting for the recommendation result of the recommendation module; the new user pool stores newly registered second users or second users who have completed conversion; the rumination pool stores second users who have been distributed but have not been successfully converted.

[0019] In one embodiment, the second user is encapsulated as a task distribution object, wherein the task contains the following information:

[0020] 1) The second user ID to be distributed;

[0021] 2) The user group ID to which the second user to be assigned belongs;

[0022] 3) The number of times the distribution needs to be done;

[0023] 4) The interval between each distribution;

[0024] 5) Distribution pool types; Distribution pool types include new pools and ruminant pools;

[0025] 6) Start time of distribution.

[0026] In one embodiment, it also includes:

[0027] The task scheduling module is used to obtain the current task that needs to be scheduled based on the development and distribution time in the task, and to generate specific distribution events and send them to the delay queue according to the time interval and number of distributions defined in the task.

[0028] In one embodiment, the distribution event includes the following information:

[0029] (1) The second user ID to be distributed;

[0030] (2) The user group ID to which the second user to be assigned belongs;

[0031] (3) Distribution pool type;

[0032] (4) Retry information;

[0033] (5) Event discarding time threshold; the event discarding time threshold is: if the event has not been processed when the time threshold is reached, the event is directly discarded.

[0034] In one embodiment, it also includes:

[0035] The stop-distribution module is used to stop distributing to the first user if, after the second user is distributed to the first user, the number of people who have not responded to the first user reaches N.

[0036] In one embodiment, it also includes:

[0037] The elimination module is used to rank all the top users' data for each metric. If the total number of all metrics of one of the top users is in the bottom x% of the overall ranking for M consecutive days, then the user is eliminated.

[0038] As can be seen from the above technical solution, compared with the prior art, this invention discloses an intelligent user automatic distribution system for dating apps, applied to the matching and interaction scenarios between a first user and a second user. The system includes: an indicator module for setting various indicator data for the first user; these indicator data include: message reply rate, message reply speed, average number of paid messages, video conversion rate, and average video duration; a grouping module for grouping second users according to their registration and account opening time; each second user in a group is configured with indicator data corresponding to the first user they follow; and a recommendation module for weighting the indicator data of each second user in each group corresponding to the first user they follow, calculating a weighted recommendation score; and recommending the first user with the highest weighted recommendation score to the corresponding second user. This system can solve the quality problem of inaccurate matching in user recommendations in social products, helping to improve the user experience. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0040] Figure 1This is a structural block diagram of the intelligent user automatic distribution system for dating apps provided by the present invention;

[0041] Figure 2 This is a schematic diagram of the intelligent user automatic distribution system for dating apps provided by the present invention.

[0042] Figure 3 A schematic diagram of the adjustment module provided by the present invention;

[0043] Figure 4 A schematic diagram of the distribution module provided for this invention;

[0044] Figure 5 A schematic diagram of the task distribution scheduling module provided by the present invention;

[0045] Figure 6 A schematic diagram of the event distribution processing logic provided by this invention;

[0046] Figure 7 This is a diagram illustrating the maintenance logic for the female user metrics and the female user recommendation list within the male user group provided by this invention. Detailed Implementation

[0047] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0048] Reference Figure 1 As shown in the figure, this invention discloses an intelligent user automatic distribution system for dating apps, applied to matching and interaction scenarios between a first user and a second user. The system includes:

[0049] The metrics module is used to set various metrics for the first user; these metrics include: message reply rate, message reply speed, average number of paid messages, video conversion rate, and average video duration.

[0050] The grouping module is used to group the second user according to the length of time the account has been registered; the second user in each group is configured with corresponding indicator data that focuses on the first user;

[0051] The recommendation module is used to calculate a weighted recommendation score by assigning the corresponding indicator data of the second user in each group to the first user and performing weighted recommendation data on the second user. The first user with the highest weighted recommendation score is then recommended to the corresponding second user.

[0052] Taking a matching scenario between male and female users as an example, where the first user is female and the second user is male, this will be the basis for all subsequent examples. Figure 2 As shown, in the overall system architecture, each female user will have multiple metrics, such as message reply rate, message reply speed, average number of paid messages, video conversion rate, and average video duration. The system groups male users, for example, those registered within 1 hour, within 24 hours, and within 7 days. Each group of male users can configure the metrics they follow for female users, such as... Figure 2 As shown in the diagram, "Male User Group 1" focuses on the message response rate and video conversion rate metrics of female users. When a male user in this group becomes active, the system will calculate a weighted recommendation score based on the message response rate and video conversion rate metrics of all currently online female users. The male user will be prioritized to be matched with the female user with the highest recommendation score, thereby improving the accuracy of the match and ensuring that the male user can interact with high-quality female users, thus increasing the conversion rate.

[0053] like Figure 1 As shown, the system also includes:

[0054] The allocation module is used to assign the second user in each group to the new user pool and the ruminant pool according to the allocation strategy, and wait for the distribution recommendation.

[0055] The configuration module is used to configure the validity period and capacity for the first user; the validity period is the interaction duration between the first user and any second user; the capacity is the number of second users that the first user can connect to simultaneously.

[0056] Among them, male users in each group will be distributed and recommended to the new pool and the ruminant pool according to certain strategies.

[0057] Female users will have different default capacity and time limits in different male user groups. When a female user's metrics change, a new weighted recommendation score will be calculated. The new recommendation score will be compared with the female user's base recommendation score (initial weighted recommendation score), and the capacity will be adjusted based on the change ratio, so that female users with good data can get more recommendation opportunities and improve the user experience.

[0058] To prevent female users from receiving a large number of recommendation opportunities in a short period, leading to an imbalance in traffic distribution, the system assigns female users the concepts of timeliness and distribution capacity. Different male user groups can be configured with the timeliness duration and capacity for distributing information to female users within that group. For example... Figure 3As shown: If the default validity period for female users in group A is 6 minutes and the capacity is 3, and the system recommends 1 male user from group A to female user A at 00:10 and 2 male users from group A to female user A at 00:12, then the remaining capacity for female user A is 2 from 00:10 to 00:12, 0 from 00:12 to 00:16, 1 from 00:16 to 00:18, and 3 after 00:18.

[0059] In one embodiment, such as Figure 1 As shown, the system also includes:

[0060] The adjustment module is used to adjust the time-limit and capacity of each first user according to the ratio of the weighted recommendation score to the base score; wherein, the base score is the average of the indicators of each first user as the base score for different group distribution configurations of the first user; the weighted recommendation score is the weighted recommendation score of the recommendation module.

[0061] Specifically, in implementation, for example, the system calculates the base recommendation score for female users in each male user group by taking the average of various female user metrics. Each different male user group can be configured with a base duration and capacity for distributing female users. The system will compare the ratio of each female user's recommendation score to the base recommendation score using a specific algorithm, thereby increasing or decreasing the response value based on the base duration and capacity. This ensures that high-quality female users have more opportunities to be distributed to male users.

[0062] In one embodiment, such as Figure 1 As shown, the system also includes:

[0063] The distribution module is used to encapsulate the second user who has successfully logged in as a distribution task object according to the judgment rules and add it to the new user pool or the rumination pool, waiting for the recommendation result of the recommendation module; the new user pool stores newly registered second users or second users who have completed conversion; the rumination pool stores second users who have been distributed but have not been successfully converted.

[0064] After a user logs in, male users will be added to a new user pool and a ruminant pool of a certain male user group based on certain judgment logic. The difference between the new user pool and the ruminant pool lies in the frequency and number of distributions.

[0065] For example, male users are grouped as "Registered for 1 hour without paying" and "Registered for 1 hour and paid 10 yuan+". Newly registered male users will initially be in the "Registered for 1 hour without paying" group. If, within 10 minutes, the system distributes male users to 3 female users and then the male user recharges 20 yuan, the system will detect this and automatically move the male user to the "Registered for 1 hour and paid 10 yuan+" group, and start distributing and recommending users based on the metrics followed by "Registered for 1 hour and paid 10 yuan+".

[0066] For example: new users in the pool are distributed every 5 minutes for a total of 8 times; users in the rumination pool are distributed every 10 minutes for a total of 3 times. After each distribution, it is determined whether the conversion is completed. If the conversion is not completed, the user will be added to the rumination pool for distribution on the next login.

[0067] The information added to the new pool or ruminant pool is an encapsulated dispatch task object, which contains the following information:

[0068] 1. Male user ID to be distributed;

[0069] 2. The ID of the male user group to which the male user to be assigned belongs;

[0070] 3. The number of times the distribution needs to be completed;

[0071] 4. The interval between each distribution;

[0072] 5. Distribution pool type;

[0073] 6. Start time of distribution;

[0074] The specific logic for determining the user's group and distribution pool type is as follows: Figure 4 As shown.

[0075] In one embodiment, such as Figure 1 As shown, the system also includes a distribution task scheduling module, which is used to obtain the distribution task that needs to be scheduled at present according to the development and distribution time in the distribution task, and generate specific distribution events and send them to the delay queue according to the time interval and number of distributions defined in the distribution task.

[0076] Once a distribution task is added to the distribution pool, the system will obtain the distribution task that needs to be scheduled based on the start distribution time in the distribution task. After obtaining the distribution task, the system will generate specific distribution events and send them to the delay queue according to the interval time and number of distributions defined in the distribution task. The time interval for processing these events is the interval time defined in the distribution task, and the number of distributions in the distribution task is the number of distribution events generated.

[0077] The distribution event contains the following information

[0078] 1. Male user ID to be distributed;

[0079] 2. The ID of the male user group to which the male user to be assigned belongs;

[0080] 3. Distribution pool type;

[0081] 4. Retry information;

[0082] 5. Event discarding time threshold (if the event has not been processed by the specified time, it will be discarded directly).

[0083] The specific task distribution scheduling logic and process are as follows: Figure 5 As shown.

[0084] Event dispatching logic: Each dispatching event execution means that the male user in the event will be dispatched to a female user. The specific event dispatching process is as follows: Figure 6 As shown.

[0085] In one embodiment, the female user metric and recommendation list maintenance logic are explained as follows:

[0086] As explained above, the system maintains data metrics for each female user, including message reply rate, message reply speed, average number of paid messages, video conversion rate, and average video duration. The specific calculation methods for these metrics are as follows:

[0087] 1. Message response rate = Number of messages replied / Total number of messages.

[0088] 2. Message response speed = Total message response time / Number of messages responded to.

[0089] 3. Average number of paid messages = Total number of paid messages / Paid message tasks

[0090] 4. Video conversion rate = Number of people converting from video / Total number of people interacting in chat

[0091] 5. Average video duration = Total video duration / Number of videos

[0092] Each male user group can specify the types of metrics they focus on and calculate weighted recommendation scores for these metrics for female users to maintain their female user recommendation list. For example, if male user group A focuses on message response rate and message response speed, the corresponding weighted recommendation score for female users would be calculated as follows:

[0093] The weighted recommendation score for female user 1 in male user group A = female user 1's message reply rate * message reply rate weight + female user 1's message reply speed * message reply speed weight.

[0094] The maintenance logic for female user metrics and the female user recommendation list within the male user group is as follows: Figure 7 As shown.

[0095] Figure 7 Each time a new recommendation score is calculated, the change ratio is calculated with the base recommendation score, and the capacity and timeliness information of female users in the male user group are maintained based on this change ratio, so that those with higher recommendation scores can get more recommendation opportunities.

[0096] In one embodiment, such as Figure 1 As shown, the system also includes:

[0097] The stop-dispatch module is used to stop distributing to the first user if, after the second user is distributed to the first user, the number of first users who have not responded reaches N. This ensures that the first user has sufficient matching and interaction capabilities, thus improving the experience for the second user.

[0098] The elimination module ranks all top users based on various metrics. If a top user's total metrics rank in the bottom x% of the overall ranking for M consecutive days, they are eliminated. This can indirectly incentivize top users.

[0099] In practice, for example, when the system distributes male users to female users, if the number of female users who have not responded reaches N (which can be configured in the male user group), the system will stop distributing to that female user.

[0100] In addition, it also has a female user elimination logic function: the system will rank female users according to various indicators. If a female user's indicators are in the bottom x% (e.g., 5%) of the total number of indicators for M consecutive days (e.g., 7 days), she will be eliminated.

[0101] This invention provides an intelligent user automatic distribution system for dating apps. By using various metrics, it enhances the ability of a first user to attract a second user, thereby improving user activity and user experience on the dating app. Simultaneously, the system groups second users through a grouping module, assigning different attention configurations to users within different groups, and uses a recommendation module to calculate a weighted recommendation score, improving recommendation accuracy and user matching, ultimately increasing user satisfaction and retention rates on the dating app.

[0102] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.

[0103] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A smart user automatic distribution system for a dating app, characterized in that, The system, applied to matching and interaction scenarios between a first user and a second user, includes: The metrics module is used to set various metrics for the first user; these metrics include: message reply rate, message reply speed, average number of paid messages, video conversion rate, and average video duration. The grouping module is used to group the second user according to the length of time the account has been registered; the second user in each group is configured with corresponding indicator data that focuses on the first user; The recommendation module is used to calculate a weighted recommendation score by assigning the corresponding indicator data of the second user in each group to the first user and performing weighted recommendation data on the second user. The first user with the highest weighted recommendation score is then recommended to the corresponding second user. The allocation module is used to allocate the second users in each group to the new user pool and the rumination pool according to the allocation strategy, and wait for the distribution recommendation; the new user pool stores newly registered second users or second users who have completed the conversion; the rumination pool stores second users who have been distributed but have not successfully converted. The configuration module is used to configure the validity period and capacity for the first user; the validity period is the interaction duration between the first user and any second user; the capacity is the number of second users that the first user can connect to simultaneously. The adjustment module is used to adjust the time-limit and capacity of each first user according to the ratio of the weighted recommendation score to the base score; wherein, the base score is the average of the indicators of each first user as the base score for different group distribution configurations of the first user; the weighted recommendation score is the weighted recommendation score of the recommendation module.

2. The intelligent user automatic distribution system for a dating app according to claim 1, characterized in that, Also includes: The distribution module is used to encapsulate the second user who has successfully logged in as a distribution task object according to the judgment rules and add it to the corresponding new pool or ruminant pool, waiting for the recommendation result of the recommendation module. The newly added pool stores newly registered second users or second users who have completed the conversion; the rumination pool stores second users who have been distributed but have not been successfully converted.

3. The intelligent user automatic distribution system for a dating app according to claim 2, characterized in that, The second user is encapsulated as a task distribution object, and the task contains the following information: 1) The second user ID to be distributed; 2) The user group ID to which the second user to be assigned belongs; 3) The number of times the distribution needs to be completed; 4) The interval between each distribution; 5) Distribution pool types; Distribution pool types include new pools and ruminant pools; 6) Start time of distribution.

4. The intelligent user automatic distribution system for a dating app according to claim 3, characterized in that, Also includes: The task scheduling module is used to obtain the current task to be scheduled based on the start time of the task, and generate specific distribution events according to the time interval and number of distributions defined in the task, and send them to the delay queue.

5. The intelligent user automatic distribution system for a dating app according to claim 4, characterized in that, The distribution event includes the following information: (1) The second user ID to be distributed; (2) The user group ID to which the second user to be assigned belongs; (3) Distribution pool type; (4) Retry information; (5) Event discarding time threshold; the event discarding time threshold is: if the event has not been processed when the time threshold is reached, the event is directly discarded.

6. The intelligent user automatic distribution system for a dating app according to claim 5, characterized in that, Also includes: The stop-distribution module is used to stop distributing to the first user if, after the second user is distributed to the first user, the number of people who have not responded to the first user reaches N.

7. The intelligent user automatic distribution system for a dating app according to claim 1, characterized in that, Also includes: The elimination module is used to rank all the top users' various metrics. If the total number of metrics of one of the top users is in the bottom x% of the overall ranking for M consecutive days, then the user is eliminated.

Images