CN113343089A

CN113343089A - User recall method, device and equipment

Info

Publication number: CN113343089A
Application number: CN202110653646.XA
Authority: CN
Inventors: 陈瑽; 寇京博; 庄涛; 田吉亮
Original assignee: Beijing Perfect Chijin Technology Co Ltd
Current assignee: Beijing Perfect Chijin Technology Co Ltd
Priority date: 2021-06-11
Filing date: 2021-06-11
Publication date: 2021-09-03

Abstract

The invention provides a user recall method, a device and equipment, wherein the method comprises the following steps: the method comprises the steps of obtaining game data of a plurality of users, taking users meeting a recall target in the plurality of users as reference users, and taking users meeting a reference target and not meeting the recall target in the plurality of users as users to be analyzed; inputting game data into a recall analysis model so that the recall analysis model calculates users to be analyzed based on reference users to obtain the similarity of the users to be analyzed and the reference users in game performance; and determining the users to be analyzed with the similarity meeting the preset conditions as target recall users, and pushing game resources corresponding to the recall targets to the target recall users. According to the method, through automatic screening of the target recall users, manual experience analysis is not needed, users with large recall potential are avoided from being omitted, recall efficiency is improved, the scale of the target recall users is controlled, and recall cost is reduced. And the recall effect is improved by pushing game resources more suitable for target recall users.

Description

User recall method, device and equipment

Technical Field

The invention relates to the technical field of internet, in particular to a user recall method, a device and equipment.

Background

With the development of cloud technology and mobile terminal technology, game release iteration is increasingly accelerated, and users are easily distracted by new games, so that the proportion of lost users in the whole registered users is higher and higher. Therefore, how to recall lost users becomes an important means for ensuring stable operation of the game.

At present, all lost users are generally acquired by adopting a data analysis method, and then recall information is sent to all lost users according to contact ways such as a mobile phone number and a mailbox, so that the lost users can log in the game again through rewards or new function introduction in the recall information. However, the method not only has high recall cost, such as huge short message expense, reward expense and the like, but also lacks pertinence and has poor recall effect.

In the related art, users meeting the policy can be selected from all lost users based on the recall policy set by the technical staff, and recall information is sent to the users. However, the method has high requirements on the capability and experience of technicians, some retrievable users are easy to miss, and the recall effect is difficult to guarantee.

Therefore, a solution for user recall is desired to solve at least one of the above technical problems.

Disclosure of Invention

The embodiment of the invention provides a user recall method, a user recall device and user recall equipment, which are used for selecting users with larger recall potential and providing more applicable game resources for the users, so that the user recall efficiency is improved, and the user recall effect is ensured.

In a first aspect, an embodiment of the present invention provides a user recall method, where the method includes:

the method comprises the steps of obtaining game data of a plurality of users, wherein users meeting recall targets in the users serve as reference users, and users meeting reference targets and not meeting the recall targets in the users serve as users to be analyzed;

inputting game data into a recall analysis model so that the recall analysis model calculates users to be analyzed based on reference users to obtain the similarity of the users to be analyzed and the reference users in game performance;

and determining the users to be analyzed with the similarity meeting the preset conditions as target recall users, and pushing game resources corresponding to the recall targets to the target recall users.

In a second aspect, an embodiment of the present invention provides a user recall apparatus, including:

the acquisition module is used for acquiring game data of the target object, wherein the game data comprises game system data of the target object in a first time period;

the analysis module is used for inputting the game data into the recall analysis model so as to enable the recall analysis model to calculate the user to be analyzed based on the reference user and obtain the similarity of the user to be analyzed and the reference user in game performance;

and the pushing module is used for determining the users to be analyzed, the similarity of which meets the preset conditions, as target recall users and pushing game resources corresponding to the recall targets to the target recall users.

In a third aspect, an embodiment of the present invention provides an electronic device, which includes a processor and a memory, where the memory stores executable code, and when the executable code is executed by the processor, the processor is enabled to implement at least the user recall method in the first aspect.

In a fourth aspect, embodiments of the present invention provide a non-transitory machine-readable storage medium having executable code stored thereon, which when executed by a processor of an electronic device, causes the processor to implement at least the user recall method of the first aspect.

In the technical scheme provided by the embodiment of the invention, the game data of a plurality of users can be obtained, wherein the users meeting the recall target in the plurality of users are taken as reference users, and the users meeting the reference target and not meeting the recall target in the plurality of users are taken as users to be analyzed, so that the initial division of the user group is realized through the recall target and the reference target. In order to predict the recall potential of the user to be analyzed, the game data can be input into the recall analysis model, so that the recall analysis model calculates the user to be analyzed based on the reference user to obtain the similarity of the user to be analyzed and the reference user in game performance, and thus, the recall potential of the user to be analyzed can be predicted through the similarity. And finally, determining the user to be analyzed, the similarity of which to the game performance of the reference user meets the preset conditions, as a target recall user, and pushing game resources corresponding to the recall target to the target recall user. In the embodiment of the invention, the similarity of the user to be analyzed and the reference user on the game performance is analyzed through the recall analysis model, and the recall potential of the user to be analyzed is judged based on the similarity, so that the automatic screening of the target recall user is realized, the manual experience analysis is not required, the omission of the user with larger recall potential is avoided, the recall efficiency is greatly improved, the scale of the target recall user can be controlled, and the recall cost is reduced. Meanwhile, game resources corresponding to the recall target are pushed to the target recall users, so that the target recall users under different recall targets can acquire more interesting game resources, and the recall effect is improved.

Drawings

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

FIG. 1 is a flowchart of a user recall method according to an embodiment of the present invention;

FIG. 2 is a flow chart of a method for determining game resources according to an embodiment of the present invention;

FIG. 3 is a schematic structural diagram of a user recall device according to an embodiment of the present invention;

fig. 4 is a schematic structural diagram of an electronic device corresponding to the user recall device provided in the embodiment shown in fig. 3.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

The terminology used in the embodiments of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in the examples of the present invention and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise, and "a plurality" typically includes at least two.

The words "if", as used herein, may be interpreted as "at … …" or "at … …" or "in response to a determination" or "in response to a detection", depending on the context. Similarly, the phrases "if determined" or "if detected (a stated condition or event)" may be interpreted as "when determined" or "in response to a determination" or "when detected (a stated condition or event)" or "in response to a detection (a stated condition or event)", depending on the context.

It is also noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a good or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such good or system. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a commodity or system that includes the element.

In addition, the sequence of steps in each method embodiment described below is only an example and is not strictly limited.

The user recall method provided by the embodiment of the invention can be executed by an electronic device, and the electronic device can be a terminal device such as a PC (personal computer), a notebook computer, a smart phone and the like, and can also be a server. The server may be a physical server including an independent host, or may also be a virtual server carried by a host cluster, or may also be a cloud server.

In practical applications, the user recall method provided by the embodiment of the present invention can be applied to any games, such as player recall scenes in open world games, shooting games, card games, and the like.

In fact, in different games, in order to adapt to the characteristics of different games, there are differences in the setting modes of recall targets. For example, in the open world game, a recall target is set according to data such as game copy level and the like, in the shooting game, the recall target is set according to the weapon use condition, and in the card game, the recall target is set according to card selection preference. These recall target setting bases are examples, and the recall target setting bases of the above-described game in actual application may be different.

The following describes the execution of the user recall method in conjunction with the following embodiments.

For this reason, in the related art, a paid user and a high-level user may be selected from all the attrition users based on a recall policy set by a technician, and recall information may be sent to these users. Although the method improves the pertinence of user recall, the method has higher requirements on the capability and experience of technicians, easily omits some recallable users, and the recall effect is difficult to guarantee.

In view of at least one of the above technical problems, the following describes an implementation process of the user recall method provided herein with reference to the following embodiments.

Fig. 1 is a flowchart of a user recall method according to an embodiment of the present invention, as shown in fig. 1, the method includes the following steps:

101. and obtaining game data of a plurality of users, wherein users meeting the recall target in the plurality of users are used as reference users, and users meeting the reference target and not meeting the recall target in the plurality of users are users to be analyzed.

102. And inputting the game data into the recall analysis model so that the recall analysis model calculates the user to be analyzed based on the reference user to obtain the similarity of the user to be analyzed and the reference user in game performance.

103. And determining the users to be analyzed with the similarity meeting the preset conditions as target recall users, and pushing game resources corresponding to the recall targets to the target recall users.

In the method provided by fig. 1, the similarity of the user to be analyzed and the reference user in the game performance is analyzed through the recall analysis model, and the recall potential of the user to be analyzed is judged based on the similarity, so that the automatic screening of the target recall user is realized, the manual experience analysis is not required, the omission of the user with large recall potential is avoided, the recall efficiency is greatly improved, the scale of the target recall user can be controlled, and the recall cost is reduced. Meanwhile, the method also pushes the game resources corresponding to the recall target to the target recall users, so that the target recall users under different recall targets can acquire more interesting game resources, and the recall effect is improved.

Specific implementations of the steps are described below with reference to specific examples.

First, game data of a plurality of users is acquired in 101. Since the plurality of users include a user who remains in the game (i.e., a reference user described below) and a user who does not remain in the game (i.e., an attrition user described below), among the plurality of users, the preliminary division of the user group is achieved by the recall target and the reference target. Optionally, the recall target and the reference target are determined according to the judgment condition of the attrition user. For example, if a user who has not logged in a game for 7 days or has not completed a given game performance for 7 days is determined to be an attrition user, the recall target and the reference target may be established based on the game login time or the execution of a given game performance (e.g., completion of a certain task). For example, the recall target may be a user who logged in for 15 days and completed a specified game performance, and the reference target may be a user who logged in for 3 days or completed a specified game performance.

Specifically, a reference user is screened out through the recall target, and the game data of the reference user is the reference standard in the recall process. In practical applications, the recall target is set according to one or a combination of the following factors: user grade, login days, friend number, friend activeness, recharging condition, game system data, prop use condition, weapon use condition, battle force change condition and copy level. The game system is, for example, a pet system, a face-pinching system, a fighting system, a replica task system, and a collection system. The game system data is, for example, usage of various game systems. Specifically, the gaming system data includes, but is not limited to, the following examples: the system comprises a pet system, a pet cultivation numerical value, a pet attribute, a task access condition, a completion progress, a task success rate and a task failure rate in the task system, a user arrival area, a user arrival frequency, a user preference route and a user preference place type in the map system.

For example, the recall target is set to log in for 7 days and reach level 40, so that attrition users with the potential to reach a higher level are selected by the recall target. Alternatively, the recall target is set to log in for 7 days, the number of friends exceeds 10 and the recharge exceeds 100 dollars, so that attrition users with the potential to reach higher liveness are selected through the recall target. Optionally, in order to screen out lost users with more recall potential, the lost users are classified by setting a recall target. In fact, the selected target recall users can be optimized by iteratively adjusting the recall target, so that the recall effect is further improved.

For example, it is assumed that the recall target is that the player level reaches 40 levels, which means that, during the recall, the users whose player level reaches 40 levels belong to the users who meet the recall target.

To facilitate recall analysis, optionally, a minimum login interval that the surviving user should reach is also set in the recall target. For example, the minimum login interval is set to 3 days, then a user who has logged in the game within 3 days can be regarded as a remaining user, and a user who has not logged in the game for 3 or more days can be regarded as an attrition user.

It is worth noting that attrition users refer to users that are not remaining in the game. In short, the lost user is the user whose time difference between the last login time and the current time exceeds the set threshold. For example, a user who is not logged into the game for n days may be determined to be an attrition user. Wherein n is a set time threshold, and the threshold can be set according to different application scenarios.

Taking the field of hand-swimming as an example, a user who has no active behavior within 7 days is generally called an attrition user. Active behavior refers to logging in or completing a specified game performance (e.g., 30 minutes online, a daily punch task, a specified task, etc.). The designated game behaves differently from game to game. Optionally, the correlation between various game performances and the retention users is acquired, and the game performance with the highest correlation is used as the designated game performance for judging the attrition users.

In fact, some users with low recall potential, such as users who do not log in again after registration, users who have online time less than 30 minutes, or users who stay below level 10, also exist among the attrition users, and obviously, these users not only have low recall potential, but also lack game data for analyzing the recall potential. In the embodiment of the invention, in order to improve the recall efficiency, a reference target for screening out the users to be analyzed from all lost users is further arranged, and then, in 101, the users meeting the reference target and not meeting the recall target are taken as the users to be analyzed. In this way, users with low recall potential (i.e., users who do not reach the reference condition) among a plurality of users can be excluded, and the data volume of the game data to be analyzed is reduced, thereby improving the efficiency of subsequent recall analysis.

Generally, the baseline target is implemented with a difficulty lower than the recall target, such as the recall target set to log in for 7 days and reach 40 levels, and the baseline target may be set to log in for 3 days and reach 15 levels. Of course, in practical applications, the setting of the reference target does not only refer to the recall target, but also can be set according to specific application occasions, such as specific game types, play characteristics, and the like.

Further, after the reference user and the user to be analyzed are divided, in 102, the game data is input into the recall analysis model, so that the recall analysis model calculates the user to be analyzed based on the reference user, and the similarity of the user to be analyzed and the reference user in the game performance is obtained.

In short, from the perspective of game performance, the higher the similarity with the reference user is, that is, the more similar the behavior of the user to be analyzed and the reference user in the game is, the higher the possibility that the user to be analyzed reaches the recall target is, the greater the recall potential of the user to be analyzed is, and therefore, the recall potential of the user to be analyzed can be quantitatively represented by the similarity between the user to be analyzed and the reference user in the game performance, so as to assist in the subsequent screening of the target recall user.

The game performance in the invention comprises the interaction behavior of the user and the game system. In particular, game performance includes, but is not limited to, login behavior, interaction behavior with the NPC, interaction behavior with other players, and behavior in various gaming systems (e.g., combat systems, collection systems, weaponry systems, skills systems, face-pinching systems, culture systems, etc.).

The recall analysis model is selected according to the recall requirement of the user. For example, the recall analysis model may be a machine learning model used for similarity calculation, such as various similarity models.

Specifically, assuming that the recall analysis model is a machine learning model for similarity calculation, based on this, game data is input into the recall analysis model, so that the recall analysis model calculates the user to be analyzed based on the reference user to obtain the similarity between the user to be analyzed and the reference user, which can be specifically implemented as:

selecting first game data generated before the reference user meets a reference target from the game data of the reference user; selecting third game data generated before the user to be analyzed meets a reference target from the game data of the user to be analyzed; and calculating the similarity between the third game data and the first game data through a machine learning model for similarity calculation, and taking the calculated similarity between the third game data and the first game data as the similarity between the user to be analyzed and the reference user.

In the above steps, it is assumed that the recall target is set to log in for 7 days and reaches 40 levels, and the reference target is set to log in for 3 days and reaches 15 levels.

Based on the above assumptions, users who log in for 7 days and reach 40 levels are used as reference users, and users who log in for more than 3 days and less than seven days and reach 15 levels but less than 40 levels are used as users to be analyzed. From the game data of the reference user, the game data generated 3 days before the reference user is logged in or 15 levels before the reference user is selected as the first game data. Similarly, from the game data of the user to be analyzed, game data generated 3 days before the user to be analyzed is logged in or 15 levels before the user to be analyzed is selected as third game data.

And then, inputting the first game data and the third game data into a similarity model, calculating the similarity between the third game data and the first game data through the similarity model, and taking the calculated similarity between the third game data and the first game data as the similarity between the user to be analyzed and the reference user.

Optionally, to simplify the similarity calculation process and improve the calculation efficiency, the similarity may be calculated by using one or more types of game data: the number of friends, the activity of friends, the recharging condition, the game system data, the use condition of props, the use condition of weapons, the change condition of the battle effectiveness, and the level of the copy.

Of course, in order to ensure the accuracy of the similarity analysis, the similarity can be calculated by using the whole amount of game data. For example, the similarity is calculated using all game data generated before the user satisfies the reference target.

Finally, after the similarity between the user to be analyzed and the reference user is obtained, in 103, the user to be analyzed, whose similarity satisfies the preset condition, is determined as the target recall user, and the game resource corresponding to the recall target is pushed to the target recall user.

The preset condition that the similarity needs to meet includes, but is not limited to, any one of the following: the similarity is not less than a similarity threshold, and the sequence obtained based on similarity sorting belongs to a preset target sequence.

For example, assume that the preset condition is that the similarity is not less than the similarity threshold. The similarity threshold is assumed to be 60%. Suppose that the similarity of each of the users a, b, c to be analyzed to the reference user is 83%, 40%, 66%. Based on the above assumptions, in 103, the users a and c to be analyzed with similarity not less than 60% are determined as target recall users, and the game resources corresponding to the recall targets are pushed to the target recall users a and c.

In another example, it is assumed that the preset condition is that the orders obtained based on the similarity ranking belong to a preset target order. Assume a target order of 100. Assuming that the number of users to be analyzed is m (where m is greater than 100), based on the above assumptions, users 1, 2, 3, … …, m to be analyzed are sorted from large to small according to similarity.

Further, assuming that the ranking results are users 1, 2, 3, … …, and m to be analyzed, based on this, the top 100 are selected as target recall users, that is, users 1 to 100 to be analyzed are selected as target recall users.

It is worth noting that if there are multiple target recall users, the same type of game resource can be pushed to the multiple target recall users. In order to increase the interest of the game, optionally, the game resources are randomly selected from the same type of game resources and pushed to different target recall users.

Or the target recall users can be divided according to other dimension data, so that different types of game resources are pushed for different target recall user groups. For example, according to the game area where the target recall user is located, the game resources defined by the area are pushed for the target recall user in the same area. Or pushing game resources related to respective roles, such as skins, activities, pets and the like, for target recall users with different role types according to the role types of the target recall users.

Here, the manner of determining the game resource corresponding to the recall target may be described below, and will not be expanded here.

Optionally, after 103, the number of successfully recalled users in the target recall user and the corresponding game data are obtained, and the obtained game data and the historical game data are subjected to iterative analysis to optimize the recall analysis model. Wherein, the number of successfully recalled users in the target recall users can be used as one of the reference data of the iterative analysis.

Optionally, the iterative analysis result may also be used to adjust a display mode or display content of the game resource. For example, suppose that the game resource is a plurality of groups of new function introduction characters, based on the successful recall rate of the target recall user corresponding to each group of new function introduction characters, the new function introduction character with higher success rate is selected as the subsequently used game resource.

In practical application, optionally, the game resources can be dynamically sequenced by adopting the iteration analysis result so as to increase the push proportion of the game resources with higher recall success rate, thereby enriching the game resources and further improving the recall efficiency.

Through 101 to 103, through the autofilter to the target recall user, not only need not to rely on the human experience analysis, avoid omitting the great user of recall potentiality, promote the efficiency of recalling greatly, still can control the scale of target recall user, reduce the cost of recalling. Meanwhile, the method also pushes the game resources corresponding to the recall target for the target recall users, so that the target recall users under different recall targets can acquire more interesting game resources, the game playability is improved, and the recall effect is improved.

In the foregoing or the following embodiments, in order to improve the pertinence of the game resource, the game resource determining method shown in fig. 2 may be further performed, and the specific steps are as follows:

201. selecting historical game data generated before a reference target is met from game data of a plurality of users, wherein the historical game data comprises: generating first game data before a reference user meets a reference target, and generating second game data before a user who does not meet a recall target meets the reference target;

202. respectively marking the first game data and the second game data by adopting different numerical values to obtain marked historical game data;

203. inputting the marked historical game data into a correlation analysis model so that the correlation analysis model calculates the correlation between the historical game data and the recall target;

204. game resources are determined based on correlations between historical game data and recall targets.

Wherein, the game data of the historical game data includes but is not limited to any one or combination of the following: user grade, login days, friend number, friend activeness, recharging condition, game system data, prop use condition, weapon use condition, battle force change condition and copy level. The specific parameters are similar to the above and are not expanded here.

In the above steps, continuing the above example, it is still assumed that the recall target is set to log in for 7 days and reaches 40 levels, and the reference target is set to log in for 3 days and reaches 15 levels.

Based on the above assumptions, users who log in for 7 days and reach level 40 are used as reference users. In 201, historical game data generated before a reference object is satisfied is extracted from game data of a plurality of users. Specifically, from the game data of the above-described reference user, game data generated 3 days before the reference user satisfies the login or 15 levels before the reference user is selected as the first game data. From the game data of the users who do not satisfy the recall target (including the users who reach the reference target and the users who do not reach the reference target), the game data generated 3 days before the user to be analyzed satisfies the login or 15 levels is selected as the second game data. In 202, the marked historical game data is obtained by marking the first game data as 1 and the second game data as 0.

Further, assuming that the correlation analysis model is a regression analysis model, at 203, the marked historical game data is input into the regression analysis model so that the regression analysis model calculates a correlation matrix between the historical game data and the recall target, wherein the correlation matrix reflects the correlation between each historical game data and the recall target.

Finally, assuming that there are a plurality of historical game data, based on this, in 204, determining game resources based on the correlation of the historical game data and the recall target may be implemented as:

arranging the historical game data from large to small according to the correlation between the historical game data and a recall target, and acquiring the data types of the historical game data in the previous target position; and taking the game resource corresponding to the acquired data type as the game resource.

For example, the plurality of historical game data are arranged from large to small in the correlation with the recall target, and the data types to which the historical game data in the top 2-digit order belong are acquired. Assuming that the data types to which the top 2 pieces of historical game data belong are pet culture data and pet capture data, it is known that the correlation between the usage of the pet system and the recall target is strong, and the game resource corresponding to the pet system is used as the game resource.

In practical applications, the correlation analysis model may be implemented by using other models or algorithms besides the regression analysis model, and the present invention is not limited thereto.

In this embodiment, the relevance between the game data and the recall target is analyzed from each dimension through the relevance analysis model, so that the game data with stronger relevance is used as a basis for selecting the game resources, the pertinence of the game resources is improved, and the recall effect is further improved.

In the foregoing or following embodiments, optionally, a plurality of recall targets are set to recall user groups of different types and different levels, so as to further improve recall efficiency and enhance recall effect.

Specifically, if the recall analysis model is provided with a plurality of recall targets, the plurality of users are divided into a plurality of reference user groups and a plurality of user groups to be analyzed corresponding to the reference user groups based on the plurality of recall targets.

Furthermore, after the user groups to be analyzed corresponding to the plurality of reference user groups and the plurality of reference user groups are divided, in 102, the recall analysis model calculates the user to be analyzed based on the reference user to obtain the similarity between the user to be analyzed and the reference user, which can be realized as follows:

and calculating the corresponding user groups to be analyzed based on the plurality of reference user groups through the recall analysis model to obtain the similarity between each user group to be analyzed and the corresponding reference user group. The similarity between each user group to be analyzed and the corresponding reference user group comprises the following steps: and the similarity between each user to be analyzed in each user group to be analyzed and the corresponding reference user group.

Furthermore, based on the similarity between each user group to be analyzed and the corresponding reference user group output by the recall analysis model, in 103, the user to be analyzed whose similarity satisfies the preset condition is determined as the target recall user, which can be realized as follows:

and selecting users to be analyzed with similarity meeting preset conditions from all the user groups to be analyzed to form a target recall user group corresponding to each user group to be analyzed.

Optionally, assuming that each user group to be analyzed corresponds to each recall target one by one, based on this, in 103, before pushing the game resource corresponding to the recall target to the target recall user, it is further determined whether the same target recall user exists in different target recall user groups.

If the same target recall users exist in different target recall user groups, the situation that the different target recall user groups have repeated data is indicated, in order to avoid harassment phenomena caused by pushing of multiple game resources, the target recall user groups corresponding to all user groups to be analyzed can be subjected to re-ranking processing based on the difficulty level of the recall target, so that the different target recall user groups do not have the same target recall user, and the game resources are pushed only once for each target recall user.

Of course, if some target recall users need to be pushed for multiple times to increase the recall probability, other re-ranking processing modes can be set. For example, a plurality of recall targets with stronger relevance with the target recall user are reserved, and game resources corresponding to the recall targets are pushed for the recall targets.

In the embodiment, the types and the levels of target recall users are enriched by setting a plurality of recall targets, so that the recall effect of the users is further improved.

In the above or below embodiments, assuming that there are a plurality of reference users among the plurality of users, optionally, the game data of the plurality of reference users is acquired, and the reference performance is selected from the game performances of the plurality of reference users based on the game data of the plurality of reference users. Wherein the reference performance includes, but is not limited to, one or a combination of the following: the behavior executed by all reference users exceeding the preset number of people, the behavior of the reference user reaching the recall target in the set time, the behavior of the reference user with higher grade, the behavior of the reference user with higher activity and the behavior of the reference user with more recharge.

For example, suppose there are 1000 reference users, based on which the game data of the 1000 reference users are obtained, and based on the game data of the 1000 reference users, the game performances associated with the respective reference users are respectively marked, and from the game performances of the 1000 reference users, the game performances marked with the number of times ranking top target positions are selected as the reference performances. Or selecting the game performance associated with the reference user with the target position before the liveness ranking and the game performance of the reference user reaching the recall target within 2 days from the game performances of 1000 reference users to form the reference performance.

In this way, further, the recall analysis model in 102 calculates the game performance of the user to be analyzed based on the reference performance, and obtains the similarity between the user to be analyzed and the reference user in the game performance. Or screening the game resources to be pushed based on the correlation between the game data of the user and the reference performance.

In practical applications, the game performance in this embodiment may refer to the game performance described above, i.e., the interaction between the user and the game system. Wherein the reference user-associated game performance is, for example, one or a combination of the following: login behavior, interaction with the NPC, interaction with other players, and behavior in various gaming systems (e.g., combat systems, collection systems, weaponry systems, skills systems, face-pinching systems, training systems, etc.).

In the above or below embodiments, in addition to the recall target introduced in the above embodiments, in practical applications, the recall target may be optionally set in the manner of a recall target guide map. Specifically, the recall target guide graph comprises a tree-shaped hierarchical structure divided based on recall targets and a plurality of nodes, wherein the nodes respectively represent recall conditions in different dimensions.

Based on the recall target guide map introduced above, optionally, behavior features in the game data of each user to be analyzed are extracted, and the recall target guide map matched with each user to be analyzed is determined according to the behavior features of each user to be analyzed. Therefore, the corresponding recall target guide picture is matched for each user to be analyzed, so that the recall analysis is more pertinent, the judgment accuracy of the user to be analyzed is further improved, and the omission of the user to be analyzed which is possibly recalled is avoided.

In the above steps, firstly, the behavior characteristics in the game data of each user to be analyzed are extracted through a machine learning model or a preset algorithm. For example, game data of each user to be analyzed is input into the classification prediction model, and behavior feature labels of each user to be analyzed are output through the classification prediction model.

Furthermore, in the above step, determining the recall target guidance chart matched with each user to be analyzed according to the behavior characteristics of each user to be analyzed may be implemented as:

according to the behavior characteristics of each user to be analyzed, inquiring a guide map branch or node corresponding to the behavior characteristics of each user to be analyzed in a preset standard recall target guide map; and establishing a recall target guide graph matched with each user to be analyzed based on the inquired guide graph branches or nodes.

In the preset standard recall target guide graph, the plurality of nodes comprise root nodes, middle nodes and leaf nodes. The next level node (i.e. the intermediate node) corresponding to the root node comprises recall policy nodes, and each recall policy node is used for indicating a user churn reason or a user backflow reason respectively. In fact, the intermediate nodes can also be subdivided into multiple layers, so that a more targeted recall analysis strategy is provided. Optionally, the recall policy node comprises an attrition recall policy node for indicating the cause of attrition of the user and/or a backflow maintenance policy node for indicating the cause of backflow of the user. Each recall strategy node corresponds to at least one leaf node, and the leaf nodes are used for indicating game resources corresponding to the user loss reasons or the user backflow reasons.

Based on the above example, in the above step, the guidance map branches or nodes matched with the behavior feature labels are queried from the preset standard recall target guidance map through the behavior feature labels of the users to be analyzed, and the guidance map branches or nodes are used as a plurality of nodes corresponding to the behavior features of the users to be analyzed. And then, constructing a recall target guide graph matched with each user to be analyzed based on the plurality of inquired nodes.

In practical applications, the recall target map determined by the above steps and matched by each user to be analyzed may be a part or all of branches of a standard recall target map.

For example, in the standard recall target graph, the cause of user churn in the churn recall policy node includes, but is not limited to, one or a combination of the following: insufficient version content, severe churn, poor interactive experience, poor gaming experience, hardware problems, unexpected games, regular churn. Poor game experience is, for example, poor pay experience, poor value experience, severe combat faults (game ecology imbalance). Hardware problems are, for example, unstable server environment, unsmooth operation, stuttering, flash back, too high memory occupancy rate, and serious equipment heating. Game non-compliance is, for example, advertising being very different from games and the game being very different from the design prototype experience. Routine churn is, for example, in-game item trading problems, game account problems, lost friends, no time to play a game. The reason for user reflow in the reflow maintenance policy node includes, but is not limited to, one or a combination of the following: version content updates (e.g., new profession, new play, new game system), reflow benefits (e.g., reflow gift bags, recharge benefits, reflow game privileges, etc.).

Based on the above example, in the above steps, it is assumed that the behavior feature tag of the user a to be analyzed is that the number of active friends is small, and the game is flashed back. Based on this, through the behavior feature tag of the user a to be analyzed, in the preset standard recall target guide map, the guide map branch corresponding to the behavior feature tag of the user a to be analyzed is queried, such as hardware problem, flash back, conventional loss and friend loss. And then, a recall target guide map matched with the user a to be analyzed is established based on the guide map branches, namely the recall target guide map a' consisting of the two branches.

Furthermore, after the recall target guidance chart matched with each user to be analyzed is determined, the game data is input into the recall analysis model in 102, so that the recall analysis model calculates the user to be analyzed based on the reference user to obtain the similarity of the user to be analyzed and the reference user in the game performance, and the method can also be realized as follows:

determining a reference user which accords with the recall target guide picture matched with each user to be analyzed in the plurality of users as the reference user matched with each user to be analyzed; and inputting the behavior characteristics of each user to be analyzed, the recall target guide picture matched with each user to be analyzed and the game data of the reference user into a recall analysis model, so that the recall analysis model calculates and obtains the similarity of each user to be analyzed and the game performance of the matched reference user.

Continuing with the above example, the reference user in accordance with the recall target guide map a' in the plurality of users is determined as the reference user matched with the user a to be analyzed. And then, inputting the behavior characteristics of the user a to be analyzed, the recall target guide map a' matched with the user to be analyzed and the game data of the reference user into a recall analysis model, so that the recall analysis model calculates and obtains the similarity of the user a to be analyzed and the matched reference user on the game performance.

Therefore, by introducing the recall target guide diagram, more targeted reference users can be screened in advance, the analysis process of the recall analysis model is assisted, and more accurate recall analysis results are provided for the users to be analyzed.

Optionally, the recall target guide map matched with each user to be analyzed further includes game resources corresponding to the recall targets. If the user to be analyzed is determined to be the target recall user, determining a recall target guide map matched with each user to be analyzed according to the behavior characteristics of each user to be analyzed, and then determining a recall strategy node with the highest matching degree from the matched recall target guide map according to the behavior characteristics of the target recall user; and taking the game resource indicated in the recall strategy node as the game resource corresponding to the recall target.

Continuing with the above example, the recall target guide map a' further includes game resources (i.e. leaf nodes) corresponding to the above two branches, hardware problem-flash back-version update package, regular churn-friend churn-new friend match.

In practical applications, the game resources corresponding to the attrition recall policy node include, but are not limited to, one or a combination of targeted recall information, call back, task designation, benefit charging, and game area designation. For example, the game resources corresponding to the reflow maintenance policy node include, but are not limited to, one or a combination of benefits activities, bonus packages, designated tasks, top-up benefits, designated game areas. It should be noted that the reflow maintenance policy node can also be used for making a game resource issuing decision after the target recall user logs in the game again.

The targeted recall information includes, for example, advertisements, short messages, and push information targeted to the targeted recall user. Optionally, the directional recall information carries a return gift package activation mode for improving recall possibility. In practical applications, the content of the targeted recall information may be set according to the correlation between the historical game data and the recall target (see the above embodiment). Furthermore, the format and the number of words of the directional recall information are preset, so that the automatic generation of the auxiliary directional recall information is realized, the simplicity of the file is ensured, the selling is avoided, and the user enjoyment is further improved.

Optionally, the recall phone content is determined based on a correlation between historical game data and recall targets. For example, based on the correlation between historical game data and recall objectives, game items or tasks preferred by the targeted recall user are determined and the user is prompted for relevant content in the recall phone. The related content is, for example, game items that give a preference to the user, or follow-up tasks that update the user's preference, or call the target recall user in a career that the user likes to select.

Furthermore, voice information in the call-back telephone can be generated by adopting the preferred character sound of the user so as to hook up the game experience memory of the user and improve the interest of the user. Alternatively, the voice information in the return call can be generated by using the voice material with higher historical recall rate (such as the voice packet of the character).

The user recall means of one or more embodiments of the present invention will be described in detail below. Those skilled in the art will appreciate that these user recall means may be constructed using commercially available hardware components configured by the steps taught in the present scheme.

Fig. 3 is a schematic structural diagram of a user recall device according to an embodiment of the present invention, and as shown in fig. 3, the user recall device includes: the device comprises an acquisition module 11, an analysis module 12 and a pushing module 13.

The acquisition module 11 is configured to acquire game data of a target object, where the game data includes game system data of the target object in a first time period;

the analysis module 12 is configured to input the game data into the recall analysis model, so that the recall analysis model calculates, based on the reference user, a user to be analyzed, and obtains similarity between the user to be analyzed and the reference user in game performance;

and the pushing module 13 is configured to determine a user to be analyzed, of which the similarity meets a preset condition, as a target recall user, and push a game resource corresponding to a recall target to the target recall user.

Optionally, the apparatus further comprises a determining module configured to:

extracting historical game data generated before the reference target is met from the game data of the plurality of users, wherein the historical game data comprises: first game data generated before the reference user satisfies the benchmark target, and second game data generated before a user who does not satisfy the recall target satisfies the benchmark target; respectively marking the first game data and the second game data by adopting different numerical values to obtain marked historical game data; inputting the marked historical game data into a correlation analysis model, so that the correlation analysis model calculates the correlation between the historical game data and the recall target; determining the game resource based on a correlation between the historical game data and the recall target.

Optionally, if there are a plurality of historical game data, the determining module is specifically configured to, when determining the game resource based on the correlation between the historical game data and the recall target:

arranging the historical game data from large to small according to the relevance of the historical game data and the recall target, and acquiring the data types of the historical game data in the previous target position; and taking the game resource corresponding to the acquired data type as the game resource.

Wherein the game data of the historical game data comprises any one or a combination of the following: user grade, login days, friend number, friend activeness, recharging condition, game system data, prop use condition, weapon use condition, battle force change condition and copy level.

Optionally, the recall analysis model is a machine learning model for similarity calculation.

The analysis module 12 inputs the game data into a recall analysis model, so that when the recall analysis model calculates the user to be analyzed based on the reference user, and obtains the similarity of the user to be analyzed and the reference user in game performance, the recall analysis model is specifically configured to:

selecting first game data generated before the reference user meets the reference target from the game data of the reference user;

selecting third game data generated before the user to be analyzed meets the reference target from the game data of the user to be analyzed;

and calculating the similarity between the third game data and the first game data through the machine learning model for similarity calculation, and taking the calculated similarity between the third game data and the first game data as the similarity of the user to be analyzed and the reference user in game performance.

Optionally, if the recall analysis model is provided with a plurality of recall targets, the apparatus further includes a preprocessing module configured to:

and dividing the plurality of users into a plurality of reference user groups and user groups to be analyzed corresponding to the plurality of reference user groups respectively based on the plurality of recall targets.

Optionally, each user group to be analyzed corresponds to each recall target one to one.

Before the pushing module 13 pushes the game resource corresponding to the recall target to the target recall user, the preprocessing module is further configured to:

judging whether the same target recall user exists in different target recall user groups or not; and if so, performing re-ranking processing on the target recall user groups corresponding to all the user groups to be analyzed based on the difficulty level of the recall target, so that the same target recall users do not exist in different target recall user groups.

Optionally, the preset condition includes any one or a combination of the following: the similarity is not less than a similarity threshold; the orders obtained based on the similarity sorting belong to preset target orders.

The user recall device shown in fig. 3 may execute the methods provided in the foregoing embodiments, and portions not described in detail in this embodiment may refer to the related descriptions of the foregoing embodiments, which are not described herein again.

In one possible design, the structure of the user recall device shown in fig. 3 and described above may be implemented as an electronic device. As shown in fig. 4, the electronic device may include: a processor 21 and a memory 22. Wherein the memory 22 has stored thereon executable code which, when executed by the processor 21, at least enables the processor 21 to implement a user recall method as provided in the preceding embodiments.

The electronic device may further include a communication interface 23 for communicating with other devices or a communication network.

In addition, an embodiment of the present invention provides a non-transitory machine-readable storage medium having stored thereon executable code, which, when executed by a processor of an electronic device, causes the processor to execute the user recall method provided in the foregoing embodiments.

The above-described apparatus embodiments are merely illustrative, wherein the various modules illustrated as separate components may or may not be physically separate. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.

Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by adding a necessary general hardware platform, and of course, can also be implemented by a combination of hardware and software. With this understanding in mind, the above-described aspects and portions of the present technology which contribute substantially or in part to the prior art may be embodied in the form of a computer program product, which may be embodied on one or more computer-usable storage media having computer-usable program code embodied therein, including without limitation disk storage, CD-ROM, optical storage, and the like.

Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims

1. A user recall method, comprising:

inputting the game data into a recall analysis model so that the recall analysis model calculates the user to be analyzed based on the reference user to obtain the similarity of the user to be analyzed and the reference user in game performance;

2. The method of claim 1, further comprising:

extracting historical game data generated before the reference target is met from the game data of the plurality of users, wherein the historical game data comprises: first game data generated before the reference user satisfies the benchmark target, and second game data generated before a user who does not satisfy the recall target satisfies the benchmark target;

respectively marking the first game data and the second game data by adopting different numerical values to obtain marked historical game data;

inputting the marked historical game data into a correlation analysis model, so that the correlation analysis model calculates the correlation between the historical game data and the recall target;

determining the game resource based on a correlation between the historical game data and the recall target.

3. The method of claim 2, wherein if there are a plurality of the historical game data, then

The determining the game resource based on the correlation of the historical game data to the recall target comprises:

arranging the historical game data from large to small according to the relevance of the historical game data and the recall target, and acquiring the data types of the historical game data in the previous target position;

and taking the game resource corresponding to the acquired data type as the game resource.

4. A method according to claim 2 or 3, wherein the game data of the historical game data comprises any one or a combination of:

user grade, login days, friend number, friend activeness, recharging condition, game system data, prop use condition, weapon use condition, battle force change condition and copy level.

5. The method of claim 1, wherein the recall analysis model is a machine learning model for similarity calculation;

the inputting the game data into a recall analysis model so that the recall analysis model calculates the user to be analyzed based on the reference user to obtain the similarity of the user to be analyzed and the reference user in game performance comprises:

6. The method of claim 1, wherein if the recall analysis model is provided with a plurality of recall targets, the method further comprises:

7. The method of claim 6, wherein each user group to be analyzed is in one-to-one correspondence with each recall target;

before the pushing the game resource corresponding to the recall target to the target recall user, the method further includes:

judging whether the same target recall user exists in different target recall user groups or not;

and if so, performing re-ranking processing on the target recall user groups corresponding to all the user groups to be analyzed based on the difficulty level of the recall target, so that the same target recall users do not exist in different target recall user groups.

8. The method of claim 1, further comprising:

extracting behavior characteristics in game data of each user to be analyzed;

according to the behavior characteristics of each user to be analyzed, determining a recall target guide map matched with each user to be analyzed;

the recall target guide graph comprises a tree-shaped hierarchical structure divided based on recall targets and a plurality of nodes, wherein the nodes respectively represent recall conditions under different dimensions.

9. The method of claim 8, wherein the inputting the game data into a recall analysis model to enable the recall analysis model to calculate the similarity of the user to be analyzed and the reference user in game performance based on the reference user comprises:

determining the reference users which accord with the recall target guide picture matched with each user to be analyzed in the plurality of users as the reference users matched with each user to be analyzed;

and inputting the behavior characteristics of each user to be analyzed, the recall target guide picture matched with each user to be analyzed and the game data of the reference user into the recall analysis model, so that the recall analysis model calculates to obtain the similarity of each user to be analyzed and the game performance of the matched reference user.

10. The method according to claim 8, wherein the determining the recall target guide map matched with each user to be analyzed according to the behavior characteristics of each user to be analyzed comprises:

according to the behavior characteristics of each user to be analyzed, inquiring a guide map branch or node corresponding to the behavior characteristics of each user to be analyzed in a preset standard recall target guide map;

and establishing a recall target guide graph matched with each user to be analyzed based on the inquired guide graph branches or nodes.

11. The method of claim 8, wherein the recall target guide map matched by each user to be analyzed further comprises game resources corresponding to the recall targets;

if it is determined that the user to be analyzed is the target recall user, after determining the recall target guide map matched with each user to be analyzed according to the behavior characteristics of each user to be analyzed, the method further comprises the following steps:

according to the behavior characteristics of the target recall users, recall strategy nodes with the highest matching degree are determined from the matched recall target guide graphs;

and taking the game resource indicated in the recall strategy node as the game resource corresponding to the recall target.

12. The method according to claim 11, wherein the recall policy node comprises an churn recall policy node for indicating a cause of churn for a user and/or a backflow maintenance policy node for indicating a cause of backflow for a user;

the game resources corresponding to the loss recall strategy node comprise one or a combination of directional recall information, a recall telephone, a designated task, a recharge welfare and a designated game area;

the game resources corresponding to the reflow maintenance strategy node comprise one or a combination of welfare activities, reward gift bags, designated tasks, recharging welfare and designated game areas.

13. The method according to any one of claims 1 to 12, wherein the preset conditions include any one or a combination of:

the similarity is not less than a similarity threshold;

the orders obtained based on the similarity sorting belong to preset target orders.

14. A user recall apparatus, comprising:

the system comprises an acquisition module, a retrieval module and a processing module, wherein the acquisition module is used for acquiring game data of a plurality of users, users meeting recall targets in the plurality of users are used as reference users, and users meeting reference targets and not meeting the recall targets in the plurality of users are used as users to be analyzed;

the analysis module is used for inputting the game data into a recall analysis model so that the recall analysis model can calculate the user to be analyzed based on the reference user to obtain the similarity of the user to be analyzed and the reference user in game performance;

15. An electronic device, comprising: a memory, a processor; wherein the memory has stored thereon executable code which, when executed by the processor, causes the processor to perform the user recall method of any of claims 1 to 13.