CN-115659016-B - Recall strategy screening method and related device

CN115659016BCN 115659016 BCN115659016 BCN 115659016BCN-115659016-B

Abstract

The application is suitable for the technical field of data processing, and provides a recall strategy screening method and a related device, which are used for screening a proper recall strategy for a specific user of an application program so as to improve the retrieval of users with low activity. The method mainly comprises the steps of determining the strategy number X of recall strategies in a recall strategy set, dividing a sample user group into X sub-sample user groups, wherein X is a positive integer greater than 0, pushing the recall strategies in the same recall strategy set to sample users in the same sub-sample user group aiming at the X sub-sample user groups, calculating the recall rate of each sub-sample user group according to the response user number of the recall strategies in the response recall strategy set in each sub-sample user group, screening out target recall strategies corresponding to target recall rates meeting preset standards, wherein the target recall rates are one or more of the recall strategies, and the target recall strategies are one or more of the recall strategies.

Inventors

WANG GUOBIN
LI JIANG
XIANG WEI
MA SONG

Assignees

土巴兔集团股份有限公司

Dates

Publication Date: 20260505
Application Date: 20221018

Claims (8)

1. A recall policy screening method, comprising: Determining the strategy number X of recall strategies in a recall strategy set, wherein X is a positive integer greater than 0, and the recall strategies are implementation plans for pushing specific contents to users; Dividing a sample user group into X sub-sample user groups, wherein the number of users of the sample user group is greater than or equal to X; pushing recall policies in the same recall policy set to sample users in the same sub-sample user group for X sub-sample user groups; Calculating the recall rate of each sub-sample user group according to the number of response users responding to the recall policies in the recall policy set in each sub-sample user group; Screening out target recall strategies corresponding to target recall rates meeting preset standards, wherein the target recall rates are one or more of the recall rates, and the target recall strategies are one or more of the recall strategies; after screening out the target recall strategies corresponding to the target recall rates meeting the preset standard, the method further comprises the following steps: executing the target recall strategy on all users in the user set of the same target level; Before executing the target recall policy for all users in the same target level set of users, the method further comprises: Calculating the natural recall rate of the rest users except the sample user in the user set with the same target level; Judging whether the target recall rate is greater than the natural recall rate; If the target recall rate is larger than the natural recall rate, triggering and executing the target recall strategy for all users in the user set of the same target level; And if the target recall rate is smaller than or equal to the natural recall rate, prompting to input a new recall strategy.
2. The recall policy screening method of claim 1, wherein prior to dividing the sample user population into X sub-sample user populations, the method further comprises: And extracting a specific number of sample users from the user set divided into the same target level as the sample user group.
3. The recall policy screening method of claim 2, wherein prior to extracting a particular number of the sample users from a set of users classified into the same target tier, the method further comprises: Acquiring user behavior data of each user in the whole user set of the application program in the latest time period; inputting the user behavior data of each user into a preset user loss prediction model to obtain the user loss probability of each user in the whole user set; And dividing each user in the whole user set into corresponding user loss grades according to a preset grade division standard to obtain Y user loss grades, wherein Y is a positive integer greater than 0, and the target grade is one of the Y user loss grades.
4. The recall policy screening method of claim 1 wherein calculating natural recall rates for remaining users of the same set of target levels excluding the sample user comprises: Acquiring the natural recall quantity of the natural recall users meeting the recall standard in the latest time period in the residual users; And calculating the natural recall rate, wherein the natural recall rate is equal to the natural recall number divided by the remaining user number of the remaining users.
5. The recall policy screening method of claim 1, wherein after executing the target recall policy on all users in the set of users of the same target level, the method further comprises: Calculating the overall recall rate of all users in the user set of the same target level; judging whether the overall recall rate is greater than the natural recall rate; If the overall recall rate is larger than the natural recall rate, the target recall strategy is used as a preferred recall strategy to be associated with the target grade for storage; And if the overall recall rate is equal to or smaller than the natural recall rate, stopping executing the target recall strategy on all users in the user set with the same target level.
6. A recall policy screening system, comprising: the determining unit is used for determining the strategy number X of the recall strategies in the recall strategy set, wherein X is a positive integer greater than 0, and the recall strategies are implementation plans for pushing specific contents to users; the dividing unit is used for dividing the sample user group into X sub-sample user groups, and the number of users of the sample user groups is greater than or equal to X; The pushing unit is used for pushing recall strategies in the same recall strategy set to sample users in the same sub-sample user group aiming at the X sub-sample user groups; A calculating unit, configured to calculate a recall rate of each sub-sample user group according to a number of response users in each sub-sample user group that respond to a recall policy in the recall policy set; the screening unit is used for screening out target recall strategies corresponding to target recall rates meeting preset standards, wherein the target recall rates are one or more of the recall rates, and the target recall strategies are one or more of the recall strategies; the system further comprises: The execution unit is used for executing the target recall strategy on all users in the user set of the same target level; the system further comprises: The computing unit is further used for computing the natural recall rate of the rest users except the sample user in the user set with the same target level; The judging unit is used for judging whether the target recall rate is larger than the natural recall rate; the triggering unit is used for triggering and executing the target recall strategy to all users in the user set of the same target level if the target recall rate is greater than the natural recall rate; and the reminding unit is used for reminding to input a new recall strategy if the target recall rate is smaller than or equal to the natural recall rate.
7. A computer device, comprising: a processor, a memory, a bus, an input-output interface, and a wireless network interface; the processor is connected with the memory, the input/output interface and the wireless network interface through buses; the memory stores a program; the recall policy screening method according to any one of claims 1 to 5 is implemented when the processor executes the program stored in the memory.
8. A computer readable storage medium having instructions stored therein, which when executed on a computer, cause the computer to perform the recall policy screening method of any one of claims 1 to 5.

Description

Recall strategy screening method and related device Technical Field The application belongs to the technical field of data processing, and particularly relates to a recall strategy screening method and a related device. Background For an internet service provider of an application on the internet, the longer a user invests in an application that he operates, the more dependent he proves the user is on his application, the higher the probability that the internet service provider will obtain potential benefits from the user. For example, a user who spends a longer time in an application will generally be more interested in the paid services of the application, the higher the probability of a transaction that the application will promote the user to consume their paid services. Internet service providers often need to actively push some content to the user that can increase the user's interest in order to increase the user's viscosity to their operating application, for the purpose of longer time the user puts on their operating application. Recall strategies are effective means for retrieving users with low liveness and increasing the duration of stay of the user in the application as an implementation plan for pushing specific content to the user. However, if the internet service provider does not actively execute the recall strategy for the user with low activity, the application program of the internet service provider can fade out the life of the user slowly, the dependence of the user on the application program of the internet service provider is reduced, and the success probability of the user consuming the payment service of the internet service provider is gradually reduced. Therefore, screening the appropriate recall strategy for the specific user becomes a technical problem to be solved. Disclosure of Invention The application aims to provide a recall strategy screening method and a related device, which aim to screen out a proper recall strategy for a specific user of an application program so as to improve the probability of retrieving users with low liveness and increasing the stay time of the users in the application program. In a first aspect, the present application provides a recall policy screening method, including: Determining the strategy number X of recall strategies in a recall strategy set, wherein X is a positive integer greater than 0, and the recall strategies are implementation plans for pushing specific contents to users; Dividing a sample user group into X sub-sample user groups, wherein the number of users of the sample user group is greater than or equal to X; pushing recall policies in the same recall policy set to sample users in the same sub-sample user group for X sub-sample user groups; Calculating the recall rate of each sub-sample user group according to the number of response users responding to the recall policies in the recall policy set in each sub-sample user group; Screening out target recall strategies corresponding to target recall rates meeting preset standards, wherein the target recall rates are one or more of the recall rates, and the target recall strategies are one or more of the recall strategies. Optionally, before dividing the sample user group into the X sub-sample user groups, the method further comprises: And extracting a specific number of sample users from the user set divided into the same target level as the sample user group. Optionally, before extracting a specific number of the sample users from the user set divided into the same target level, the method further includes: Acquiring user behavior data of each user in the whole user set of the application program in the latest time period; inputting the user behavior data of each user into a preset user loss prediction model to obtain the user loss probability of each user in the whole user set; And dividing each user in the whole user set into corresponding user loss grades according to a preset grade division standard to obtain Y user loss grades, wherein Y is a positive integer greater than 0, and the target grade is one of the Y user loss grades. Optionally, after screening out the target recall strategy corresponding to the target recall rate meeting the preset standard, the method further includes: And executing the target recall strategy on all users in the user set with the same target grade. Optionally, before executing the target recall policy on all users in the same target level user set, the method further includes: Calculating the natural recall rate of the rest users except the sample user in the user set with the same target level; Judging whether the target recall rate is greater than the natural recall rate; If the target recall rate is larger than the natural recall rate, triggering and executing the target recall strategy for all users in the user set of the same target level; And if the target recall rate is smaller than or equal to the natural recall rate, prompting to inpu