Game recommendation method and device, electronic equipment and storage medium

Abstract

The invention discloses a game recommendation method which can be applied to the field of artificial intelligence. According to the method, account behavior information related to a target game is converted into text information, the text information is aggregated to generate an associated game document of the target game, and the associated game document is generated based on the relation between an associated account related to the target game and the associated game. According to the method, the association account is used as a bridge to bear the relationship between the target game and the association game, and the relationship information between the games is completely migrated into the document structure, so that the analysis result of the keyword analysis technology for the document can indicate the association tendency between the games; and then a game recommendation method is obtained by improving a keyword analysis technology and combining particularity design of game recommendation. The game recommendation accuracy can be remarkably improved, the game experience is improved, game information resource spreading is facilitated, the game information resource spreading efficiency is improved, and various information interaction and information spreading with games as carriers are enhanced.

Claims

1. A game recommendation method, It is characterized in that The method comprises: In the case of obtaining a game recommendation request for a target account, obtaining an associated game document corresponding to a target game pointed to by the target account, wherein the associated game document includes an associated game having a first preset operation relationship with the associated account, and the associated account is an account having a second preset operation relationship with the target game; Determining a recommendation index corresponding to each associated game in the associated game document, wherein the recommendation index is an index obtained by performing a document keyword analysis on the associated game document and characterizing a recommendation degree of the corresponding associated game; Game recommendations are made according to the recommendation indicators corresponding to the related games.

2. The method according to claim 1, It is characterized in that The determining of the recommendation index corresponding to each associated game in the associated game document includes: For each of the related games, determining the frequency of occurrence of the related game in the related game document; Determine the number of execution accounts corresponding to the associated game, where the execution account is an account that performs a third preset operation on the associated game; Based on the frequency and the number of the execution accounts, a recommendation index corresponding to the associated game is determined.

3. The method according to claim 1 or 2, It is characterized in that The associated game document is constructed by the following method: Determining an associated account for performing the second preset operation on the target game; For each of the associated accounts, the following operations are performed: extracting game behavior information, where the game behavior information is behavior information generated by the associated account performing the first preset operation; Extracting each game identifier in the game behavior information to obtain a game identifier set corresponding to the associated account; The entire set of game identification sets corresponding to each of the associated accounts is recorded to obtain the associated game document.

4. The method according to claim 2, It is characterized in that The determining, based on the frequency and the number of the execution accounts, a recommendation indicator corresponding to the associated game includes: Determine at least one of the following optimization parameters: a balance factor, a random factor, and a smoothing factor, wherein the balance factor is used to optimize the recommendation index when the number of corresponding execution accounts is higher than a first threshold, the random factor is used to output a diversified game recommendation list in a scenario where a game recommendation list is generated, and the smoothing factor is used to optimize the recommendation index when the number of corresponding execution accounts is lower than a second threshold; The determining the recommendation index corresponding to the associated game based on the frequency and the number of the execution accounts includes: determining the recommendation index corresponding to the associated game based on the frequency, the number of the execution accounts and the optimization parameter.

5. The method according to claim 4, It is characterized in that The optimization parameters include the balance factor, the smoothing factor and the random factor, and the determination of the recommendation index corresponding to the associated game based on the frequency, the number of the execution accounts and the optimization parameters includes: Determining a first target parameter according to the number of the execution accounts and the sum of the random factor; Taking the first target parameter as a base and the balance factor as an exponent, a second target parameter is obtained; Determining a third target parameter based on the sum of the second target parameter and the smoothing factor; The recommendation index is determined based on the ratio of the frequency to the third target parameter.

6. The method according to claim 1, It is characterized in that The making game recommendations according to the recommendation indicators respectively corresponding to the associated games includes: Arrange the recommendation indicators corresponding to the associated games in descending order to obtain a game recommendation list including a preset number of associated games; The game recommendation list is displayed.

7. The method according to claim 1, It is characterized in that The first preset operation includes at least one of the following: downloading, registering, and logging in; The second preset operation includes at least one of the following: search, download, register, and log in; The third preset operation includes at least one of the following: downloading, registering, and logging in.

8. A game recommendation device, It is characterized in that The device comprises: a request acquisition module, configured to, upon obtaining a game recommendation request for a target account, obtain an associated game document corresponding to a target game pointed to by the target account, wherein the associated game document includes an associated game having a first preset operation relationship with the associated account, and the associated account is an account having a second preset operation relationship with the target game; A recommendation module, used to determine a recommendation index corresponding to each associated game in the associated game document, wherein the recommendation index is an index obtained by performing a document keyword analysis on the associated game document and characterizing the recommendation degree of the corresponding associated game; Game recommendations are made according to the recommendation indicators corresponding to the related games.

9. A computer device, It is characterized in that The computer device includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the game recommendation method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, It is characterized in that The storage medium stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the game recommendation method as described in any one of claims 1 to 7.

Images