CN-121971857-A - Data processing method, device, equipment and readable storage medium

CN121971857ACN 121971857 ACN121971857 ACN 121971857ACN-121971857-A

Abstract

Acquiring game progress information corresponding to an object virtual character controlled by an object in a game world, if the game progress information meets a state triggering condition, acquiring fact information associated with the game progress information in a fact knowledge base of the game world, and generating character state information according to the fact information and the game progress information; the method comprises the steps of inputting character state information into a search model, generating target character configuration data through the search model, obtaining historical memory behavior data, generating latest interaction data according to the target character configuration data and the historical memory behavior data, generating target fixed virtual characters according to the target character configuration data and the latest interaction data, and putting the target fixed virtual characters into a game environment associated with game progress information. By adopting the method and the device, the interactive content of the fixed virtual role can be updated efficiently and the personalized interactive service can be realized.

Inventors

ZHANG HAOYANG

Assignees

深圳奥拓盖母科技有限公司

Dates

Publication Date: 20260505
Application Date: 20260327

Claims (15)

1. A method of data processing, comprising: Acquiring game progress information corresponding to an object virtual role controlled by an object in a game world, if the game progress information meets a state triggering condition, acquiring fact information associated with the game progress information in a fact knowledge base of the game world, and generating role state information corresponding to the object virtual role according to the fact information and the game progress information; Inputting the character state information into a search model, and generating target character configuration data associated with the character state information through the search model, wherein the fact information in the character state information is used for restraining the target character configuration data in a progress range indicated by the game progress information, and the target character configuration data is used for representing character attributes corresponding to fixed virtual characters in the game world; Acquiring historical memory behavior data corresponding to the object virtual character, and generating latest interaction data for interacting with the object virtual character according to the target character configuration data and the historical memory behavior data, wherein the historical memory behavior data refers to historical interaction data between the object virtual character and a historical fixed virtual character associated with the target character configuration data in the historical game progress of the game world; and generating a target fixed virtual role for pushing the game progress associated with the target virtual role according to the target role configuration data and the latest interaction data, and putting the target fixed virtual role into the game environment associated with the game progress information.
2. The method of claim 1, wherein the game progress information includes style update information, scenario progress information, activity progress information, and duration progress information, the method further comprising: And if the style updating information indicates that the game virtual style of the object virtual character is updated to be the target style, or the scenario progress information indicates that the object virtual character is in a target scenario node, or the activity progress information indicates that the object virtual character participates in target activity, or the duration progress information indicates that the character creation duration of the object virtual character is less than or equal to a duration threshold value, determining that the game progress information meets a state triggering condition.
3. The method of claim 1, wherein the game progress information includes style update information, scenario progress information, and activity progress information, wherein the generating character status information corresponding to the object virtual character according to the fact information and the game progress information comprises: Acquiring style updating place information of the style updating information in the game world, extracting node place information of a target scenario node where the object virtual character is located in the game world from the scenario progress information, and determining the style updating place information and the node place information as character place information corresponding to the object virtual character; acquiring character attribute information corresponding to the object virtual character, and extracting the participation state of the object virtual character in relation to target activities from the activity progress information; And combining the target scenario node, the role place information, the role attribute information, the participation state and the fact information into role state information corresponding to the object virtual role.
4. The method of claim 1, wherein the inputting the character state information into a search model, generating target character configuration data associated with the character state information by the search model, comprises: Inputting the character state information into a retrieval model, and extracting features of the character state information through the retrieval model to obtain character state features, wherein the character state features comprise fact features corresponding to the fact information; Acquiring one or more pieces of fixed state information of a fixed virtual character in the game world through the retrieval model, and extracting the characteristics of the one or more pieces of fixed state information to obtain fixed state characteristics respectively corresponding to the one or more pieces of fixed state information; Respectively carrying out similarity matching on the character state characteristics and one or more fixed state characteristics to obtain state similarity corresponding to the one or more fixed state characteristics, and determining fixed state information, of the one or more fixed state information, with the state similarity being greater than or equal to a state similarity threshold, as target fixed state information; generating a fact constraint condition according to the fact feature, and carrying out constraint processing on one or more initial role configuration data associated with the target fixed state information based on the fact constraint condition to obtain target role configuration data associated with the role state information.
5. The method of claim 4, wherein the fact constraint comprises a time constraint, a relationship constraint, and a place constraint, wherein the one or more initial character configuration data comprises S initial character configuration data, wherein the S initial character configuration data comprises initial character configuration data S i , wherein S is a positive integer, wherein i is a positive integer less than or equal to S, wherein the constraint processing is performed on the one or more initial character configuration data associated with the target fixed state information based on the fact constraint to obtain target character configuration data associated with the character state information, and wherein the constraint processing comprises: The method comprises the steps of acquiring a triggerable event set, a relation network and a knowledge boundary in initial role configuration data S i , wherein the triggerable event set comprises event occurrence time corresponding to one or more triggerable events respectively, the relation network is used for representing social relations between a fixed virtual role and an object virtual role in the game world, and the knowledge boundary is used for representing an interactable data range between the fixed virtual role and the object virtual role; And if one or more event occurrence times are all before the node time of the target scenario node indicated by the time constraint condition, and the relationship network does not include a virtual relationship which is not associated with the target virtual character and is indicated by the relationship constraint condition, and one or more game location information included by the knowledge boundary belongs to the character location information indicated by the location constraint condition, determining the initial character configuration data S i as target character configuration data associated with the character state information.
6. The method of claim 1, wherein the obtaining historic memory behavior data corresponding to the object avatar comprises: acquiring an object identifier of the object virtual character, and acquiring a fixed identifier of a history fixed virtual character associated with the target character configuration data; The method comprises the steps of obtaining historical memory behavior data associated with an object virtual role and a historical fixed virtual role in a memory database according to the object identifier and the fixed identifier, wherein the memory database comprises a fixed number of one or more candidate memory behavior data, the memory database is further used for deleting one or more candidate memory behavior data with a data storage time length larger than a storage threshold value, the data storage time length refers to the storage time length of the one or more candidate memory behavior data in the memory database, and the historical memory behavior data refers to the candidate memory behavior data with a data storage time length smaller than or equal to the storage threshold value.
7. The method of claim 1, wherein generating the latest interaction data for interacting with the object avatar from the target persona configuration data and the historic memory behavior data comprises: generating a prompt word for indicating a large language model to generate interaction data according to the target role configuration data and the historical memory behavior data; Inputting the prompt word into a large language model, extracting character configuration data characteristics corresponding to the target character configuration data and memory behavior data characteristics corresponding to the history memory behavior data through the large language model, wherein the character configuration data characteristics are used for representing association relations among one or more data fields of the target character configuration data; Feature fusion is carried out on the character configuration data features and the memory behavior data features to obtain character fusion features, and dialogue interaction data and behavior interaction data for interacting with the virtual character of the object are generated according to the character fusion features; And determining the dialogue interaction data and the behavior interaction data as the latest interaction data.
8. The method of claim 7, wherein the character fusion features include a dialogue fusion feature and a behavior fusion feature, wherein the feature fusing the character configuration data feature and the memory behavior data feature to obtain a character fusion feature, and generating dialogue interaction data and behavior interaction data for interacting with the object virtual character according to the character fusion feature comprises: the method comprises the steps of obtaining a first gating vector associated with a dialogue mode from the character configuration data characteristics, and generating a dialogue fusion characteristic according to the first gating vector and the memory behavior data characteristics, wherein the first gating vector is used for enhancing the characteristics associated with the dialogue mode in the memory behavior data characteristics; Generating a candidate dialogue data set according to the dialogue fusion characteristics, acquiring data lengths respectively corresponding to one or more candidate dialogue data in the candidate dialogue data set, and determining the candidate dialogue data with the data length smaller than or equal to a length threshold value in the one or more candidate dialogue data as dialogue interaction data for interacting with the object virtual character; acquiring a second gating vector associated with a behavior mode from the character configuration data features, and generating the behavior fusion feature according to the second gating vector and the memory behavior data features, wherein the second gating vector is used for enhancing the features associated with the behavior mode in the memory behavior data features; Generating a role behavior instruction according to the behavior fusion characteristic, acquiring an associated behavior event from a triggerable event set of the target role configuration data according to the role behavior instruction, and combining the role behavior instruction and the associated behavior event into behavior interaction data for interacting with the object virtual role, wherein the role behavior instruction is used for indicating a fixed virtual role in the game world to execute the associated behavior event.
9. The method of claim 1, wherein the target character configuration data includes fixed character attribute information and character image identifications, wherein the latest interaction data includes dialogue interaction data and behavior interaction data, wherein the generating a target fixed virtual character for advancing a game progress associated with the object virtual character from the target character configuration data and the latest interaction data comprises: Acquiring a character three-dimensional model from a character knowledge base according to the character image identification, and performing model rendering on the character three-dimensional model according to the fixed character attribute information to obtain a target character three-dimensional model; Constructing dialogue interaction logic aiming at the target role three-dimensional model according to the dialogue interaction data, and constructing behavior interaction logic aiming at the target role three-dimensional model according to the behavior interaction data; and adding the dialogue interaction logic and the behavior interaction logic to the target character three-dimensional model to obtain a target fixed virtual character for propelling the game progress associated with the target virtual character.
10. The method of claim 1, wherein the most recent interaction data comprises conversational interaction data, the method further comprising: Obtaining dialogue throwing rules, carrying out feature extraction on the dialogue throwing rules to obtain throwing rule features, and carrying out feature extraction on the dialogue interaction data to obtain dialogue interaction features; Feature matching is carried out on the throwing rule features and the dialogue interaction features to obtain dialogue similarity between the dialogue throwing rules and the dialogue interaction data, and if the dialogue similarity is larger than or equal to a similarity threshold value, the target fixed virtual character is added to a throwing queue; And if the dialogue similarity is smaller than the similarity threshold, acquiring fixed dialogue data meeting the dialogue throwing rule, replacing dialogue interaction data in the latest interaction data with the fixed dialogue data to obtain updated interaction data, generating a default fixed virtual role according to the target role configuration data and the updated interaction data, and adding the default fixed virtual role to a throwing queue.
11. The method as recited in claim 1, further comprising: Acquiring interaction content between the object virtual character and the target fixed virtual character; Acquiring an interaction feedback rate, a dialogue completion rate and a scenario conflict rate between the interaction content and the game progress information aiming at the interaction content, and generating a role evaluation score corresponding to the target fixed virtual role according to the interaction feedback rate, the dialogue completion rate and the scenario conflict rate; if the role evaluation score is greater than or equal to an evaluation threshold, recording the interaction content as latest memory behavior data to a memory database corresponding to the target fixed virtual role; And if the character evaluation score is smaller than the evaluation threshold, stopping the throwing of the target fixed virtual character, and throwing the historical fixed virtual character into the game environment associated with the game progress information.
12. A data processing apparatus, comprising: The state information generation module is used for acquiring game progress information corresponding to an object virtual role controlled by an object in a game world, acquiring fact information associated with the game progress information in a fact knowledge base of the game world if the game progress information meets a state triggering condition, and generating role state information corresponding to the object virtual role according to the fact information and the game progress information; The system comprises a configuration data generation module, a character state information generation module, a configuration data generation module and a configuration data generation module, wherein the character state information is input into a retrieval model, and target character configuration data associated with the character state information is generated through the retrieval model; The interactive data generation module is used for acquiring historical memory behavior data corresponding to the object virtual character, and generating latest interactive data for interacting with the object virtual character according to the target character configuration data and the historical memory behavior data, wherein the historical memory behavior data refers to historical interactive data between the object virtual character and a historical fixed virtual character associated with the target character configuration data in the historical game progress of the game world; And the virtual character throwing module is used for generating a target fixed virtual character for propelling the game progress associated with the target virtual character according to the target character configuration data and the latest interaction data, and throwing the target fixed virtual character into the game environment associated with the game progress information.
13. A computer device comprises a processor, a memory, and a network interface; The processor is connected to the memory and the network interface, wherein the network interface is configured to provide a data communication function, the memory is configured to store a computer program, and the processor is configured to invoke the computer program to cause the computer device to perform the method of any of claims 1-11.
14. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a computer program adapted to be loaded and executed by a processor to cause a computer device having the processor to perform the method of any of claims 1-11.
15. A computer program product, characterized in that the computer program product comprises a computer program stored in a computer readable storage medium and adapted to be read and executed by a processor to cause a computer device with the processor to perform the method of any of claims 1-11.

Description

Data processing method, device, equipment and readable storage medium Technical Field The present application relates to the field of computer technologies, and in particular, to a data processing method, apparatus, device, and readable storage medium. Background An operator may develop a fixed virtual character in the game world that interacts with the object virtual character controlled by the user to be placed in the game environment in which the object virtual character is located. However, since the interactive contents of all users and the fixed virtual roles are the same, the presented interactive effect is easy to be single, and if the interactive contents of the fixed virtual roles are to be updated, operators are required to generate an update package by redevelopment of the interactive logic aiming at the fixed virtual roles, and then issue the update package to the users, so that the users experience new interactive contents by updating the games through the update package, the whole update process is complicated, and personalized interaction between the fixed virtual roles and the object virtual roles still cannot be realized. Disclosure of Invention The embodiment of the application provides a data processing method, a device, equipment and a readable storage medium, which can realize efficient updating and personalized interaction service of interaction content aiming at a fixed virtual role. In one aspect, an embodiment of the present application provides a data processing method, including: acquiring game progress information corresponding to an object virtual role controlled by an object in a game world, if the game progress information meets a state triggering condition, acquiring fact information associated with the game progress information in a fact knowledge base of the game world, and generating role state information corresponding to the object virtual role according to the fact information and the game progress information; Inputting character state information into a search model, and generating target character configuration data related to the character state information through the search model, wherein fact information in the character state information is used for restraining the target character configuration data in a progress range indicated by game progress information; The method comprises the steps of obtaining historical memory behavior data corresponding to an object virtual character, and generating latest interaction data for interacting with the object virtual character according to target character configuration data and the historical memory behavior data, wherein the historical memory behavior data refers to historical interaction data between historical fixed virtual characters of the object virtual character in the historical game progress of a game world and associated with the target character configuration data; And generating a target fixed virtual character for pushing the game progress associated with the target virtual character according to the target character configuration data and the latest interaction data, and putting the target fixed virtual character into the game environment associated with the game progress information. The game progress information comprises style updating information, scenario progress information, activity progress information and duration progress information, and the method further comprises the following steps: If the style updating information indicates that the game virtual style of the virtual character of the object is updated to be the target style, or the plot progress information indicates that the virtual character of the object is at the target plot node, or the activity progress information indicates that the virtual character of the object participates in the target activity, or the duration progress information indicates that the character creation duration of the virtual character of the object is less than or equal to the duration threshold value, then the game progress information is determined to meet the state triggering condition. The game progress information comprises style updating information, scenario progress information and activity progress information, and character state information corresponding to the virtual character of the object is generated according to the fact information and the game progress information, and the method comprises the following steps: Acquiring style updating place information of style updating information in a game world, extracting node place information of a target scenario node where an object virtual character is located in the game world from scenario progress information, and determining the style updating place information and the node place information as character place information corresponding to the object virtual character; Acquiring character attribute information corresponding to the virtual character of the object, and extracting the participation st