CN-121988039-A - Data processing method, device, equipment and readable storage medium

CN121988039ACN 121988039 ACN121988039 ACN 121988039ACN-121988039-A

Abstract

The application discloses a data processing method, a device, equipment and a readable storage medium, wherein the method comprises the steps of obtaining role information, game progress information and game service constraint information corresponding to a game virtual role, and determining the role information and the game progress information as game state data; the game state data is input into a search model, historical game configuration data is generated through the search model, the game state data, game service constraint information and the historical game configuration data are input into a large language model, a state configuration data set is output through the large language model, prop intensity constraint information is generated according to the game state data, constraint processing is conducted on the state configuration data set based on the prop intensity constraint information to obtain target state configuration data, game virtual props are generated based on the target state configuration data, and the game virtual props are put into game accounts corresponding to the game virtual roles. By adopting the method and the device, the updating efficiency and the updating accuracy for the game props can be improved.

Inventors

ZHANG HAOYANG

Assignees

深圳奥拓盖母科技有限公司

Dates

Publication Date: 20260508
Application Date: 20260327

Claims (15)

1. A method of data processing, comprising: Acquiring role information, game progress information and game service constraint information corresponding to a game virtual role in a game world, and determining the role information and the game progress information as game state data corresponding to the game virtual role; Inputting the game state data into a retrieval model, generating historical game configuration data associated with the game state data through the retrieval model, inputting the game state data, the game service constraint information and the historical game configuration data into a large language model, and outputting a state configuration data set corresponding to the game state data through the large language model, wherein the state configuration data set comprises one or more state configuration data meeting the game service constraint information; generating prop intensity constraint information according to the game state data, and carrying out constraint processing on the state configuration data set based on the prop intensity constraint information to obtain target state configuration data; And generating a game virtual prop for enhancing the attribute intensity corresponding to the game virtual character based on the target state configuration data, and throwing the game virtual prop into a game account corresponding to the game virtual character.
2. The method as recited in claim 1, further comprising: Acquiring the stay time of the game virtual character at a stay gate, the number of remaining clearance steps and the key blocking number, wherein the key blocking number refers to the number of unblocked blocking of the game virtual character at the stay gate; if the stay time length is greater than or equal to a time length threshold, or the remaining clearance steps are smaller than a step number threshold, or the key blocking number is greater than or equal to a number threshold, acquiring a checkpoint identifier, checkpoint type information and scenario node information of the stay checkpoint; And determining the gate identification, the gate type information and the scenario node information as game progress information corresponding to the game virtual character.
3. The method of claim 1, wherein the inputting the game state data into a search model, generating historical game configuration data associated with the game state data by the search model, comprises: Inputting the game state data into a retrieval model, and extracting features of the game state data through the retrieval model to obtain game state features, wherein the game state features comprise game progress features corresponding to the game progress information; acquiring one or more game level information associated with the game world in a knowledge base through the retrieval model, and extracting features of the one or more game level information to obtain game level features respectively corresponding to the one or more game level information; Performing similarity matching on the game progress features and one or more game level features respectively to obtain level similarity corresponding to the one or more game level features, and determining game level information, in the one or more game level information, with the level similarity being greater than or equal to a level similarity threshold, as target game level information; Extracting features of one or more initial state configuration data associated with the target game level information to obtain initial configuration features respectively corresponding to the one or more initial state configuration data; and respectively carrying out similarity matching on the game state characteristics and one or more initial configuration characteristics to obtain configuration similarity corresponding to the one or more initial configuration characteristics, and determining initial state configuration data, of the one or more initial state configuration data, the configuration similarity is greater than or equal to a configuration similarity threshold value as historical game configuration data associated with the game state data.
4. The method of claim 1, wherein the inputting the game state data, the game service constraint information, and the historical game configuration data into a large language model, outputting a state configuration data set corresponding to the game state data through the large language model, comprises: Generating a configuration prompt word for indicating a large language model to perform configuration generation according to the game state data, the game service constraint information and the historical game configuration data; The configuration prompt word is input into the large language model, the game state characteristics corresponding to the game state data in the configuration prompt word are extracted through the large language model, the configuration semantic characteristics and the configuration range characteristics corresponding to the historical game configuration data are extracted, the game state characteristics comprise the role characteristics corresponding to the role information and the game progress characteristics corresponding to the game progress information, the configuration semantic characteristics are used for representing the association relation between one or more data fields of the historical game configuration data, and the configuration range characteristics are used for reflecting the content range of the one or more data fields; Feature fusion is carried out on the configuration semantic features and the character features to obtain character fusion features, and a character configuration data set corresponding to the character information is generated according to the character fusion features; Performing feature fusion on the configuration range features and the game progress features to obtain progress fusion features, and generating a progress configuration data set corresponding to the game progress information according to the progress fusion features; generating a role configuration threshold condition for constraining the role configuration data set and a progress configuration threshold condition for constraining the progress configuration data set according to the game service constraint information, acquiring a target role configuration data set meeting the role configuration threshold condition from the role configuration data set, and acquiring a target progress configuration data set meeting the progress configuration threshold condition from the progress configuration data set; And acquiring a state configuration data set corresponding to the game state data from the target role configuration data set and the target progress configuration data set.
5. The method of claim 2, wherein generating prop intensity constraint information from the game state data comprises: Determining prop price constraint information and prop frequency constraint information configured by a game account corresponding to the game virtual character according to the character payment frequency corresponding to the game virtual character in the game state data; Determining additional step number constraint information configured by the game account according to the remaining clearance step number corresponding to the game virtual character; generating physical superposition constraint rules according to physical states included in the character information in the game state data, and generating prop issuing constraint rules according to the game progress information; and combining the prop price constraint information, the prop frequency constraint information, the additional step number constraint information, the physical superposition constraint rule and the prop issuing constraint rule into prop intensity constraint information.
6. The method of claim 1, wherein the prop intensity constraint information comprises prop price constraint information, prop number constraint information, extra step number constraint information, physical superposition constraint rules and prop issuing constraint rules, the state configuration data set comprises S state configuration data, the S state configuration data comprises state configuration data S i , S is a positive integer, i is a positive integer less than or equal to S, and the constraint processing is performed on the state configuration data set based on the prop intensity constraint information to obtain target state configuration data, wherein the method comprises the following steps: Obtaining prop price configuration data, prop use times, extra steps, physical superposition conditions and prop issuing conditions in the state configuration data S i ; If the prop price configuration data is lower than the lowest price indicated by the prop price constraint information, or the prop use frequency is greater than the maximum frequency indicated by the prop frequency constraint information, or the extra step number is greater than the maximum step number indicated by the extra step number constraint information, or the physical superposition condition does not meet the physical superposition constraint rule, or the prop issuing condition does not meet the prop issuing constraint rule, deleting the state configuration data S i ; and if the prop price configuration data, the prop use times, the extra steps, the physical superposition conditions and the prop issuing conditions all meet the prop intensity constraint information, carrying out parameter calibration on the configuration parameters of the state configuration data S i to obtain target state configuration data matched with the prop intensity constraint information.
7. The method of claim 6, wherein performing parameter calibration on the configuration parameters of the state configuration data S i to obtain target state configuration data matched with the prop intensity constraint information includes: Performing data evaluation on the state configuration data S i to obtain configuration prediction data; If the configuration prediction data is out of the standard numerical range, inputting the game state data and the state configuration data S i into a simulation model, and outputting simulation sample data corresponding to the game state data through the simulation model, wherein the simulation sample data comprises a clearance rate used for representing a stay gate indicated by the game progress information and cost data required for clearance of the stay gate; if the clearance or the cost data is outside the standard numerical range, adjusting the configuration parameters of the state configuration data S i to obtain candidate state configuration data corresponding to the state configuration data S i ; And continuously inputting the candidate state configuration data and the game state data into the simulation model until simulation sample data output by the simulation model is in the standard numerical range, and determining the latest candidate state configuration data as target state configuration data matched with the prop intensity constraint information.
8. The method of claim 7, wherein the standard numerical range includes a clearance numerical range and a cost numerical range, and wherein adjusting the configuration parameters of the state configuration data S i to obtain candidate state configuration data corresponding to the state configuration data S i if the clearance or the cost data is outside the standard numerical range includes: If the clearance is greater than the maximum clearance value indicated by the clearance value range or the cost data is greater than the maximum cost value indicated by the cost value range, reducing the configuration parameters of the state configuration data S i according to a fixed parameter step length to obtain candidate state configuration data corresponding to the state configuration data S i ; And if the clearance is smaller than the minimum clearance value indicated by the clearance value range, increasing the configuration parameters of the state configuration data S i according to the fixed parameter step length to obtain candidate state configuration data corresponding to the state configuration data S i .
9. The method of claim 1, wherein the target state configuration data includes attribute enhancement data, a prop name identifier, and a prop icon identifier, wherein the generating a game virtual prop for enhancing an attribute strength corresponding to the game virtual character based on the target state configuration data comprises: Obtaining a prop name from a file knowledge base according to the prop name identification, and obtaining a prop icon from an image knowledge base according to the prop icon identification; Combining the prop name, the prop icon and the attribute enhancement data into a game virtual prop corresponding to the target state configuration data, wherein the attribute enhancement data is used for enhancing the attribute strength corresponding to the game virtual role.
10. The method of claim 1, wherein the target state configuration data further comprises a prop text identifier, the method further comprising: acquiring a text putting rule, and acquiring prop text data for describing the game virtual prop in a file knowledge base according to the prop text identifier; Extracting features of the text input rule to obtain input rule features, and extracting features of the prop text data to obtain prop text features; Performing feature matching on the throwing rule features and the prop text features to obtain text similarity between the text throwing rule and the prop text data, inputting the game virtual prop into a simulation model, and outputting defect detection data for reflecting design defects of the game world through the simulation model; and if the text similarity is smaller than a text similarity threshold and the defect detection data meets the defect detection condition, adding the game virtual prop into a release queue, wherein the release queue is used for releasing the game virtual prop to a game account corresponding to the game virtual character.
11. The method of claim 1, wherein the number of characters of the game avatar is at least two, the method further comprising: dividing the at least two game virtual roles into comparison virtual roles and test virtual roles according to the role identifications respectively corresponding to the at least two game virtual roles, wherein the number of the roles of the comparison virtual roles is larger than that of the test virtual roles; Acquiring fixed virtual props from a prop configuration library according to game state data of the comparison virtual roles, putting the fixed virtual props into game accounts corresponding to the comparison virtual roles, and synchronously acquiring the comparison attribute strength of the comparison virtual roles after the fixed virtual props are used; acquiring the test attribute strength of the test virtual character after the game virtual prop is used; If the difference between the test attribute intensity and the comparison attribute intensity is out of the attribute intensity range, stopping the throwing of the game virtual prop, and throwing the fixed virtual prop to a game account corresponding to the test virtual character; And if the difference between the test attribute intensity and the comparison attribute intensity is within the attribute intensity range, stopping the throwing of the fixed virtual prop, and throwing the game virtual prop to a game account corresponding to the comparison virtual character.
12. A data processing apparatus, comprising: The receiving and transmitting module is used for acquiring role information, game progress information and game service constraint information corresponding to a game virtual role in a game world, and determining the role information and the game progress information as game state data corresponding to the game virtual role; The system comprises a data set generation module, a large language model, a state configuration data set and a game state data set, wherein the data set generation module is used for inputting the game state data into a retrieval model, generating historical game configuration data associated with the game state data through the retrieval model, inputting the game state data, the game service constraint information and the historical game configuration data into the large language model, and outputting the state configuration data set corresponding to the game state data through the large language model; The set constraint processing module is used for generating prop intensity constraint information according to the game state data, and carrying out constraint processing on the state configuration data set based on the prop intensity constraint information to obtain target state configuration data; And the virtual prop throwing module is used for generating game virtual props for enhancing the attribute intensity corresponding to the game virtual roles based on the target state configuration data and throwing the game virtual props into game accounts corresponding to the game virtual roles.
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 When the game virtual character controlled by the user fails for a plurality of times at a certain level, an operator can update and develop the game prop based on the condition of the user, and an update package obtained by development is issued to the user, so that the user can update the game according to the update package to obtain a new game prop. However, since the whole process from developing an update package to updating a game by a user is relatively long, the update efficiency of game props is low, and the problem of user blocking is difficult to solve in time, and operators usually analyze common blocking problems, namely, the content of the update package issued to all users is uniform, so that props acquired by different users are completely consistent, but the blocking problems of part of users do not belong to the common blocking problems, and the problem of blocking is still difficult to solve by means of new game props, so that the update of the game props is not accurate enough. Disclosure of Invention The embodiment of the application provides a data processing method, a device, equipment and a readable storage medium, which can improve the updating efficiency and updating accuracy for game props. In one aspect, an embodiment of the present application provides a data processing method, including: Acquiring character information, game progress information and game service constraint information corresponding to a game virtual character in a game world, and determining the character information and the game progress information as game state data corresponding to the game virtual character; The game state data is input into a search model, historical game configuration data related to the game state data is generated through the search model, the game state data, game service constraint information and the historical game configuration data are input into a large language model, and a state configuration data set corresponding to the game state data is output through the large language model; The method comprises the steps of generating prop intensity constraint information according to game state data, and carrying out constraint processing on a state configuration data set based on the prop intensity constraint information to obtain target state configuration data; and generating game virtual props for enhancing the attribute intensity corresponding to the game virtual roles based on the target state configuration data, and throwing the game virtual props into game accounts corresponding to the game virtual roles. Wherein, still include: acquiring the stay time of the game virtual character at the stay level, the number of remaining clearance steps and the key blocking number, wherein the key blocking number refers to the number of unblocked blocking of the game virtual character at the stay level; If the stay time length is greater than or equal to a time length threshold, or the number of remaining clearance steps is less than a step number threshold, or the number of key barriers is greater than or equal to a number threshold, acquiring a checkpoint identifier, checkpoint type information and scenario node information of a stay checkpoint; and determining the gate identification, the gate type information and the scenario node information as game progress information corresponding to the game virtual character. Wherein inputting the game state data to the search model, generating historical game configuration data associated with the game state data by the search model, comprises: Inputting the game state data into a retrieval model, and extracting features of the game state data through the retrieval model to obtain game state features, wherein the game state features comprise game progress features corresponding to game progress information; Acquiring one or more game level information associated with a game world in a knowledge base through a retrieval model, and extracting features of the one or more game level information to obtain game level features corresponding to the one or more game level information respectively; Performing similar matching on the game progress features and one or more game level features respectively to obtain level similarity corresponding to the one or more game level features respectively, and determining game level information with the level similarity being greater than or equal to a level similarity threshold value in the one or more game level information as target game level information; extracting features of one or more initial state configuration data associated with the target game level information to obtain initial configuration features corresponding to the one or more initial st