Game value simulation method and terminal

Abstract

The application provides a game value simulation method and a terminal, which comprises the following steps: obtaining historical value change data of each attribute in a game, constructing and training a value simulation model based on the historical value change data; verifying the value simulation model, inputting current values of each attribute into the value simulation model for multiple iterations in real time, and outputting value change results; and adjusting the current values of each attribute in real time according to the value change results. The application constructs and trains a value simulation model through a machine learning algorithm, analyzes a large amount of player data, iteratively predicts the influence of different value adjustments on game experience, and thus realizes personalized and real-time game balance optimization, improves the scientificity of game value design and the game experience of players, provides more abundant and diverse game content, and enhances the immersion and satisfaction of players.

Description

Technical Field

[0001] This invention relates to the field of AI technology, and in particular to a game numerical simulation method and terminal. Background Technology

[0002] AI-based game simulation technology typically involves using artificial intelligence algorithms, such as machine learning, deep learning, and reinforcement learning, to simulate and optimize various dynamic processes within a game. These technologies can be used to train non-player characters (NPCs), generate game content, optimize game performance, and enhance the player experience.

[0003] While AI-based game numerical simulation technology has significant advantages in providing simulations of highly complex and dynamic game environments, it still has some drawbacks: First, these systems typically require significant computing resources to train and run complex AI models, which can lead to high costs and computational bottlenecks.

[0004] Secondly, existing technologies may face challenges in handling large-scale multi-agent interactions, especially in game scenarios that require coordinating complex strategies and behaviors.

[0005] Third, the generalization and robustness of AI models are also current research challenges. Models may perform well when trained in specific game environments, but may not perform well when faced with changes in the environment or new games.

[0006] Furthermore, in terms of real-time performance, deep learning models may not meet the demands of fast response and low latency in games, limiting the application of AI in real-time gaming. Security and privacy protection are critical factors that must be considered when designing these systems, as AI models may process sensitive user data. Therefore, appropriate safeguards must be implemented to ensure data security and user privacy. Summary of the Invention

[0007] The technical problem to be solved by this invention is to provide a game numerical simulation method and terminal to realize the automatic adjustment of game numerical values, so as to improve the scientific nature of game numerical design and the player's gaming experience.

[0008] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: A game numerical simulation method includes the following steps: S1. Obtain historical numerical change data of each attribute in the game, and construct and train a numerical simulation model based on the historical numerical change data. S2. Verify the numerical simulation model, and input the current values ​​of each attribute into the numerical simulation model in real time for multiple iterations, and output the numerical change results. S3. Based on the numerical change results, adjust the current values ​​of each attribute in real time.

[0009] To solve the above-mentioned technical problems, another technical solution adopted by the present invention is as follows: A game simulation terminal includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it performs the following steps: S1. Obtain historical numerical change data of each attribute in the game, and construct and train a numerical simulation model based on the historical numerical change data. S2. Verify the numerical simulation model, and input the current values ​​of each attribute into the numerical simulation model in real time for multiple iterations, and output the numerical change results. S3. Based on the numerical change results, adjust the current values ​​of each attribute in real time.

[0010] The beneficial effects of this invention are as follows: It provides a game numerical simulation method and terminal, which uses machine learning algorithms to construct and train numerical simulation models, analyzes a large amount of player data, and iteratively predicts the impact of different numerical adjustments on the game experience, thereby achieving personalized and real-time game balance optimization, improving the scientific nature of game numerical design and the player's game experience, providing richer and more diverse game content, and enhancing the player's immersion and satisfaction. Attached Figure Description

[0011] Figure 1 This is an overall flowchart of a game numerical simulation method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of a game numerical simulation terminal according to an embodiment of the present invention.

[0012] Label Explanation: 1. A game numerical simulation terminal; 2. Memory; 3. Processor. Detailed Implementation

[0013] To explain in detail the technical content, objectives, and effects of the present invention, the following description is provided in conjunction with the embodiments and accompanying drawings.

[0014] Please refer to Figure 1 A method for game numerical simulation, comprising the following steps: S1. Obtain historical numerical change data of each attribute in the game, and construct and train a numerical simulation model based on the historical numerical change data. S2. Verify the numerical simulation model, and input the current values ​​of each attribute into the numerical simulation model in real time for multiple iterations, and output the numerical change results. S3. Based on the numerical change results, adjust the current values ​​of each attribute in real time.

[0015] As can be seen from the above description, the beneficial effects of the present invention are as follows: It provides a game numerical simulation method, which uses machine learning algorithms to construct and train a numerical simulation model, analyzes a large amount of player data, and iteratively predicts the impact of different numerical adjustments on the game experience, thereby achieving personalized and real-time game balance optimization, so as to improve the scientific nature of game numerical design and the player's game experience, provide richer and more diverse game content, and enhance the player's immersion and satisfaction.

[0016] Furthermore, before constructing and training the numerical simulation model, the following steps are also included: The historical numerical change data is cleaned and formatted, and key features are extracted.

[0017] As described above, the acquired data is cleaned and formatted to ensure data quality and facilitate subsequent processing.

[0018] Furthermore, the construction and training of the numerical simulation model based on the historical numerical change data specifically includes: A numerical simulation model is constructed using machine learning algorithms, and the numerical simulation model is trained based on the historical numerical change data. The machine learning algorithms include linear regression, decision trees, and neural networks.

[0019] As described above, the acquired data is used to train the model using machine learning algorithms, and the model parameters are adjusted to improve simulation accuracy.

[0020] Furthermore, the verification of the numerical simulation model in step S2 specifically involves: The generalization ability and accuracy of the numerical simulation model are verified through cross-validation or test set validation.

[0021] As described above, the generalization ability and accuracy of the model are verified to ensure that the game values ​​obtained from subsequent simulations are scientific and reasonable.

[0022] Furthermore, after step S3, the method further includes: S4. Based on the numerical change results and player feedback, adjust the model parameters of the numerical simulation model in reverse.

[0023] As described above, iterative optimization of the model is performed to ensure that the game values ​​obtained from the simulation remain in dynamic equilibrium, thereby optimizing the game experience.

[0024] Please refer to Figure 2A game numerical simulation terminal includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it performs the following steps: S1. Obtain historical numerical change data of each attribute in the game, and construct and train a numerical simulation model based on the historical numerical change data. S2. Verify the numerical simulation model, and input the current values ​​of each attribute into the numerical simulation model in real time for multiple iterations, and output the numerical change results. S3. Based on the numerical change results, adjust the current values ​​of each attribute in real time.

[0025] As can be seen from the above description, the beneficial effects of the present invention are as follows: Based on the same technical concept, and in conjunction with the above-mentioned game numerical simulation method, a game numerical simulation terminal is provided. The numerical simulation model is constructed and trained through machine learning algorithms, a large amount of player data is analyzed, and the impact of different numerical adjustments on the game experience is iteratively predicted. This achieves personalized and real-time game balance optimization, thereby improving the scientific nature of game numerical design and the player's game experience, providing richer and more diverse game content, and enhancing the player's immersion and satisfaction.

[0026] Furthermore, before constructing and training the numerical simulation model, the following steps are also included: The historical numerical change data is cleaned and formatted, and key features are extracted.

[0027] As described above, the acquired data is cleaned and formatted to ensure data quality and facilitate subsequent processing.

[0028] Furthermore, the construction and training of the numerical simulation model based on the historical numerical change data specifically includes: A numerical simulation model is constructed using machine learning algorithms, and the numerical simulation model is trained based on the historical numerical change data. The machine learning algorithms include linear regression, decision trees, and neural networks.

[0029] As described above, the acquired data is used to train the model using machine learning algorithms, and the model parameters are adjusted to improve simulation accuracy.

[0030] Furthermore, the verification of the numerical simulation model in step S2 specifically involves: The generalization ability and accuracy of the numerical simulation model are verified through cross-validation or test set validation.

[0031] As described above, the generalization ability and accuracy of the model are verified to ensure that the game values ​​obtained from subsequent simulations are scientific and reasonable.

[0032] Furthermore, after step S3, the method further includes: S4. Based on the numerical change results and player feedback, adjust the model parameters of the numerical simulation model in reverse.

[0033] As described above, iterative optimization of the model is performed to ensure that the game values ​​obtained from the simulation remain in dynamic equilibrium, thereby optimizing the game experience.

[0034] This invention provides a game numerical simulation method and terminal, mainly applied to scenarios involving dynamic adjustment of various attribute values ​​in games. The following is a detailed description with reference to specific embodiments: Please refer to Figure 1 Embodiment 1 of the present invention is as follows: A game numerical simulation method, such as Figure 1 As shown, the steps include: S1. Obtain historical numerical change data of various attributes in the game, including but not limited to historical numerical change data of character attributes, equipment attributes, and skill attributes, and then build and train a numerical simulation model based on the historical numerical change data.

[0035] S2. Verify the numerical simulation model and input the current values ​​of each attribute into the numerical simulation model in real time for multiple iterations, and output the numerical change results.

[0036] S3. Adjust the current values ​​of each attribute in real time based on the numerical change results.

[0037] In this embodiment, a numerical simulation model is constructed and trained using machine learning algorithms. This model analyzes a large amount of player data and iteratively predicts the impact of different numerical adjustments on the game experience. This enables personalized and real-time game balance optimization, improving the scientific nature of game numerical design and the player's gaming experience. It also provides richer and more diverse game content, enhancing player immersion and satisfaction.

[0038] Embodiment 2 of the present invention is as follows: A game numerical simulation method, based on the above embodiment one, further includes the following step before constructing and training the numerical simulation model: Historical numerical change data is cleaned and formatted to ensure data quality and facilitate subsequent processing.

[0039] Then, based on the needs of game design and numerical balance, key features are extracted from the cleaned and formatted data.

[0040] In this embodiment, a numerical simulation model is constructed and trained based on historical numerical change data, specifically as follows: If we use machine learning algorithms to build numerical simulation models, such as linear regression, decision trees and neural networks, and train the numerical simulation models based on historical numerical change data; The acquired data is used to train the model using machine learning algorithms, and the model parameters are adjusted to improve simulation accuracy.

[0041] Meanwhile, in this embodiment, the verification of the numerical simulation model in step S2 specifically involves: The generalization ability and accuracy of the numerical simulation model are verified by methods such as cross-validation or test sets.

[0042] This involves verifying the model's generalization ability and accuracy to ensure that the game values ​​obtained from subsequent simulations are scientifically sound and reasonable.

[0043] Finally, the validated model is applied to game numerical simulation to simulate numerical changes and outcomes in the game. The simulation results are analyzed to evaluate the balance and rationality of the game's numerical values. Based on the analysis results, game values ​​or model parameters can be adjusted to optimize the game experience. Adjusting model parameters can be achieved by iteratively updating the numerical simulation model through player feedback or game updates. In this embodiment, step S3 is followed by: S4. Adjust the model parameters of the numerical simulation model in reverse based on the numerical change results and player feedback.

[0044] This involves iteratively optimizing the model to maintain a dynamic balance in the game values ​​obtained from the simulation, thereby optimizing the game experience.

[0045] In this embodiment, the essence of the entire solution is to simulate a game character DEMO platform, then put the game rules and game values ​​into the game character DEMO platform, and then give a rule so that the platform can perform multiple iterations and batch runs based on the initial data and the rule, and finally obtain the numerical changes after N iterations. This is used to simulate the changes in game values ​​obtained by the player after performing the game multiple times according to the game rules, and to verify the game numerical balance issue.

[0046] The following are specific numerical examples for illustration: For example, a game has a rule for normal enhancement, and each enhancement has the following possible outcomes: 1. Success, enhancement level +1; 2. If it fails, the enhancement level remains unchanged; 3. Failure, enhancement level -1; 4. Failure results in the equipment disappearing.

[0047] We aim to see both the big picture and the boundaries to grasp the player experience, which can be illustrated using Table 1 below: Table 1:

[0048] The existing numerical modeling rules may be as follows: Level 1.0 Enhancement: 50% chance to upgrade to level 11, 50% chance to remain at level 10; Level 2.11 Enhancement: 30% chance to upgrade to level 12, 50% chance to remain at level 11, and 20% chance to revert to level 10; Level 3.12 Enhancement: 20% chance to increase to level 13, 30% chance to remain at level 12, and 50% chance to return to level 11; Level 4.13: Enhancement complete.

[0049] Based on the solution in this embodiment, by inputting this rule and the initial values ​​of the game, and setting 10 iterations, it is found that the number of iterations is too dispersed; some players can complete it in 5 steps, while others need 95 steps. Therefore, it is possible to intuitively simulate the subsequent gameplay experience of players based on this numerical system, which means optimizing this numerical system and changing it to: 1. After each level 10 enhancement, the probability of enhancing to level 11 on the next attempt increases by 5%, while the probability of maintaining level 10 decreases by 5%. 2. After each level 11 enhancement, the probability of enhancing to level 12 on the next attempt increases by 2%, while the probability of maintaining level 11 decreases by 2%. 3. After each level 12 enhancement, the probability of enhancing to level 13 on the next attempt increases by 1%, while the probability of maintaining level 12 decreases by 1%. Then, prediction simulation is performed again using this embodiment to ultimately achieve optimization.

[0050] The specific values ​​are shown in Table 2 below: Table 2:

[0051] As can be seen from Table 2: 1. The optimized version has fewer average steps and fewer enhancement stones, indicating that the overall process has been streamlined. 2. The optimized maximum number of steps and the number of enhancement stones are smaller, indicating that the values ​​are more convergent.

[0052] Please refer to Figure 2 Embodiment 3 of the present invention is as follows: A game numerical simulation terminal 1 includes a memory 2, a processor 3, and a computer program stored on the memory 2 and executable on the processor 3. When the processor 3 executes the computer program, it completes the steps in the game numerical simulation method described in Embodiment 1 or Embodiment 2.

[0053] In summary, the game numerical simulation method and terminal provided by this invention constructs and trains a numerical simulation model through machine learning algorithms, analyzes a large amount of player data, and iteratively predicts the impact of different numerical adjustments on the game experience, thereby achieving personalized and real-time game balance optimization. This improves the scientific nature of game numerical design and the player's gaming experience, provides richer and more diverse game content, and enhances player immersion and satisfaction.

[0054] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent modifications made based on the content of the present invention's specification and drawings, or direct or indirect applications in related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A game numerical simulation method, characterized in that, Including the following steps: S1. Obtain historical numerical change data of each attribute in the game, and construct and train a numerical simulation model based on the historical numerical change data. S2. Verify the numerical simulation model, and input the current values ​​of each attribute into the numerical simulation model in real time for multiple iterations, and output the numerical change results. S3. Based on the numerical change results, adjust the current values ​​of each attribute in real time.

2. The game numerical simulation method according to claim 1, characterized in that, Before constructing and training the numerical simulation model, the following steps are also included: The historical numerical change data is cleaned and formatted, and key features are extracted.

3. The game numerical simulation method according to claim 2, characterized in that, The numerical simulation model constructed and trained based on the historical numerical change data is specifically as follows: A numerical simulation model is constructed using machine learning algorithms, and the numerical simulation model is trained based on the historical numerical change data. The machine learning algorithms include linear regression, decision trees, and neural networks.

4. The game numerical simulation method according to claim 1, characterized in that, The verification of the numerical simulation model in step S2 specifically involves: The generalization ability and accuracy of the numerical simulation model are verified through cross-validation or test set validation.

5. The game numerical simulation method according to claim 1, characterized in that, Following step S3, the following is also included: S4. Based on the numerical change results and player feedback, adjust the model parameters of the numerical simulation model in reverse.

6. A game numerical simulation terminal, characterized in that, Includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, performs the following steps: S1. Obtain historical numerical change data of each attribute in the game, and construct and train a numerical simulation model based on the historical numerical change data. S2. Verify the numerical simulation model, and input the current values ​​of each attribute into the numerical simulation model in real time for multiple iterations, and output the numerical change results. S3. Based on the numerical change results, adjust the current values ​​of each attribute in real time.

7. A game numerical simulation terminal according to claim 6, characterized in that, Before constructing and training the numerical simulation model, the following steps are also included: The historical numerical change data is cleaned and formatted, and key features are extracted.

8. A game numerical simulation terminal according to claim 7, characterized in that, The numerical simulation model constructed and trained based on the historical numerical change data is specifically as follows: A numerical simulation model is constructed using machine learning algorithms, and the numerical simulation model is trained based on the historical numerical change data. The machine learning algorithms include linear regression, decision trees, and neural networks.

9. A game numerical simulation terminal according to claim 6, characterized in that, The verification of the numerical simulation model in step S2 specifically involves: The generalization ability and accuracy of the numerical simulation model are verified through cross-validation or test set validation.

10. A game numerical simulation terminal according to claim 6, characterized in that, Following step S3, the following is also included: S4. Based on the numerical change results and player feedback, adjust the model parameters of the numerical simulation model in reverse.

Images