Real-time dynamic game behavior decision method and apparatus thereof

Abstract

The application provides a real-time dynamic game behavior decision method and device, and relates to the fields of artificial intelligence and data processing technology.The method comprises the following steps: in response to receiving a surrender instruction, collecting multi-dimensional game behavior characteristics; for each target object in the team, constructing an index evaluation value of multiple indexes according to the multi-dimensional game behavior characteristics, wherein the multiple indexes at least include operation density, operation accuracy, survival behavior and historical resilience; dynamically weighting the index evaluation value of each index to obtain a game behavior evaluation value of the target object; obtaining a relative performance index and a dynamic decision weight of each target object, obtaining a decision evaluation value corresponding to the surrender instruction according to the game behavior evaluation value, the relative performance index and the dynamic decision weight, and processing the surrender instruction according to the decision evaluation value.The application realizes the upgrade of game behavior decision from traditional win rate prediction to behavior-driven intelligent arbitration, and significantly improves the decision accuracy and player experience.

Description

Technical Field

[0001] This application relates to the fields of artificial intelligence and data processing technology, and in particular to a real-time dynamic game behavior decision-making method and apparatus. Background Technology

[0002] In existing technologies, real-time predictions are made using deep learning models (such as spatiotemporal graph convolutional networks and Transformers) based on current match progress data and player historical data. Reinforcement learning is then used to dynamically adjust feature weights and calculate win rates in real time, enabling game behavior decisions based on surrender commands. However, existing technologies only process objective game data, neglecting the impact of varying player psychological resilience on game win rates. Furthermore, surrender decision-making mechanisms suffer from problems such as a single-dimensional approach, rigid static thresholds, and a lack of team coordination. Summary of the Invention

[0003] This application aims to at least partially address one of the technical problems in the related art. The technical solution of this application is as follows:

[0004] The first aspect of this application proposes a real-time dynamic game behavior decision-making method, including: In response to receiving a surrender command, multi-dimensional game behavior characteristics are collected. These multi-dimensional game behavior characteristics include at least historical behavior data from past game matches, real-time behavior data from the current game match, and team coordination behavior data. For each target in the team, based on multidimensional game behavior characteristics, construct evaluation values ​​for multiple indicators, including at least operation density, operation accuracy, survival behavior, and historical resilience. The game behavior evaluation value of the target object is obtained by dynamically weighting the evaluation value of each indicator. Obtain the relative effectiveness index and dynamic decision weight for each target object. Based on the game behavior evaluation value, relative effectiveness index, and dynamic decision weight, obtain the decision evaluation value corresponding to the surrender command, and process the surrender command based on the decision evaluation value.

[0005] This application constructs a four-factor evaluation value based on multi-dimensional data collection, including real-time operations, historical behavior, and team collaboration. This value assesses operational density, operational accuracy, survival behavior, and historical resilience, capturing real-time fluctuations in player status and reducing misjudgment rates. By combining relative efficiency indices and dynamic weights at the team decision-making level, it upgrades game behavior decision-making from traditional win rate prediction to behavior-driven intelligent adjudication, significantly improving decision-making accuracy and player experience.

[0006] A second aspect of this application provides a real-time dynamic game behavior decision-making device, comprising: The data acquisition module is used to collect multi-dimensional game behavior features in response to receiving a surrender command. These multi-dimensional game behavior features include at least historical behavior data from past game matches, real-time behavior data from the current game match, and team collaboration behavior data. The building module is used to construct evaluation values ​​for multiple indicators for each target object in the team based on multidimensional game behavior characteristics. These indicators include at least operation density, operation accuracy, survival behavior, and historical resilience. The acquisition module is used to dynamically weight the game behavior evaluation value of the target object based on the evaluation value of each indicator; The decision-making module is used to obtain the relative effectiveness index and dynamic decision weight of each target object. Based on the game behavior evaluation value, relative effectiveness index and dynamic decision weight, it obtains the decision evaluation value corresponding to the surrender command and processes the surrender command based on the decision evaluation value.

[0007] A third aspect of this application provides an electronic device, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute the instructions to implement a real-time dynamic game behavior decision-making method as provided in the first aspect of this application.

[0008] A fourth aspect of this application provides a computer-readable storage medium that, when the instructions in the computer-readable storage medium are executed by a processor of an electronic device, enables the electronic device to perform the real-time dynamic game behavior decision-making method provided in the first aspect of this application.

[0009] To achieve the above objectives, a fifth aspect of this application provides a computer program product, including a computer program that, when executed by a processor, implements the real-time dynamic game behavior decision-making method provided in the first aspect of this application.

[0010] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0011] Figure 1 This is a flowchart of a real-time dynamic game behavior decision-making method according to an embodiment of this application; Figure 2 This is a flowchart of an embodiment of the construction operation density index evaluation value of this application; Figure 3 This is a flowchart of an embodiment of the construction operation accuracy evaluation value of this application; Figure 4 This is a flowchart illustrating the evaluation of survival behavior metrics according to one embodiment of this application; Figure 5This is a flowchart illustrating the evaluation value of historical resilience metrics in one embodiment of this application; Figure 6 This is a flowchart of a real-time dynamic game behavior decision-making method according to an embodiment of this application; Figure 7 This is a structural block diagram of a real-time dynamic game behavior decision-making device according to an embodiment of this application; Figure 8 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0012] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0013] The acquisition, storage, and application of information and data involved in the technical solution of this application all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0014] The examples in this application may involve user data, data acquisition, and / or use. All of these aspects comply with relevant laws, regulations, and provisions. In the examples of this application, all data collection, acquisition, processing, manipulation, forwarding, and use are conducted with the user's knowledge and confirmation. Accordingly, in implementing each example, the type, scope of use, and usage scenarios of any data or information that may be involved should be communicated to the user and their authorization obtained through appropriate means in accordance with relevant laws and regulations. The specific methods of notification and / or authorization can vary depending on the actual situation and application scenario, and the scope of the solution is not limited in this regard.

[0015] In this manual and the sample solutions, any processing of personal information will be conducted only under legal grounds (such as obtaining the consent of the data subject or being necessary for the performance of a contract) and will only be carried out within the scope stipulated or agreed upon. A user's refusal to process personal information beyond what is necessary for basic functions will not affect the user's use of basic functions.

[0016] The following description, in conjunction with the accompanying drawings, outlines a real-time dynamic game behavior decision-making method and apparatus according to embodiments of this application.

[0017] Figure 1 This is a flowchart of a real-time dynamic game behavior decision-making method according to an embodiment of this application, such as... Figure 1 As shown, the method includes the following steps: S101, in response to receiving a surrender command, collects multi-dimensional game behavior characteristics, which include at least historical behavior data of past game matches, real-time behavior data of the current game match, and team collaborative behavior data.

[0018] In response to surrender commands received from the system or players, in order to accurately determine the rationality and authenticity of surrender requests, it is necessary to collect multi-dimensional game behavior feature data to build a comprehensive and dynamic evaluation system.

[0019] Multidimensional game behavior characteristics include at least the following key data categories: for example, acquiring historical game behavior data, covering the player's and / or the entire server's past win rate, economy, and operation records, to analyze their habitual style and stability level; for another example, collecting real-time behavior data of the current game, such as the total number and hit count of non-targeted operations, economy, and operation records, to reflect the immediate state and abnormal fluctuations; and for yet another example, integrating team coordination behavior data, including the number of team fight coordinations, signal communication density, position complementarity, and support efficiency, to assess the overall level of cooperation.

[0020] By integrating these three types of data, the system can effectively distinguish between rational surrender and emotional or malicious surrender behavior, thereby improving the accuracy and fairness of surrender judgments.

[0021] S102, for each target object in the team, construct indicator evaluation values ​​for multiple indicators based on multi-dimensional game behavior characteristics. These multiple indicators include at least operation density, operation accuracy, survival behavior, and historical resilience.

[0022] The target audience can be players in a game. For each target audience member in the team (e.g., each player character), in order to accurately characterize their behavioral tendencies and credibility in surrender decision-related scenarios, it is necessary to systematically construct indicator evaluation values ​​corresponding to multiple indicators based on the collected multi-dimensional game behavior characteristics.

[0023] These metrics should cover at least the following four core dimensions: First, operational density, which measures the number and frequency of effective operations performed by players per unit of time, such as skill release, positioning adjustment, perspective switching, and command issuance, reflecting their participation and activity level; Second, operational precision, which assesses the hit rate and timing accuracy of key operations (such as aiming, hitting, interrupting, and dodging), characterizing their control and accuracy under high-pressure conditions; Third, survival behavior, including positioning to avoid danger, reasonable use of healing or shield skills, timing of retreat, and life resource management in team battles, reflecting their self-preservation and risk assessment capabilities; Fourth, historical resilience, which refers to a player's continued fighting willingness when facing unfavorable situations in past matches, the frequency of refusing to surrender, and historical performance in successfully turning the tide, used to measure their psychological resilience and perseverance in disadvantageous situations.

[0024] By quantifying and scoring the above indicators, a fine-grained and comparable behavioral profile can be established for each individual in the team, thereby providing a data foundation for subsequent surrender judgments and decision-making interventions.

[0025] S103: Dynamically weight the evaluation values ​​of each indicator to obtain the game behavior evaluation value of the target object.

[0026] For each target object, the multiple indicator evaluation values ​​that have been constructed, such as operation density, operation accuracy, survival behavior and historical resilience, need to be further integrated through dynamic weighting methods to obtain a game behavior evaluation value that can comprehensively reflect the rationality of its current game behavior.

[0027] The core of dynamic weighting lies in the fact that the weights of each indicator are not fixed, but rather adaptively adjusted based on the real-time game situation and the stage of the match. For example, in the early game, operation density and operation accuracy have a greater impact on the situation and can be given higher weights; as the mid-to-late game team fights progress, the importance of survival behavior and historical resilience increases significantly, and their weights are increased accordingly. In addition, when a certain indicator value shows extreme anomalies, such as a sudden drop in operation density or serious errors in survival behavior, the system can dynamically reduce its weight to avoid outliers interfering with the overall judgment.

[0028] The game behavior evaluation value calculated by this dynamic weighting method can more sensitively reflect the player's true behavior in the current game environment, providing more accurate and individualized data support for subsequent judgment on whether to respond to the surrender command.

[0029] S104: Obtain the relative effectiveness index and dynamic decision weight of each target object; based on the game behavior evaluation value, relative effectiveness index and dynamic decision weight, obtain the decision evaluation value corresponding to the surrender command; and process the surrender command based on the decision evaluation value.

[0030] For each target within the team, the system further obtains their relative effectiveness index and dynamic decision weight. Optionally, the relative effectiveness index measures the ratio of the member's overall contribution in the current game to the average level of other members in the team; the dynamic decision weight represents the individual influence coefficient adjusted in real time. Subsequently, the system integrates each target's game behavior evaluation value, relative effectiveness index, and dynamic decision weight, and calculates the decision evaluation value corresponding to the surrender command through a weighted aggregation method. Finally, based on the comparison result of this decision evaluation value with a preset threshold, the system automatically approves, rejects, or delays the processing of the surrender command.

[0031] In some implementations, if the decision evaluation value is greater than or equal to a preset decision evaluation value threshold, a vote is allowed, and the surrender command is processed based on the voting results. If the decision evaluation value is less than the decision evaluation value threshold, the surrender command is rejected, and a cooldown period is set.

[0032] This application constructs a four-factor evaluation value based on multi-dimensional data collection, including real-time operations, historical behavior, and team collaboration. This value assesses operational density, operational accuracy, survival behavior, and historical resilience, capturing real-time fluctuations in player status and reducing misjudgment rates. By combining relative efficiency indices and dynamic weights at the team decision-making level, it upgrades game behavior decision-making from traditional win rate prediction to behavior-driven intelligent adjudication, significantly improving decision-making accuracy and player experience.

[0033] Figure 2 This is a flowchart of a real-time dynamic game behavior decision-making method according to an embodiment of this application, such as... Figure 2 As shown, the process of constructing an evaluation value for operation density based on multidimensional game behavior characteristics includes the following steps: S201. Based on multidimensional game behavior characteristics, obtain the effective operation score of the target object within a unit of time, and the mean and standard deviation of the effective operation score of the entire server's group of objects of the same skill level within a unit of time, and calculate the operation intensity index.

[0034] S202. Based on the multidimensional game behavior characteristics, obtain the mean and standard deviation of the effective operation scores of the target object in its historical wins and losses, and calculate the historical operation stability index.

[0035] S203. Based on the multidimensional game behavior characteristics, obtain the mean and standard deviation of the effective operation scores of the target object in the advantageous and disadvantageous periods, and calculate the operation resilience index.

[0036] S204. Based on the operational intensity index, historical operational stability index, and operational resilience index, calculate the immediate operational intensity score, operational mindset stability score, and adversity response effectiveness score, respectively.

[0037] S205 dynamically adjusts the weights of the immediate operational intensity score, operational mindset stability score, and adversity response effectiveness score based on the positive or negative historical operational stability index, generating an evaluation value for operational density.

[0038] In some implementations, a dynamic comparison benchmark can be established by constructing a database of player actions of similar skill levels across the entire server. In the game, certain effective actions correspond to specific scores, and players receive scores for each effective action. First, the total score of effective actions earned by a player within the most recent minute of the current match is collected and denoted as... Optionally, the unit of time can be 1 minute or other time; this application does not impose any restrictions on this.

[0039] Furthermore, valid operation scores are collected from the entire server. The collection scope can be the complete match data of the past 30 natural days for a group of players of the same skill level across the entire server. For example, players of the same competitive rank (rank difference ≤ 1) or competitive score difference < 5% across the entire server. Optionally, to improve data accuracy, filtering conditions can be set, such as removing AFK matches and abnormal data matches (operation frequency < threshold).

[0040] Collect the effective operation score sequence of all players in the games that meet the above conditions, and record it as the server-wide effective operation score.

[0041] Furthermore, the core parameters are calculated, and the average operation score per minute is obtained based on the average score of all effective operations across the entire server (unit: points / minute). The operation fluctuation coefficient is obtained based on the standard deviation of the effective operation scores across the entire server. Optionally, the averaging operation in this application represents the average score of valid operations.

[0042] Furthermore, the operational intensity index is obtained by using the following formula for the Shanghai Standardized Index. :

[0043] The following describes the process of calculating the historical operation stability index using an operation performance stability evaluation framework built based on players' historical game data.

[0044] In this embodiment of the disclosure, the win dataset and the loss dataset are processed in a hierarchical manner. Based on the win dataset, the effective operation score sequence of all the player's wins is extracted; based on the loss dataset, the effective operation score sequence of all the player's losses is extracted.

[0045] Furthermore, key parameters are extracted. These key parameters include winning move characteristic values ​​and losing move characteristic values. The winning move characteristic values ​​include the average score of moves per point in winning games. And the volatility coefficient used to indicate the standard deviation of the player's winning action score. The losing action characteristic value includes the average score of actions scored per losing action for that player. And the volatility coefficient used to indicate the standard deviation of the player's losing play score. Alternatively, the stability can be quantified using the following formula to obtain the historical operational stability index. :

[0046] Alternatively, the following boundary condition constraints can be used: when hour, The default value is 0.

[0047] The following describes the process of calculating the operational resilience index, which is used to assess operational resilience in adversity. In other words, it assesses a player's ability to adjust their actions in adversity by monitoring changes in the game situation in real time.

[0048] In some implementations, a disadvantageous state in a game is determined by monitoring key scores and setting a disadvantage threshold. Optionally, key scores can be monitored using core indicators such as economic difference, kill difference, or map control rate. A disadvantage threshold can be set, for example, triggering a disadvantageous state when the deficit in a core indicator exceeds 15%. (Timestamp) This is the time point at which the first adversity was recorded.

[0049] Furthermore, operational sequence segmentation is performed on the dominant period sequence and the adversity period sequence. For the dominant period sequence, segmentation is performed from 0 to... Minutes of operation score; for adversity sequences, segmentation The score for each minute of operation up to the current time.

[0050] Furthermore, resilience is quantified by calculating the operational resilience index. The mean of the operational scores during the advantageous period is calculated based on the advantageous period and adverse period sequences. and standard deviation And the mean of the average operation score during the disadvantage period. and standard deviation The operational resilience index is obtained using the following formula. :

[0051] Optionally, data can be corrected using dynamic correction terms, for example, when the duration of adversity is less than 2 minutes. Multiply by a 0.8 attenuation factor.

[0052] In some implementations, the real-time operation intensity score is used to quantify the player's operation intensity level in real-time games. In this embodiment, the score ranges from 0 to 10, and the following formula can be used to quantify the real-time operation intensity score based on the operation intensity index and the historical operation stability index. :

[0053] in, This represents the standard normal cumulative function, which maps the combined exponent to a specified standardized interval (e.g., 0-1).

[0054] In some implementations, the adversity response performance score is used to assess a player's ability to adjust their actions under adverse conditions. In this embodiment, a score range of 0-10 is used, and the following formula can be employed to quantify the adversity response performance score based on the operational resilience index. :

[0055] Alternatively, extreme value processing can also improve data accuracy, for example, if The manual review process is automatically triggered at any time.

[0056] In some implementations, the operational mentality stability score is used to reflect the difference in operational stability between winning and losing games. In this embodiment, a score range of 0-10 is used, and the operational mentality stability score can be quantified using the following formula. :

[0057] Mental stability can be assessed using the following methods, for example, when At that time, the defeat compensation mechanism is activated. At that time, the operation fluctuation penalty mechanism is triggered.

[0058] The compensation coefficient matrix includes the loss compensation coefficient and the operation fluctuation penalty compensation coefficient, where the loss compensation coefficient is... ,according to Multiply by the loss compensation factor to get the compensated result. The operational volatility penalty compensation coefficient is: ,according to Subtracting the loss compensation factor, we get the compensated result. .

[0059] Alternatively, the boundary conditions can be adjusted accordingly. Apply constraints, for example, Gradient compression is performed when the value exceeds the limit.

[0060] Based on dynamic weights, obtain the evaluation value of operation density. The operational real-time density stability score can be obtained from the following formula: if This indicates stable historical operations, increasing the weighting of immediate intensity. The formula for the total operation score is:

[0061] Here, 0.4 and 0.35 represent weights, which can also be other values ​​in other implementations.

[0062] if This indicates that historical operational fluctuations are significant, and the weighting of adversity response should be increased. The formula for the overall operational score is:

[0063] Here, 0.4 and 0.3 represent weights, which can also be other values ​​in other implementations.

[0064] In this embodiment, the stability of real-time operation density is measured in terms of operation intensity by using an operation density index evaluation value. This can improve the accuracy of surrender decisions and significantly enhance competitive fairness and the gaming experience.

[0065] Figure 3 This is a flowchart of a real-time dynamic game behavior decision-making method according to an embodiment of this application, such as... Figure 3 As shown, the process of constructing an evaluation value for operational accuracy based on multidimensional game behavior characteristics includes the following steps: S301, based on multi-dimensional game behavior characteristics, obtains the total number of non-targeting operations and the number of hits of the target within the preset time range of the current game, and calculates the real-time hit rate.

[0066] S302: Based on multi-dimensional game behavior characteristics, obtain the historical hit data of the target object in the most recent specified game, and calculate the historical hit rate using time decay weighting.

[0067] S303 calculates a standardized score for real-time hit rate based on the mean and standard deviation of the hit rate over the same time range of the most recent specified game.

[0068] S304 calculates a standardized score for historical hit rate based on real-time hit rate and historical hit rate.

[0069] S305: Calculate the adverse hit response bonus score based on the change in hit rate before and after the current adverse state in the game.

[0070] S306, based on the comparison results of real-time hit rate and historical hit rate, dynamically allocates the weights of real-time hit rate standardized score, historical hit rate standardized score and adversity hit response bonus score to generate an indicator evaluation value of operational accuracy.

[0071] For example, in a game, the number of times a player performs a key non-targeted action (non-targeted, meaning not guaranteed to hit) and the number of successful hits are recorded as hit data. Hit data from the player's most recent 3 minutes of the game is collected and recorded as the player's real-time hit data. The number of hits is... Total number of times The real-time hit rate can be determined based on the ratio of the number of hits to the total number of hits. Optionally, the preset time range can be 3 minutes or other time ranges, which are not limited in this application.

[0072] Furthermore, a historical decay weighted pool is constructed. The time decay function can be represented by the following formula:

[0073] Wherein, λ is a hyperparameter preset based on prior knowledge. Optionally, in this application, λ = 0.1 / day. In other implementations, it can also be other values, and this application does not impose any restrictions on it. For the first The number of days since the match, in days.

[0074] The weighted pool is constructed with complete match data for the player's most recently specified matches (e.g., the most recent 50 matches). In case of an exception where the most recently specified matches are less than 50, all matches are taken as the most recently specified matches.

[0075] The historical hit rate is calculated using the following formula. :

[0076] in, These are weighting coefficients. For the first Number of hits in a game For the first The number of misses in a game.

[0077] Optionally, the standardized score for real-time hit rate can be obtained using the following formula. :

[0078] in, and These are the mean and standard deviation of the hit rate of players using the same character in the same time period over the last 50 matches. Φ is the standard normal distribution function.

[0079] Optionally, a standardized score for historical hit rates can be obtained using the following formula. :

[0080] The following settings use dynamic weights based on the relationship between real-time hit rate and historical hit rate to adjust... .

[0081] if / Greater than This indicates a high number of hits, and a certain number of points will be awarded as compensation. ,at this time, Add the compensation score to get the compensated score.

[0082] if / Less than This indicates that the current hit count is low, and a certain number of points will be deducted. Multiply by a preset factor (such as 0.8) to obtain the compensated score.

[0083] The following calculates the Adversity Hit Response Bonus score. ,in, This refers to the current game match. The player's hit rate afterwards This refers to the current game match. The hit rate of players previously.

[0084] like This indicates that the player's hit rate will increase in adverse situations, through a modifier (for example, the modifier at this time could be...). )right Make corrections, at this point, Updated to itself and correction factor The sum of .

[0085] Furthermore, assign contextual weights. if Greater than The formula is:

[0086] Here, 0.4 and 0.3 are pre-configured weights, and can be other values ​​in other implementations. For dynamic weights, the formula is:

[0087] in This represents the current start time of the game. Set the preset match duration.

[0088] In this embodiment, operational accuracy is measured in terms of operational performance through an evaluation value for operational precision. This can capture real-time fluctuations in the player's state and reduce the false judgment rate.

[0089] Figure 4 This is a flowchart of a real-time dynamic game behavior decision-making method according to an embodiment of this application, such as... Figure 4 As shown, the process of constructing an evaluation value for survival behavior based on multidimensional game behavior characteristics includes the following steps: S401: Based on multi-dimensional game behavior characteristics, obtain the target object's real-time survival behavior score in the current game and the target object's historical survival behavior score in the most recent specified game, and calculate the historical baseline score.

[0090] S402. Based on multidimensional game behavior characteristics, obtain the survival score distribution of game matches for all players of the same skill level across the entire server, and calculate the server-wide baseline score for game matches for all players of the same skill level across the entire server.

[0091] S403: Based on multi-dimensional game behavior characteristics, obtain the average survival score of the target team and the opposing team, and calculate the survival suppression score.

[0092] S404: Based on multidimensional game behavior characteristics, obtain the duration of the disadvantage period and the number of survivals in a specified state, and calculate the adversity compensation coefficient.

[0093] S405 generates an evaluation value for survival behavior based on historical baseline score, server-wide baseline score, survival suppression score, and adversity compensation coefficient.

[0094] The game defines several in-game survival behaviors, such as skill evasion, potion use, shield use, and status removal. These behaviors reflect a player's inclination to continue the game rather than surrender. Each behavior corresponds to a certain score, and the sum of the scores for the player's survival actions over a period of time is recorded as the player's survival behavior score. In a game, the player's survival behavior score within the first five minutes of the current time point is recorded as the real-time player survival behavior score. The real-time player survival behavior score for the current game is recorded as... .

[0095] Furthermore, calculate the historical baseline score. Let the current time point of this game be denoted as . Collect data from the player's most recent specified matches (e.g., the past 50 matches with the same character). The real-time historical data of player survival at a given point in time is denoted as , Refers to the first A game.

[0096] No. The weight of the game is ,in, It is the first The number of days since the last match.

[0097] The formula for calculating the historical mean is: .

[0098] Historical fluctuations are .

[0099] The standardized score of the historical benchmark is .

[0100] The following describes the process of calculating the server-wide baseline score, which is used to indicate the survival benchmark for players of the same rank across the entire server.

[0101] Collect data from all matches played within a specified timeframe (e.g., the past 30 days) across the entire server, comparing players of the same skill level. Real-time player survival score list at each point in time Sort the list in ascending order and calculate the mean. and standard deviation Then calculate Percentage in the distribution of this list .

[0102] The standardized score of the server-wide baseline score can be expressed as:

[0103] Where Φ is the standard normal distribution function.

[0104] The following describes the process of calculating survival suppression points, collecting data from each player on both the player's and the opposing team's teams. The real-time player survival behavior scores at each point in time are used to calculate the average score of both sides. and and relative advantages :

[0105] when At that time, the standardized score of the survival suppression score was:

[0106] when At that time, the standardized score of the survival suppression score was:

[0107] Here, 5 indicates that the team has 5 people. In other implementations, it can also be other values, and this application does not restrict it.

[0108] The adversity compensation mechanism of this application is described below. The adversity compensation coefficient may include... , In the embodiments of this application, (Duration of disadvantage / Current game time) (Number of times a character successfully survives when their health is below the health threshold / Number of times a character's health is below the health threshold).

[0109] Based on preset weights, the basic synthetic survival behavior is divided into:

[0110] After standardization using a score range of 0-10, the evaluation value of survival behavior can be expressed as:

[0111] In this embodiment of the application, the survival behavior score is measured in the tactical survival dimension by the evaluation value of the survival behavior index, which helps to improve the accuracy of surrender decisions, reduce the misjudgment rate, and optimize the player experience.

[0112] Figure 5 This is a flowchart of a real-time dynamic game behavior decision-making method according to an embodiment of this application, such as... Figure 5 As shown, the process of constructing an evaluation value for historical resilience based on multidimensional game behavior characteristics includes the following steps: S501, based on multi-dimensional game behavior characteristics, obtains the surrender rate, surrender rate, comeback rate, and average duration of losing games of the target object in the most recent specified game.

[0113] S502 generates a historical resilience index value by weighting the surrender rate, the agreed surrender rate, the turnaround rate, and the average duration of the losing game.

[0114] In some implementations, data from a player's most recent specified matches (e.g., the last 100 games) is collected. If no player has fewer than 100 matches, data from all matches is collected to obtain four metrics, including the surrender rate, the surrender rate, the comeback rate, and the average duration of losing games.

[0115] The surrender rate is calculated by dividing the number of matches in which players initiated surrender by the total number of matches counted. ; The surrender rate is calculated by dividing the number of matches in which a player agreed to their teammate's surrender by the total number of matches counted. ; A game in which a player's team falls behind the opposing team by a significant margin in crucial moments is called a losing game. A game in which the player ultimately wins is called a comeback game. Comeback rate. The number of comeback games divided by the number of games trailing.

[0116] In a losing game, the duration from when a player's team falls behind the opponent by a certain margin in crucial scores until the game ends is called the losing game duration. The percentage of the losing game duration to the total game time is called the losing game duration percentage. The average losing game duration percentage... It is the average duration of losing games in the statistically analyzed matches.

[0117] Standardized scores are calculated based on the surrender rate, the rate of agreeing to surrender, the comeback rate, and the average duration of lasting games in losing games, resulting in deductions for surrender behavior, bonuses for comeback rate, and bonuses for duration of lasting games.

[0118] The deduction criteria for surrender behavior include the standard value for initiating surrender and the standard value for agreeing to surrender. The standard value for initiating surrender is... The standard value for agreeing to surrender is .

[0119] The formula for the comeback rate bonus is as follows: .

[0120] The bonus points for persistence are: .

[0121] The score is calculated by weighting the deductions for surrender, the bonuses for comeback rate, and the bonuses for duration of perseverance, thus generating an evaluation value for historical resilience. .

[0122] In this embodiment, historical resilience is measured in terms of long-term stress resistance through a historical resilience index evaluation value. This upgrades surrender decision-making from traditional win rate prediction to behavior-driven intelligent adjudication, which is beneficial to improving the accuracy of surrender decisions.

[0123] Figure 6 This is a flowchart of a real-time dynamic game behavior decision-making method according to an embodiment of this application, such as... Figure 6 As shown, the method includes the following steps: S601 calculates the basic score of game behavior based on the evaluation value of each indicator.

[0124] S602 generates a balance factor based on the mean and standard deviation of the evaluation values ​​of each indicator.

[0125] S603 uses the product of the basic game behavior score and the balance factor as the game behavior evaluation value of the target object.

[0126] S604. Based on the game behavior evaluation value of each target object, obtain the mean and standard deviation of the game behavior evaluation value, and calculate the relative effectiveness index.

[0127] S605, calculate dynamic decision weights based on the relative effectiveness index of each target object and the basic weight coefficient of the role type.

[0128] Calculate the basic score for game behavior based on the evaluation value of each indicator. for:

[0129] Furthermore, in this embodiment, a dynamic balancing factor is introduced to ensure the balance of reward dimensions, thereby preventing "specialized" players from obtaining artificially high scores.

[0130] Calculate the average of the four scores. :

[0131] The value of the balance factor for:

[0132] The product of the base game behavior score and the balance factor is used as the game behavior evaluation value for the target:

[0133] The following describes the team collaboration scoring and decision-making process for this application.

[0134] Assuming the player's team includes Named target (i.e., player) , ,..., Each player's game behavior evaluation score (i.e., overall score) .

[0135] For players Define player Relative efficiency index:

[0136] in, It is the average score of the team's players' game behavior evaluation. It is the standard deviation of the team's player game behavior evaluation scores. Player The player's game behavior evaluation score, i.e. of .

[0137] For players Controlled character The base weighting coefficient is determined by the role's positioning (e.g., damage-dealing character A=1.1, support character A=0.9). Calculate the player's... Dynamic weighting coefficients: if If less than or equal to 0, then:

[0138] if If it is greater than 0, then:

[0139] in, The player who initiates the surrender is called Player. ; The player who initiated the surrender was not a player. ; Players who initiate surrender The base weight coefficient of the controlled character.

[0140] The decision evaluation value corresponding to the surrender instruction, i.e., the basic team score, is obtained as follows:

[0141] in, For efficiency convergence parameters, Strengthen the influence of high-scoring players. To mitigate the impact of extreme values, and given that players sometimes use voice chat to team up, the settings can be lowered. Players enter the game solo, and settings can be set to higher. .

[0142] A threshold can be set for agreeing to a player initiating a surrender. ,when At that time, the player Only when a surrender vote is initiated can it be passed and other players be allowed to vote; when At that time, the player A surrender vote will be rejected, and a surrender vote cannot be initiated again within five minutes.

[0143] This application addresses issues such as the single-dimensionality approach, rigid static thresholds, and lack of team collaboration in existing game behavior decision-making, enabling real-time dynamic surrender decisions based on multi-dimensional player behavioral characteristics. Through multi-dimensional behavioral data collection and quantitative analysis, this application upgrades surrender decisions from traditional win rate prediction to behavior-driven intelligent adjudication, significantly improving decision accuracy and player experience.

[0144] In terms of technical implementation, a four-factor evaluation model is first constructed, encompassing operational density, operational precision, survival behavior, and historical resilience. Operational density quantifies player activity through the effective frequency of operations per unit time, while operational precision assesses execution capability based on the deviation between the hit rate of non-targeted skills and historical benchmarks. The survival behavior module collects players' tactical actions such as skill avoidance and resource management, and combines this data with server-wide data from players of the same rank to generate a standardized score. The historical resilience dimension comprehensively considers surrender rate, comeback rate, and duration of play to characterize a player's long-term resilience under pressure. The four-factor data are dynamically weighted to form an individual resilience value, with weight allocation taking into account both real-time status and historical performance to avoid inflated scores for players with uneven performance.

[0145] At the team decision-making level, a relative effectiveness index and dynamic weights are introduced. Based on the deviation between the player's overall score and the team average, individual decision weights are dynamically adjusted to ensure that team decisions consider both individual performance and synergistic effects. For example, damage-dealing characters receive increased weight due to the accuracy of their key skills, while support characters receive additional bonuses for their contributions to survival. Ultimately, the team's overall resilience index is compared to a preset threshold to determine whether a surrender request is valid.

[0146] The technical benefits of this solution are reflected in three aspects: First, it improves decision-making accuracy by capturing real-time fluctuations in player status through a four-factor model, reducing the misjudgment rate. Second, it optimizes the player experience by preventing forced voting due to short-term disadvantages through an adversity resilience assessment mechanism, while survival behavior scoring incentivizes proactive defensive strategies. Third, it enhances the system's resistance to interference by filtering out accidental actions through historical resilience scoring and adapting dynamic weights to the needs of different game stages. The optimized mechanism can improve the accuracy of adjudication, extend the average game duration, and significantly improve competitive fairness and the overall gaming experience.

[0147] Figure 7 This is a structural block diagram of a real-time dynamic game behavior decision-making device according to an embodiment of this application, as shown below. Figure 7 As shown, the real-time dynamic game behavior decision-making device 700 includes: The data acquisition module 710 is used to collect multi-dimensional game behavior features in response to receiving a surrender command. The multi-dimensional game behavior features include at least historical behavior data of past game matches, real-time behavior data of the current game match, and team collaborative behavior data. Module 720 is used to construct evaluation values ​​for multiple indicators for each target object in the team based on multi-dimensional game behavior characteristics. These multiple indicators include at least operation density, operation accuracy, survival behavior, and historical resilience. The acquisition module 730 is used to dynamically weight the game behavior evaluation value of the target object based on the evaluation value of each indicator. The decision module 740 is used to obtain the relative effectiveness index and dynamic decision weight of each target object, obtain the decision evaluation value corresponding to the surrender command based on the game behavior evaluation value, the relative effectiveness index and the dynamic decision weight, and process the surrender command based on the decision evaluation value.

[0148] In some implementations, the construction module 720 is also used for: Based on multidimensional game behavior characteristics, the effective operation score of the target object within a unit of time and the mean and standard deviation of the effective operation score of the game match of the same level object group in the whole server are obtained, and the operation intensity index is calculated. Based on multidimensional game behavior characteristics, the mean and standard deviation of the effective operation scores in the target object's historical wins and losses are obtained, and the historical operation stability index is calculated. Based on multidimensional game behavior characteristics, the mean and standard deviation of the effective operation scores of the target object in the advantageous and disadvantageous periods are obtained, and the operation resilience index is calculated. Based on the operational intensity index, historical operational stability index, and operational resilience index, calculate the immediate operational intensity score, operational mindset stability score, and adversity response effectiveness score, respectively. Based on the positive or negative historical operational stability index, the weights of the immediate operational intensity score, operational mindset stability score, and adversity response effectiveness score are dynamically adjusted to generate an evaluation value for operational density.

[0149] In some implementations, the construction module 720 is also used for: Based on multidimensional game behavior characteristics, obtain the total number of non-targeting actions and the number of hits of the target within the preset time range of the current game, and calculate the real-time hit rate; Based on multidimensional game behavior characteristics, obtain the historical hit data of the target object in the most recent specified game, and calculate the historical hit rate using time decay weighting. Calculate the standardized score for real-time hit rate based on the mean and standard deviation of the hit rate within the same time range of the most recent specified game. Calculate a standardized score for historical hit rate based on real-time hit rate and historical hit rate; Calculate the adverse hit response bonus score based on the change in hit rate before and after the current adverse state in the game. Based on the comparison between real-time hit rate and historical hit rate, the weights of the real-time hit rate standardized score, historical hit rate standardized score, and adversity hit response bonus score are dynamically allocated to generate an evaluation value of operational accuracy.

[0150] In some implementations, the construction module 720 is also used for: Based on multidimensional game behavior characteristics, obtain the target object's real-time survival behavior score in the current game and the target object's historical survival behavior score in the most recent specified game, and calculate the historical baseline score; Based on multidimensional game behavior characteristics, obtain the survival score distribution of game matches for all players of the same skill level across the entire server, and calculate the server-wide baseline score for game matches for all players of the same skill level across the entire server. Based on multidimensional game behavior characteristics, the average survival score of the target team and the opposing team is obtained, and the survival suppression score is calculated. Based on multidimensional game behavior characteristics, the duration of the disadvantage period and the number of survivals in a specified state are obtained, and the adversity compensation coefficient is calculated. Based on historical baseline scores, server-wide baseline scores, survival suppression scores, and adversity compensation coefficients, an evaluation value for survival behavior is generated.

[0151] In some implementations, the construction module 720 is also used for: Based on multidimensional game behavior characteristics, obtain the surrender rate, surrender rate, comeback rate, and average duration of losing games of the target object in the most recent specified game. The historical resilience index is generated by weighting the surrender rate, the agreed surrender rate, the turnaround rate, and the average duration of the losing game.

[0152] In some implementations, the acquisition module 730 is further configured to: A basic score for game behavior is calculated based on the evaluation value of each indicator. A balance factor is generated based on the mean and standard deviation of the evaluation values ​​of each indicator. The product of the basic game behavior score and the balance factor is used as the game behavior evaluation value of the target object.

[0153] In some implementations, the decision module 740 is further configured to: Based on the game behavior evaluation value of each target object, the mean and standard deviation of the game behavior evaluation value are obtained, and the relative effectiveness index is calculated. Dynamic decision weights are calculated based on the relative effectiveness index of each target object and the basic weight coefficient of the role type.

[0154] In some implementations, the decision module 740 is further configured to: If the decision evaluation value is greater than or equal to the preset decision evaluation value threshold, a vote is allowed, and the surrender command is processed according to the voting results. If the decision evaluation value is less than the decision evaluation value threshold, the surrender command is rejected and a cooldown period is set.

[0155] This application constructs a four-factor evaluation value based on multi-dimensional data collection, including real-time operations, historical behavior, and team collaboration. This value assesses operational density, operational accuracy, survival behavior, and historical resilience, capturing real-time fluctuations in player status and reducing misjudgment rates. By combining relative efficiency indices and dynamic weights at the team decision-making level, it upgrades game behavior decision-making from traditional win rate prediction to behavior-driven intelligent adjudication, significantly improving decision-making accuracy and player experience.

[0156] Figure 8 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application.

[0157] like Figure 8 As shown, the electronic device 800 includes: The memory 801 and the processor 802 are connected by a bus 803, which connects different components (including the memory 801 and the processor 802). The memory 801 stores a computer program. When the processor 802 executes the program, it implements the real-time dynamic game behavior decision-making method of this application embodiment.

[0158] Bus 803 represents one or more of several bus architectures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of the various bus architectures. For example, these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect (PCI) bus.

[0159] Electronic device 800 typically includes a variety of electronic device readable media. These media can be any available media that can be accessed by electronic device 800, including volatile and non-volatile media, removable and non-removable media.

[0160] Memory 801 may also include computer system readable media in the form of volatile memory, such as random access memory (RAM) 804 and / or cache memory 805. Electronic device 800 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 806 may be used to read and write non-removable, non-volatile magnetic media (… Figure 8 Not shown; usually referred to as a "hard drive"). Although Figure 8 As not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk") and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 803 via one or more data media interfaces. Memory 801 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of this application.

[0161] A program / utility 808 having a set (at least one) of program modules 807 may be stored, for example, in memory 801. Such program modules 807 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment. Program modules 807 typically perform the functions and / or methods described in the embodiments of this application.

[0162] Electronic device 800 can also communicate with one or more external devices 809 (e.g., keyboard, pointing device, display 811, etc.), and with one or more devices that enable a user to interact with the electronic device 800, and / or with any device that enables the electronic device 800 to communicate with one or more other computing devices (e.g., network card, modem, etc.). This communication can be performed through input / output (I / O) interface 812. Furthermore, electronic device 800 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 813. Figure 8 As shown, network adapter 813 communicates with other modules of electronic device 800 via bus 803. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with electronic device 800, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0163] The processor 802 executes various functional applications and data processing by running programs stored in the memory 801.

[0164] It should be noted that the implementation process and technical principles of the electronic device in this embodiment are explained in the foregoing description of the real-time dynamic game behavior decision-making method of this application embodiment, and will not be repeated here.

[0165] To implement the above embodiments, this application also proposes a computer-readable storage medium.

[0166] When the instructions in the computer-readable storage medium are executed by the processor of the electronic device, the electronic device is able to execute the aforementioned real-time dynamic game behavior decision-making method. Optionally, the computer-readable storage medium may be ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0167] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the claims.

[0168] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A real-time dynamic game behavior decision-making method, characterized in that, include: In response to receiving a surrender command, multi-dimensional game behavior features are collected, including at least historical behavior data from past game matches, real-time behavior data from the current game match, and team collaborative behavior data. For each target object in the team, based on the multidimensional game behavior characteristics, a multi-indicator evaluation value is constructed, which includes at least operation density, operation accuracy, survival behavior and historical resilience. The game behavior evaluation value of the target object is obtained by dynamically weighting the evaluation value of each indicator. Obtain the relative effectiveness index and dynamic decision weight for each target object. Based on the game behavior evaluation value, the relative effectiveness index, and the dynamic decision weight, obtain the decision evaluation value corresponding to the surrender instruction, and process the surrender instruction based on the decision evaluation value.

2. The method according to claim 1, characterized in that, The process of constructing an evaluation value for operation density based on the aforementioned multidimensional game behavior characteristics includes: Based on the multidimensional game behavior characteristics, the effective operation score of the target object within a unit of time and the mean and standard deviation of the effective operation score of the game match of the same level object group in the whole server within a unit of time are obtained, and the operation intensity index is calculated. Based on the multidimensional game behavior characteristics, the mean and standard deviation of the effective operation scores in the historical wins and losses of the target object are obtained, and the historical operation stability index is calculated. Based on the multidimensional game behavior characteristics, the mean and standard deviation of the effective operation scores of the target object in the advantageous and disadvantageous periods are obtained, and the operation resilience index is calculated. Based on the operational intensity index, historical operational stability index, and operational resilience index, calculate the immediate operational intensity score, operational mindset stability score, and adversity response effectiveness score, respectively. Based on the positive or negative historical operational stability index, the weights of the immediate operational intensity score, the operational mindset stability score, and the adversity response effectiveness score are dynamically adjusted to generate an evaluation value for the operational density index.

3. The method according to claim 1, characterized in that, The process of constructing an evaluation value for operational accuracy based on the aforementioned multidimensional game behavior characteristics includes: Based on the multi-dimensional game behavior characteristics, the total number of non-targeting operations and the number of hits of the target within the current game's preset time range are obtained, and the real-time hit rate is calculated. Based on the multidimensional game behavior characteristics, the historical hit data of the target object in the most recent specified game is obtained, and the historical hit rate is calculated using time decay weighting. A standardized score for the real-time hit rate is calculated based on the mean and standard deviation of the hit rate within the same time range of the most recent specified game. Calculate a standardized score for the historical hit rate based on the real-time hit rate and the historical hit rate; Calculate the adverse hit response bonus score based on the change in hit rate before and after the current adverse state in the game. Based on the comparison between real-time hit rate and historical hit rate, the weights of the real-time hit rate standardized score, the historical hit rate standardized score, and the adversity hit response bonus score are dynamically allocated to generate an evaluation value for the operational accuracy.

4. The method according to claim 1, characterized in that, The process of constructing an evaluation value for survival behavior based on the aforementioned multidimensional game behavior characteristics includes: Based on the multidimensional game behavior characteristics, obtain the target object's real-time survival behavior score in the current game and the target object's historical survival behavior score in the most recent specified game, and calculate the historical baseline score; Based on the multidimensional game behavior characteristics, obtain the survival score distribution of the game matches of the same skill level group across the entire server, and calculate the server-wide baseline score of the game matches of the same skill level group across the entire server. Based on the multidimensional game behavior characteristics, the average survival score of the target team and the opposing team is obtained, and the survival suppression score is calculated. Based on the multidimensional game behavior characteristics, the duration of the disadvantage period and the number of survivals in a specified state are obtained, and the adversity compensation coefficient is calculated. Based on the historical baseline score, the server-wide baseline score, the survival suppression score, and the adversity compensation coefficient, an evaluation value for survival behavior is generated.

5. The method according to claim 1, characterized in that, The process of constructing an evaluation value for historical resilience based on the aforementioned multidimensional game behavior characteristics includes: Based on the multidimensional game behavior characteristics, obtain the surrender rate, surrender rate, comeback rate, and average duration of losing games of the target object in the most recent specified game. The historical resilience index is generated by weighting the surrender rate, the surrender rate, the comeback rate, and the average duration of the losing game.

6. The method according to any one of claims 1-5, characterized in that, The step of dynamically weighting the evaluation values ​​of each indicator to obtain the game behavior evaluation value of the target object includes: A basic score for game behavior is calculated based on the evaluation value of each indicator. A balance factor is generated based on the mean and standard deviation of the evaluation values ​​of each indicator. The product of the basic game behavior score and the balance factor is used as the game behavior evaluation value of the target object.

7. The method according to any one of claims 1-5, characterized in that, The step of obtaining the relative performance index and dynamic decision weight for each target object includes: Based on the game behavior evaluation value of each target object, the mean and standard deviation of the game behavior evaluation value are obtained, and the relative efficiency index is calculated. The dynamic decision weights are calculated based on the relative performance index and the basic weight coefficient of the role type for each target object.

8. The method according to any one of claims 1-5, characterized in that, The process of processing the surrender instruction based on the decision evaluation value includes: If the decision evaluation value is greater than or equal to a preset decision evaluation value threshold, a vote is allowed, and the surrender command is processed according to the voting results. If the decision evaluation value is less than the decision evaluation value threshold, the surrender instruction is rejected and a cooldown period is set.

9. A real-time dynamic game behavior decision-making device, characterized in that, include: The data acquisition module is used to collect multi-dimensional game behavior features in response to receiving a surrender command. The multi-dimensional game behavior features include at least historical behavior data of past game matches, real-time behavior data of the current game match, and team collaborative behavior data. The module is used to construct evaluation values ​​for multiple indicators for each target object in the team based on the multidimensional game behavior characteristics. The multiple indicators include at least operation density, operation accuracy, survival behavior and historical resilience. The acquisition module is used to dynamically weight the evaluation values ​​of each indicator to obtain the game behavior evaluation value of the target object; The decision module is used to obtain the relative effectiveness index and dynamic decision weight of each target object, obtain the decision evaluation value corresponding to the surrender instruction based on the game behavior evaluation value, the relative effectiveness index and the dynamic decision weight, and process the surrender instruction based on the decision evaluation value.

10. An electronic device, characterized in that, include: processor; Memory for storing the executable instructions of the processor; The processor is configured to execute the instructions to implement the method as described in any one of claims 1-8.

11. A computer-readable storage medium, characterized in that, When the instructions in the computer-readable storage medium are executed by the processor of the electronic device, the electronic device is able to perform the steps of the method as described in any one of claims 1-8.

12. A computer program product comprising a computer program that, when executed by a processor, implements the method according to any one of claims 1-8.

Images