Game cheat detection method, device, electronic device and storage medium

Abstract

The embodiments of the present application disclose a method, device, electronic device and storage medium for detecting game cheats. They are applied to the computer field. They include: obtaining a historical game behavior sequence corresponding to a game account. Determining a state transition coordinate sequence corresponding to the historical game behavior sequence based on a Markov chain state transition space coordinate system. Determining the information entropy corresponding to the Markov chain state transition space based on the distribution of at least one coordinate sequence sample in the Markov chain state transition space coordinate system. Determining the cheating status of the game account based on the information entropy corresponding to the Markov chain state transition space. The present invention is based on the characteristic that the game behavior transfer of a cheating account must show regularity, and directly determines the cheating status of a game account based on the process of game behavior transfer, so that the cheating detection process cannot be easily circumvented by game cheats.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method, device, electronic device and storage medium for detecting game cheats. Background Art

[0002] In recent years, the global gaming market has continued to expand alongside technological advancements. Within this market, the proper operation of games is closely tied to a balanced gaming environment. A balanced gaming environment means that the more time and effort players invest in the game, the stronger their characters become. However, the emergence of cheating tools, known as game plug-ins, has disrupted this balance. These tools allow characters to automatically achieve profitable goals and become stronger without player control. To maintain game fairness, timely detection and crackdown on cheats during gameplay is crucial to ensuring the proper operation of games.

[0003] There are two main methods for detecting cheats in the existing technology. One method relies on physical methods such as device information analysis, risk environment detection, and suspicious process detection. The other method identifies cheats by analyzing differences in player behavior.

[0004] In the above existing technologies, physical cheat detection requires extensive expert experience and prior knowledge to construct detection rules, which is time-consuming and labor-intensive. Detection based on analyzing differences in player behavior only considers the behavioral differences between normal players and cheating players. This oversimplifies the considerations in the detection process, allowing cheat detection to be evaded with simple circumvention measures. Therefore, efficient and concise cheat detection has become a pressing issue. Summary of the Invention

[0005] Embodiments of the present invention provide a method, device, electronic device, and storage medium for detecting game cheats. During cheat detection, the player's historical game behavior sequence can be used to determine the game behavior transitions, and based on these transitions, the player can be analyzed to determine whether they are cheating. Because a player's historical game behavior sequence cannot be tampered with, the game behavior transitions can be used to accurately analyze whether the player's game behavior meets the characteristics of game behavior in a cheating state, thereby efficiently and accurately completing game cheat detection.

[0006] A first aspect of the present invention provides a method for detecting game cheats, comprising:

[0007] Get the historical game action sequence corresponding to the game account. The historical game action sequence is used to indicate the time sequence of game actions performed by the game account within the historical time period.

[0008] A state transfer coordinate sequence corresponding to a historical game behavior sequence is determined according to a Markov chain state transfer space coordinate system, where the state transfer coordinate sequence includes at least one coordinate sequence sample.

[0009] The information entropy corresponding to the Markov chain state transition space is determined according to the distribution status of at least one coordinate sequence sample in the Markov chain state transition space coordinate system.

[0010] The cheating status of the game account is determined based on the information entropy corresponding to the Markov chain state transition space.

[0011] A second aspect of an embodiment of the present application provides a game cheat detection device, the game cheat detection device comprising:

[0012] The acquisition unit is used to acquire a historical game behavior sequence corresponding to the game account, where the historical game behavior sequence is used to indicate the time sequence of all game behaviors performed by the game account within a historical time period.

[0013] The determining unit is configured to determine a state transition coordinate sequence corresponding to a historical game behavior sequence based on a Markov chain state transition space coordinate system. The state transition coordinate sequence includes at least one coordinate sequence sample.

[0014] The calculation unit is used to determine the information entropy corresponding to the Markov chain state transition space according to the distribution of at least one coordinate sequence sample in the Markov chain state transition space coordinate system.

[0015] The determination unit is further used to determine the cheating status of the game account based on the information entropy corresponding to the Markov chain state transition space.

[0016] A third aspect of an embodiment of the present application provides an electronic device, including: a memory and a processor, wherein the memory and the processor are coupled.

[0017] The memory is used to store one or more computer instructions.

[0018] The processor is used to execute one or more computer instructions to implement the game cheat detection method described in the first aspect above.

[0019] A fourth aspect of an embodiment of the present application provides a computer-readable storage medium having one or more computer instructions stored thereon, which are executed by a processor to implement a method for detecting game plug-ins as described in the first aspect above.

[0020] The technical solution provided in the embodiments of the present application, when performing cheat detection, first obtains the historical game behavior sequence corresponding to the game account to be detected. This historical game behavior sequence is then converted into a state transition coordinate sequence based on the Markov chain state transition space coordinate system. The information entropy corresponding to the Markov chain state transition space is then determined based on the distribution of coordinate sequence samples in the state transition coordinate sequence within the Markov chain state transition space coordinate system. Finally, based on the information entropy corresponding to the Markov chain state transition space, the historical game behavior of the game account is evaluated to determine the cheating status of the game account.

[0021] The above embodiment applies the properties of Markov chains to evaluate a player's historical gaming behavior. By constructing a Markov chain state transition space coordinate system to observe the transition state of gaming behavior, the state transition coordinate sequence is used to determine information entropy, and then based on the information entropy, it is determined whether the player's gaming behavior is disordered and random. If the transition of a player's historical gaming behavior is not random and has a specific and obvious transition pattern, then it indicates that the game account is in a cheating state. Because historical gaming behavior cannot be tampered with and detection of historical gaming behavior cannot be prevented, the cheat detection method based on historical gaming behavior can accurately analyze whether a player is in a cheating state. Moreover, by using the transition of historical gaming behavior, it can be determined whether the player's gaming behavior transitions are regular, thereby simply and efficiently determining whether the player has cheated in the game. In this way, subsequent processing can be carried out based on the detection results to maintain the fairness of the game and improve the gaming experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a flowchart of a method for detecting game cheats provided in an embodiment of the present application;

[0023] Figure 2 A flowchart of a method for obtaining a status game behavior sequence corresponding to a game account provided in an embodiment of the present application;

[0024] Figure 3 A flowchart of a method for calculating information entropy provided in an embodiment of the present application;

[0025] Figure 4 A flowchart of another method for calculating information entropy provided in an embodiment of the present application;

[0026] Figure 5 This is a structural diagram of a game cheat detection device provided by the present invention;

[0027] Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0028] Embodiments of the present invention provide a method, device, electronic device, and storage medium for detecting game cheats. During cheat detection, the player's historical game behavior sequence can be used to determine the game behavior transitions, and based on these transitions, the player can be analyzed to determine whether they are cheating. Because a player's historical game behavior sequence cannot be tampered with, the game behavior transitions can be used to accurately analyze whether the player's game behavior meets the characteristics of game behavior in a cheating state, thereby efficiently and accurately completing game cheat detection.

[0029] In order to enable those skilled in the art to better understand the technical solutions of this application, the following clearly and completely describes this application in conjunction with the drawings in the embodiments of this application. However, this application can be implemented in many other ways different from the above description. Therefore, based on the embodiments provided in this application, all other embodiments obtained by those skilled in the art without creative work should fall within the scope of protection of this application.

[0030] It should be noted that the terms "first", "second", "third", etc. in the claims, description and drawings of the present application are used to distinguish similar objects and are not used to describe a specific order or sequence. The data used in this way are interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in a sequence other than that illustrated or described herein. In addition, the terms "including", "having" and their variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0031] The continuous development of electronic information has led to the expansion of the gaming industry and a booming gaming market. Once a game enters the market, its longevity is closely tied to the balance of the gaming environment. Specifically, a more balanced and fair gaming environment—that is, a positive correlation between the time and effort players invest in developing their characters and the strength of their abilities—will attract more players. Conversely, attracting fewer players will eventually lead to the game's closure due to profitability issues.

[0032] However, cheating tools like game plug-ins can directly impact the balance of the gaming environment. They can allow game characters in a game account to automatically play the game and complete profitable in-game activities, thereby increasing their power without any player interaction. For example, game plug-ins can allow game characters to automatically participate in profitable game activities, automatically defeat monsters and level up, and tamper with health points and experience points during gameplay. This allows players using game plug-ins to directly increase the power of the game character corresponding to their game account without investing any time or effort. This can severely undermine the fairness of the game and affect the gaming experience of regular players. If left unchecked, this will lead to a decrease in the number of players. Therefore, detecting and punitively handling cheating players is crucial to maintaining a healthy gaming environment.

[0033] Current methods for detecting game cheats typically rely on physical methods such as device information analysis, risk environment monitoring, and suspicious process detection. For example, cheat detection can be performed by detecting suspicious processes or tampering with the game program on the client. However, this detection method requires extensive expert experience and prior knowledge to construct detection rules, which is time-consuming and labor-intensive. Furthermore, different games require different detection rules, making it impossible to develop a universal cheat detection method. Furthermore, cheats can evade detection by forging processes or blocking detection programs, resulting in poor detection effectiveness. Another approach is to identify cheats by analyzing differences in player behavior. Specifically, this method analyzes player behavior to detect cheats, but it typically only analyzes differences in player behavior and considers only the differences between normal players and players using cheats. For example, players without cheats can move freely in the game, attack any object, and wear a variety of equipment. Players with cheats, on the other hand, follow fixed attack routes, attack objects in a fixed order, and wear fixed equipment at fixed times. Since the above detection process only considers differences in specific game behaviors, it only requires setting up simple circumvention measures to evade cheat detection. Therefore, how to detect cheats efficiently and concisely has become an urgent problem that needs to be solved.

[0034] To address the aforementioned issues, embodiments of the present application provide a method, apparatus, electronic device, and storage medium for detecting game cheats. This method, when detecting game cheats, establishes a Markov chain state transition space coordinate system, converting the player's corresponding game behavior sequence into a state transition coordinate sequence. This sequence can be viewed as a Markov chain. The randomness and disorder of the state transition coordinate sequence are then analyzed based on the properties of the Markov chain to determine whether the game account is a cheating account.

[0035] The method, device, terminal, and computer-readable storage medium described in this application are further described in detail below with reference to specific embodiments and drawings.

[0036] Before introducing the embodiments of this application, a brief introduction to Markov chains is provided. Markov chains are random processes in probability theory and mathematical statistics that exhibit the Markov property and exist within a discrete set of exponentials and state space. The Markov property refers to the fact that, given the current state and all past states, the conditional probability distribution of a random process's future state depends solely on the current state and is independent of past states. In the embodiments of this application, if a user's gaming behaviors are arranged chronologically, the resulting sequence of gaming behaviors can be considered a Markov chain. Information entropy is a parameter used to measure the randomness and disorder of a Markov chain. Normal players typically engage in a wide variety of gaming behaviors during gaming, and the transitions between these behaviors are relatively random, lacking a fixed pattern, resulting in an overall appearance of "disorder" and "chaos." Therefore, the gaming behaviors of normal players exhibit a high-entropy distribution, characterized by a wide variety of behaviors and random transitions between them. However, cheaters, whose characters are controlled by automated programs, tend to focus on highly profitable gaming activities, such as daily activities and experience dungeons. Therefore, cheating players engage in a relatively small number of behaviors during gameplay, with numerous repetitive and similar combinations. The transitions between these behaviors exhibit distinct patterns, giving the overall appearance of being "ordered" and "clean." Generally, cheating players exhibit a low-entropy distribution, with a small number of behaviors and non-random transitions between them, but rather distinct patterns.

[0037] Based on the above description, Figure 1 A flowchart of a method for detecting game cheats is provided for an embodiment of the present application. It should be noted that the steps shown in the flowchart can be executed in a computer system, such as a set of computer-executable instructions, and in some cases, the steps shown can be executed in a different logical order than shown in the flowchart.

[0038] like Figure 1 As shown, the method for detecting game cheats includes the following steps:

[0039] 101. Obtain the historical game behavior sequence corresponding to the game account.

[0040] When testing a game account, you first need to obtain the historical game behavior sequence corresponding to the game account. The historical behavior sequence is the time sequence of all game behaviors performed by the game account within the historical time period.

[0041] Here, a game behavior refers to the actions taken by a game account to control a virtual character to participate in and complete game events while playing. For example, a game action can be defined as a game action, such as logging in, accepting a mission, or moving to a map. Alternatively, completing a complete game activity can be defined as a game action, such as completing a game mission, unlocking a game achievement, or obtaining a game reward. It should be understood that game actions can be defined based on the characteristics and rules of each game and are not specifically limited. A game action sequence is a collection of game actions sorted by the time they occurred. Specifically, to obtain a historical game action sequence corresponding to a game account, first retrieve the historical game actions that occurred for that game account over a period of time based on historical record data. For example, if a game account performed game actions A, B, and C over the past day, and game action A occurred at 10:00 AM, game action B at 5:00 PM, and game action C at 12:00 PM, then the historical game action sequence corresponding to that game account can be determined as {game action A, game action C, game action B}. It is understood that each game action can be recorded using a game action identifier. Within a game, a game action identifier corresponds to a unique game action. Then the above game behavior sequence can be {A, C, B}.

[0042] 102. Determine the state transition coordinate sequence corresponding to the historical game behavior sequence based on the Markov chain state transition space coordinate system.

[0043] After obtaining the game behavior sequence, it is necessary to convert it into a state transition coordinate sequence based on the Markov chain state transition space coordinate system. The purpose of this is to evaluate the randomness and disorder of the behavior transitions between each game behavior in the game behavior sequence based on the distribution of the state transition coordinates (coordinate sequence samples) in the Markov chain state transition space coordinate system. This allows us to determine whether there is a clear transition pattern in the game account's historical game behavior and ultimately determine whether the game account is using cheats.

[0044] 103. According to the distribution of coordinate sequence samples in the Markov chain state transition space coordinate system, the information entropy corresponding to the Markov chain state transition space is determined.

[0045] After obtaining the state transition coordinate sequence corresponding to the historical gaming behavior sequence in step 102, the information entropy corresponding to the Markov chain state transition space can be determined based on the distribution of the coordinate sequence samples in the Markov chain state transition space coordinate system. Information entropy is a fundamental concept in information theory, used to characterize the degree of disorder in information about a phenomenon. In this embodiment, this value is used to represent the randomness exhibited by the transitions in gaming behavior of a gaming account. More specifically, a larger value of this information entropy indicates a more random and disordered gaming account. A smaller value of this information entropy indicates a more orderly and organized gaming account.

[0046] 104. Determine the cheating status of the game account based on the information entropy corresponding to the Markov chain state transition space.

[0047] As you can understand, after obtaining information entropy, we can select game accounts with low entropy, determine the cheating status of the game account, and complete the cheat detection. Once a game account containing cheats is detected, we can subsequently take punitive measures against the game account, such as banning or demotion, to maintain game fairness and improve the gaming experience for normal players.

[0048] The above-mentioned embodiment of the present application applies the properties of Markov chains to evaluate a player's historical gaming behavior. By constructing a Markov chain state transition space coordinate system to observe the transition state of the gaming behavior, the information entropy is determined using the state transition coordinate sequence, and then the information entropy is used to determine whether the player's gaming behavior is disordered and random. If the transition of the player's historical gaming behavior is not random and has a specific and obvious transition pattern, then it means that the game account is in a cheating state. Since historical gaming behavior cannot be tampered with and detection of historical gaming behavior cannot be prevented, the cheat detection method based on historical gaming behavior can accurately analyze whether the player is in a cheating state. Moreover, through the transition of historical gaming behavior, it can be determined whether the player's gaming behavior transition is regular, thereby simply and efficiently determining whether the player has a game cheat. In this way, subsequent processing can be performed based on the detection results to maintain the fairness of the game and improve the gaming experience.

[0049] In combination with the above embodiment, the process of obtaining the historical game behavior sequence and the process of converting the historical game behavior sequence into the state transition coordinate sequence are described in detail below. Figure 2 This is a flow chart of a method for obtaining a game account's corresponding state game behavior sequence according to an embodiment of the present application. The specific method is as follows:

[0050] 201. Obtain game logs of a game account within a historical time period.

[0051] The game log is used to map the time at which a game action occurred to the game action identifier corresponding to that time. Typically, a game log stores game actions and the corresponding game time. All game actions generated by a game account participating in a game are recorded in the game log. Therefore, a historical game action sequence can be determined based on the game log of a game account over a historical time period. This historical game action sequence can be viewed as a Markov chain. By analyzing the Markov properties of this historical game action sequence, it can be determined whether the game actions generated by the game account follow a fixed pattern.

[0052] 202. Establish a historical game behavior sequence based on the time when the game behavior occurs.

[0053] After obtaining the game log, it is necessary to create a historical game behavior sequence based on the time when the game behavior occurred. In the historical game behavior sequence, each game behavior can be referred to by a game behavior identifier.

[0054] Specifically, the game behavior identifiers in the game log can be sorted by the time the game behavior occurred to obtain a historical game behavior sequence. For example, if the game log records game behavior A at 10:30, game behavior B at 10:32, game behavior A again at 10:34, and game behavior D at 10:35, the resulting historical game behavior sequence would be {A, B, A, D}.

[0055] 203. Count the frequency of each game behavior identifier appearing in the historical game behavior sequence.

[0056] After obtaining the historical game behavior sequence, the frequency of each game behavior identifier appearing in the historical game behavior sequence is first counted. The frequency is the number of times the game event corresponding to the game identifier appears in the historical time period.

[0057] 204. Sort all game behavior identifiers according to frequency, and obtain the order value corresponding to each game behavior identifier.

[0058] After obtaining the frequency of each game behavior identifier in the historical game behavior sequence, the game behavior identifiers can be sorted according to the frequency, and the order value corresponding to each game behavior identifier can be obtained respectively. This step is to convert the historical game behavior sequence into a digital sequence, that is, to digitize the information. In this way, digital coordinates can be used to represent the transfer between each game behavior, and the information entropy can be determined based on the distribution of digital coordinates in the coordinate axis. Among them, there are many ways to sort, such as positive order, reverse order, etc. In order to make the distribution of game behaviors with a large number of occurrences in the spatial coordinates concentrated in the center of the coordinate axis, the game behaviors may fall in the center of the coordinate axis. The embodiment of the present application provides an interpolation sorting method with a normal distribution, specifically:

[0059] First, sort all game behavior identifiers in descending order by frequency. Then, insert the highest-frequency game behavior identifier into the array as the first element. Then, in descending order, insert the remaining behavior identifiers into the array in a chronological order. For example, the second-highest-frequency game behavior identifier is inserted to the left of the first element, the second-highest-frequency game behavior identifier is inserted to the right of the first element, the fourth-highest-frequency game behavior identifier is inserted to the left of the second-highest-frequency game behavior identifier, and so on, until all game behavior identifiers are sorted. Finally, obtain the order value of each sorted game behavior identifier. This order value represents the coordinate value of the behavior in the Markov chain state transition space coordinate system.

[0060] For example, there are a total of 5 game behaviors in the historical game behavior sequence, and the corresponding behavior identifiers are: A, B, C, D, and E, and their corresponding frequencies are 1, 5, 10, 20, and 15, respectively. Therefore, the result after descending sorting is DECBA. Next, first arrange the behavior identifier D with the largest frequency as the first element to obtain D; then insert the game behavior E with the second largest frequency before D to obtain ED; then insert the behavior C with the third largest frequency after D to obtain EDC; repeat the insertion steps to insert the behavior identifier B with the fourth largest frequency before E and the behavior identifier A with the fifth largest frequency after C to obtain BEDCA. According to the arranged array BEDCA, it can be determined that the order values ​​corresponding to each behavior identifier A, B, C, D, and E are 5, 1, 4, 3, and 2, respectively.

[0061] 205. Replace each game behavior identifier in the historical game behavior sequence with the sequence value corresponding to each game behavior identifier to obtain a digital sequence.

[0062] After obtaining the sequence value, each game action identifier in the historical game action sequence is replaced with the sequence value corresponding to each game action identifier to obtain a numerical sequence. For example, in the historical action sequence {A, B, C, D, E}, the game action identifiers A, B, C, D, and E have corresponding sequence values ​​of 5, 1, 4, 3, and 2, respectively, resulting in the numerical sequence {5, 1, 4, 3, 2}.

[0063] This step is used to establish the correspondence between each game identifier and coordinate value in the historical game behavior sequence. If the determination method is other, this step will be replaced.

[0064] 206. Establish the Markov chain state transition space coordinate system.

[0065] Next, we need to establish a Markov chain state transition space coordinate system. This system is used to measure the transitions between various game behaviors. For example, we first determine a preset dimension, which indicates the number of state transitions between game behaviors. Then, we construct the Markov chain state transition space coordinate system based on this preset dimension.

[0066] The following is a brief introduction to the Markov chain state transition:

[0067] Traditional Markov chains are generally first-order Markov chains. However, based on the Markov property, combining multiple Markov chains as components results in an n-order Markov chain, which possesses n-order memory and can be considered a generalization of the Markov chain. For example, consider the weather. Considering the next day's weather based on today's weather is a model for a first-order Markov chain. However, considering the weather for the third day, starting from today's weather, one must first consider the transition from today's weather to the next, and then consider the transition from the second to the third. This process involves considering two weather transitions, given the known conditions of the day, and is therefore a model for a second-order Markov chain. Similarly, the n-order memory of an n-order Markov chain can be viewed as considering the transition of n events based on the currently known conditions.

[0068] This step incorporates Markov chain principles into cheat detection, analyzing the transitions between gaming behaviors. By analyzing the transitions between gaming behaviors, we can determine whether there are patterns in the transitions. If so, it indicates that the gaming behaviors follow a certain, relatively orderly pattern, indicating that the gaming behaviors are being performed using a cheat. If not, the gaming behaviors are being performed by a normal player.

[0069] In order to achieve the desired effect while minimizing the computational complexity, in this application, it is preferred to consider between 0 and 2 behavior transitions. Specifically, if behavior transitions are not considered and only behavior types are considered, the historical game behavior sequence is considered a 0th-order Markov chain. If 1 game behavior transition is considered, the historical game behavior sequence should be converted to a 1st-order Markov chain. If 2 game behavior transitions are considered, the historical game behavior sequence should be converted to a 2nd-order Markov chain.

[0070] To simplify the process of converting historical game behavior sequences into a Markov chain and facilitate analysis of each behavior transition, this step establishes a Markov chain state transition space coordinate system of corresponding dimensions based on the number of game behaviors involved in each behavior transition in the Markov chain. This involves using the coordinate values ​​of each axis of the space coordinate system to locate the location of a behavior transition in the Markov chain state transition space coordinate system, thereby obtaining the corresponding state transition coordinate sequence for the historical game behavior sequence.

[0071] 207. Determine the coordinate dimension of the coordinate sequence sample according to the dimension of the Markov chain state transition space coordinate system.

[0072] The dimension of the Markov chain state transition space coordinate system is consistent with the coordinate dimension of the coordinate sequence samples. That is, the dimension of the Markov chain state transition space coordinate system is the same as the coordinate dimension of the coordinate sequence samples.

[0073] 208. Determine a state transition coordinate sequence based on the dimension and digital sequence of the coordinate sequence samples.

[0074] The following is a detailed description of the process of converting the historical game behavior sequence into the transfer coordinate sequence in the Markov chain state transition space coordinate system:

[0075] For example, a historical game behavior sequence is {A, B, C, D, B, A, E, B}, and the corresponding coordinate value sequence is {3, 1, 5, 2, 1, 3, 4, 1}.

[0076] When the transfer of historical game behaviors is not considered, the corresponding Markov chain is of order 0, that is:

[0077] T={(A), (B), (C), (D), (B), (A), (E), (B)}

[0078] The corresponding Markov chain state transition space coordinate system is a one-dimensional space coordinate system, and its coordinate sequence is t = {(3), (1), (5), (2), (1), (3), (4), (1)}. In this case, the Markov chain state transition space coordinate system is a one-dimensional number axis, and each coordinate sequence sample is distributed in the form of a point on this one-dimensional number axis.

[0079] For example, when considering a single transition of historical game behavior, the corresponding Markov chain is of order 1. In this case, the Markov chain state transition space coordinate system is a two-dimensional coordinate axis. In the above example, the state transition sequence corresponding to the historical game behavior sequence {A, B, C, D, B, A, E, B} is {(A, B), (B, C), (C, D), (D, B), (B, A), (A, E), (E, B)}. In this order 1 expression, each set of behavior identifiers indicates that the game behavior has undergone a single behavior transition, such as (A, B) indicating the transition process from game behavior A to B. At this time, the state transition coordinate sequence is t = {(3, 1), (1, 5), (5, 2), (2, 1), (1, 3), (3, 4), (4, 1)}. At this time, the sample points of each coordinate sequence are distributed in a two-dimensional space coordinate system. The values ​​of the two coordinate axes represent the game behavior before the transfer and the game behavior after the transfer, respectively. For example, the coordinate value (3, 1) corresponding to (A, B) is 3, which is the coordinate value of the game behavior before the transfer, and 1 is the coordinate value of the game behavior after the transfer.

[0080] Exemplarily, when considering two transfers of historical game behavior, the corresponding Markov chain is of order 2. At this time, the Markov chain state transfer space coordinate system is a three-dimensional space coordinate axis. In the above example, the state transfer sequence corresponding to the historical game behavior sequence {A, B, C, D, B, A, E, B} is {(A, B, C), (B, C, D), (C, D, B), (D, B, A), (B, A, E), (A, E, B)}; in this second-order expression, each group of behavior identifiers indicates that the game behavior has undergone two behavior transfers, such as (A, B, C), which means that the game behavior is transferred from A to B and then to C. At this time, the state transfer coordinate sequence is t = {(3, 1, 5), (1, 5, 2), (5, 2, 1), (2, 1, 3), (1, 3, 4), (3, 4, 1)}; at this time, each coordinate sequence sample is distributed in the three-dimensional space coordinate system in the form of a point, and the values ​​of the three coordinate axes in the coordinate system represent the initial game behavior coordinates, the game behavior coordinates after the first behavior transfer, and the game behavior coordinates after the second transfer, respectively. For example, the coordinate value (3, 1, 5) corresponding to (A, B, C) is 3 for the game behavior coordinate value before the transfer, 1 for the game behavior coordinate value after the first behavior transfer, and 5 for the game behavior coordinate value after the second transfer.

[0081] In the above three cases, when obtaining the corresponding Markov chain expression based on the historical game behavior sequence, each game behavior that has occurred is considered as a known condition. Different consideration methods can also be set during the implementation process. Different dimensions can be selected according to different needs during the specific implementation process, and no specific limitations are imposed here.

[0082] After summarizing the above three cases, we can obtain the general expression of the state transition coordinate sequence as follows:

[0083]

[0084] Among them, M is the dimension of the Markov chain state transition space coordinate system, which actually means that the game behavior has been transferred M times, and N represents the Nth group of game behavior transfer process. This is the first group game behavior transfer process, in which Indicates the coordinate position on the first coordinate axis corresponding to the game behavior before the transfer; Indicates the coordinate position on the second coordinate axis corresponding to the game behavior after one transfer; The coordinates of the game behavior after transferring N-1 times correspond to the coordinate position on the M-th coordinate axis.

[0085] The state transition coordinate sequence is the digitized result of the historical game behavior sequence. Once the state transition coordinate sequence is obtained, the state transition after each game behavior can be analyzed based on the information entropy corresponding to the state transition coordinate sequence, thereby determining whether the player's historical game behavior is regular and detecting cheating players.

[0086] The following is a detailed introduction to the calculation process of information entropy and the process of determining the plug-in status based on information entropy. Figure 3 This is a flow chart of a method for calculating information entropy provided in an embodiment of the present application. Figure 3 As shown, the method comprises the following steps:

[0087] 301. Determine the frequency of each coordinate sequence sample appearing in the state transition coordinate sequence.

[0088] When calculating information entropy, we first need to determine the frequency of occurrence of each coordinate sequence sample, and then determine the information entropy based on the frequencies corresponding to all coordinate sequence samples. To calculate the frequency, we first need to count the frequency of each coordinate sequence sample appearing in the state transition coordinate sequence, that is, count the number of times each coordinate sequence appears.

[0089] 302. Determine the total frequency according to the frequencies corresponding to all coordinate sequence samples.

[0090] After obtaining the frequencies corresponding to all coordinate sequence samples, all frequencies can be added together to obtain the total frequency value. Of course, the total frequency of all coordinate sequence samples can also be directly counted.

[0091] 303. Determine a frequency value corresponding to each coordinate sequence sample according to the frequency and total frequency of each coordinate sequence sample appearing in the state transition coordinate sequence.

[0092] After obtaining the frequency of each coordinate sequence sample appearing in the state transition coordinate sequence, the frequency value corresponding to each coordinate sequence sample can be obtained. That is, it can be calculated using the probability calculation formula, which is:

[0093] P(x i )=C i / ∑ i∈I C i

[0094] Among them, i represents different coordinate sequence samples, corresponding to When i=1, the coordinate sequence sample is Among them, C i is the frequency corresponding to i, ∑ i∈I C i is the total frequency of occurrence of all coordinate sequence samples. Dividing the two, we can get the frequency value P(xi ).

[0095] 304. Determine the information entropy corresponding to the Markov chain state transition space based on the frequency values ​​corresponding to all coordinate sequence samples.

[0096] After obtaining the frequency value of each coordinate sequence according to the previous step, the corresponding information entropy value can be calculated based on it. The calculation method uses the following formula:

[0097]

[0098] Among them, H(x) is the information entropy value; P(x i ) is the probability of occurrence of each coordinate sequence sample, thus completing the information entropy value corresponding to the historical game behavior sequence.

[0099] Furthermore, when calculating information entropy, the above method centrally analyzes all coordinate sequence samples within the entire coordinate space. This means that the Markov chain state transition space is not partitioned, and the cheating status of a game account is determined based on the distribution of all coordinate sequence samples within the entire Markov chain state transition space.

[0100] Specifically, once the information entropy corresponding to the Markov chain state transition space is obtained, the cheating status of the game account can be determined accordingly. For example, a method of setting a preset threshold can be used to determine whether the game account corresponding to the information entropy is a cheating account.

[0101] The preset threshold can be a value obtained after extensive analysis of game data, used to determine whether a game account is in a cheating state. If the information entropy value corresponding to a game account is less than the preset threshold, it indicates that the player engages in a small number of gaming behaviors during gameplay, with a large number of repeated and similar gaming behavior combinations, and the transitions between gaming behaviors show a clear pattern. In this case, the player is a cheating player. Conversely, if the information entropy value corresponding to a game account is greater than or equal to the preset threshold, it indicates that the player typically engages in a wide variety of gaming behaviors during gameplay, and the transitions between gaming behaviors are relatively random and do not follow a fixed pattern. In this case, the player is a normal gamer.

[0102] For example, when there are too many coordinate sequence samples, the entire Markov chain state transition space can be divided into multiple subspaces, and then multiple sub-information entropies are determined based on the distribution of coordinate sequence samples in each subspace. Finally, based on the multiple sub-information entropies, it is determined whether the game account is a cheating account. The specific method is as follows: Figure 4 The embodiment shown.

[0103] This embodiment includes the following steps:

[0104] 401. Divide the Markov chain state transition space coordinate system according to preset parameters to obtain N state transition subspaces.

[0105] First, preset parameters can be determined according to the detection requirements, and then the Markov chain state transition space coordinate system can be divided according to the preset parameters to obtain N state transition subspaces, where N is a positive integer greater than or equal to 1. For example, when dividing the Markov chain state transition space coordinate system, since the coordinate values ​​corresponding to each game behavior identifier in the historical game behavior sequence are already obtained, the coordinate range of each coordinate axis can be determined first, and then the entire Markov chain state transition space can be divided based on the preset parameters and the coordinate range of the coordinate axis to obtain N sub-coordinate spaces.

[0106] For example, if the preset parameter is n, it means that the coordinate axis within each range is divided into n segments based on the coordinate range of each coordinate axis. Then, the entire Markov chain state transition space will be divided into n segments. M sub-coordinate spaces, M is the dimension of the Markov chain state transition space coordinate system. For example, in a one-dimensional Markov chain state transition space coordinate system, each coordinate axis is divided into n coordinate intervals, and the corresponding sub-coordinate spaces are n; in a two-dimensional Markov chain state transition space coordinate system, each coordinate axis is divided into n coordinate intervals, and the corresponding sub-coordinate spaces are n 2 In the three-dimensional Markov chain state transition space coordinate system, each coordinate axis is divided into n coordinate intervals, and the corresponding sub-coordinate space is n 3 Of course, this step also supports other subspace division methods, which can be set according to needs during specific implementation.

[0107] 402. Determine a target coordinate sequence sample in each of the N state transition subspaces according to the coordinate values ​​of the coordinate sequence sample.

[0108] After dividing the Markov chain state transition space coordinate system into N state transition subspaces, the coordinate range of each subspace is obtained. Then, based on the coordinate values ​​of the coordinate sequence samples, the target coordinate sequence samples in each of the N state transition subspaces can be determined.

[0109] 403. Determine the sub-information entropy corresponding to each state transition subspace according to the distribution of target coordinate sequence samples in each state transition subspace.

[0110] Then, it is necessary to determine the sub-information entropy corresponding to each state transition subspace based on the distribution of target coordinate sequence samples in each state transition subspace. The specific implementation method can be similar to steps 301 to 304 to calculate the respective sub-information entropy, specifically:

[0111] First, based on the target coordinate sequence samples in each state transition subspace, the state transition subspace coordinate sequence corresponding to each state transition subspace is determined. The frequency of each target coordinate sequence sample in the state transition subspace coordinate sequence is then determined. Finally, the total frequency corresponding to each state transition subspace is determined based on the frequencies corresponding to all target coordinate sequence samples. Finally, the target frequency value corresponding to each target coordinate sequence sample is determined based on the frequency of each target coordinate sequence sample in the state transition subspace coordinate sequence and the frequency corresponding to each state transition subspace.

[0112] After determining the target frequency value corresponding to each target coordinate sequence sample, the frequency calculation formula can be used to determine the sub-information entropy corresponding to the Markov chain state transition subspace based on the target frequency values ​​corresponding to all target coordinate sequence samples. Since there are N Markov chain state transition subspaces, the corresponding sub-information entropy is also N.

[0113] After obtaining N sub-information entropies, the cheat status needs to be determined based on the N sub-information entropies. For example, the mean sub-information entropy can be determined based on the N sub-information entropies, and then compared with a preset threshold. When the mean sub-information entropy is less than the preset threshold, the game account can be determined to be a cheating game account.

[0114] In addition, to ensure a more comprehensive analysis of all sub-information entropies, a neural network model can be used to determine the cheating status. The input to the neural network model can be a combined feature obtained from N sub-information entropies. After the combined feature is input into the cheating detection model (neural network model), the cheating detection model will output the cheating status of the game account. The cheating detection model can determine the randomness of the state transition of the game behavior based on the N sub-information entropies. The obtained N sub-information entropies can be analyzed centrally to determine the degree of match between the game account and the cheating account, thereby determining its cheating status.

[0115] During implementation, the combined features can be combined with the captured external tag information and used in a supervised learning model, or they can be used in an unsupervised learning model without using the external tag information. The specific implementation method can vary based on actual needs and conditions and is not specifically limited here.

[0116] Based on the above method embodiment, Figure 5 This is a structural diagram of a game cheat detection device provided by the present invention, as shown in FIG. Figure 5 As shown, the game display device includes:

[0117] The acquisition unit 501 is used to acquire a historical game behavior sequence corresponding to a game account. The historical game behavior sequence is used to indicate the time sequence of all game behaviors performed by the game account within a historical time period.

[0118] The determining unit 502 is configured to determine a state transition coordinate sequence corresponding to a historical game behavior sequence based on a Markov chain state transition space coordinate system. The state transition coordinate sequence includes at least one coordinate sequence sample.

[0119] The calculation unit 503 is configured to determine the information entropy corresponding to the Markov chain state transition space according to the distribution of at least one coordinate sequence sample in the Markov chain state transition space coordinate system.

[0120] The determination unit 502 is further configured to determine the cheating status of the game account based on the information entropy corresponding to the Markov chain state transition space.

[0121] In an optional embodiment, the acquisition unit 501 is specifically configured to acquire game logs for a game account within a historical time period. The game logs include the time at which the game activity occurred and the game activity identifier corresponding to the time at which the game activity occurred. A historical game activity sequence is established based on the time at which the game activity occurred, and the historical game activity sequence includes the game activity identifier.

[0122] In an optional embodiment, the game cheat detection device further includes a processing unit 504 .

[0123] Processing unit 504 is configured to count the frequency of each gaming behavior identifier appearing in the historical gaming behavior sequence, sort all gaming behavior identifiers by frequency, and obtain a sequence value corresponding to each gaming behavior identifier. Furthermore, each gaming behavior identifier in the historical gaming behavior sequence is replaced with the sequence value corresponding to each gaming behavior identifier to obtain a digital sequence.

[0124] In an optional embodiment, the determination unit 502 is specifically used to establish a Markov chain state transition space coordinate system, determine the coordinate dimension of the coordinate sequence sample according to the dimension of the Markov chain state transition space coordinate system, and determine the state transition coordinate sequence according to the dimension and digital sequence of the coordinate sequence sample.

[0125] In an optional embodiment, determination unit 502 is specifically configured to construct a Markov chain state transition space coordinate system based on a preset dimension, where the preset dimension is used to indicate the number of state transitions of the gaming behavior. A coordinate dimension of the coordinate sequence sample is determined based on the preset dimension, where the coordinate dimension is the same as the dimension of the Markov chain state transition space coordinate system.

[0126] In an optional embodiment, calculation unit 503 is configured to determine the frequency of occurrence of each coordinate sequence sample in the state transition coordinate sequence. A total frequency is determined based on the frequencies corresponding to all coordinate sequence samples. A frequency value corresponding to each coordinate sequence sample is determined based on the frequency of occurrence of each coordinate sequence sample in the state transition coordinate sequence and the total frequency. Thus, information entropy corresponding to the Markov chain state transition space is determined based on the frequency values ​​corresponding to all coordinate sequence samples.

[0127] In an optional embodiment, the determining unit 502 is specifically configured to determine that the game account is a cheat game account when the information entropy corresponding to the Markov chain state transition space is less than a preset threshold.

[0128] In an optional embodiment, the calculation unit 503 is specifically configured to divide the Markov chain state transition space coordinate system according to preset parameters to obtain N state transition subspaces, where N is a positive integer greater than or equal to 1. A target coordinate sequence sample in each of the N state transition subspaces is determined based on the coordinate values ​​of the coordinate sequence samples. A sub-information entropy corresponding to each state transition subspace is determined based on the distribution of the target coordinate sequence samples in each state transition subspace.

[0129] In an optional embodiment, the calculation unit 503 is specifically configured to determine the state transition subspace coordinate sequence corresponding to each state transition subspace based on the target coordinate sequence samples in each state transition subspace. The frequency corresponding to each state transition subspace is determined based on the frequencies corresponding to all target coordinate sequence samples. The target frequency value corresponding to each target coordinate sequence sample is determined based on the frequency of each target coordinate sequence sample appearing in the state transition subspace coordinate sequence and the frequency corresponding to each state transition subspace. The sub-information entropy corresponding to the Markov chain state transition subspace is determined based on the target frequency values ​​corresponding to all target coordinate sequence samples.

[0130] In an optional embodiment, the determining unit 502 is specifically configured to determine a mean sub-information entropy value based on the N sub-information entropies, and determine that the game account is a cheating game account when the mean sub-information entropy value is less than a preset threshold.

[0131] In an optional embodiment, the determination unit 502 is specifically configured to combine the N sub-information entropies to obtain a combined feature. The combined feature corresponding to the N sub-information entropies is input into a cheat detection model. The cheat detection model determines the cheating status of the game account based on the results of the cheat detection model. The cheat detection model is configured to determine the randomness of the state transition of the gaming behavior based on the N sub-information entropies.

[0132] The above-mentioned embodiment of the present application applies the properties of Markov chains to evaluate a player's historical gaming behavior. By constructing a Markov chain state transition space coordinate system to observe the transition state of the gaming behavior, the information entropy is determined using the state transition coordinate sequence, and then the information entropy is used to determine whether the player's gaming behavior is disordered and random. If the transition of the player's historical gaming behavior is not random and has a specific and obvious transition pattern, then it means that the game account is in a cheating state. Since historical gaming behavior cannot be tampered with and detection of historical gaming behavior cannot be prevented, the cheat detection method based on historical gaming behavior can accurately analyze whether the player is in a cheating state. Moreover, through the transition of historical gaming behavior, it can be determined whether the player's gaming behavior transition is regular, thereby simply and efficiently determining whether the player has a game cheat. In this way, subsequent processing can be performed based on the detection results to maintain the fairness of the game and improve the gaming experience.

[0133] It should be noted that the information interaction, execution process, etc. between the modules / units in the processing device are the same as those in this application. Figures 1 to 4 The corresponding method embodiments are based on the same concept. For specific contents, please refer to the description in the method embodiments shown above in this application, which will not be repeated here.

[0134] Next, we will introduce an electronic device provided by an embodiment of the present application. Figure 6 , Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 5 The game cheat detection device described in the corresponding embodiment is used to implement Figures 1 to 4 Specifically, the electronic device 800 includes: a receiver 801, a transmitter 802, a processor 803 and a memory 804 (wherein the number of the processor 803 in the execution device 800 can be one or more, Figure 6 (taking one processor as an example), the processor 803 may include an application processor 8031 ​​and a communication processor 8032. In some embodiments of the present application, the receiver 801, the transmitter 802, the processor 803 and the memory 804 may be connected via a bus or other means.

[0135] The memory 804 may include a read-only memory and a random access memory, and provides instructions and data to the processor 803. A portion of the memory 804 may also include non-volatile random access memory (NVRAM). The memory 804 stores processor and operation instructions, executable modules, or data structures, or subsets or extended sets thereof. The operation instructions may include various operation instructions for implementing various operations.

[0136] Processor 803 controls the operation of the execution device. In specific applications, the various components of the execution device are coupled together via a bus system. In addition to a data bus, the bus system may also include a power bus, a control bus, and a status signal bus. However, for clarity, all bus systems are referred to as a bus system in the figure.

[0137] The methods disclosed in the above embodiments of the present application can be applied to or implemented by the processor 803. The processor 803 can be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by hardware integrated logic circuits or software instructions in the processor 803. The above processor 803 can be a general-purpose processor, a digital signal processing (DSP), a microprocessor, or a microcontroller, and can further include an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The processor 803 can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of the present application can be directly implemented and executed by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium well-known in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in memory 804, and processor 803 reads information in memory 804 and, in conjunction with its hardware, completes the steps of the above method.

[0138] Receiver 801 can be used to receive input digital or character information and generate signal input related to executing device-related settings and function control. Transmitter 802 can be used to output digital or character information through the first interface. Transmitter 802 can also be used to send instructions to the disk pack through the first interface to modify data in the disk pack. Transmitter 802 can also include a display device such as a display screen.

[0139] In the embodiment of the present application, the application processor 8031 ​​in the processor 803 is used to execute Figures 1 to 4 The detection method of game cheats in the corresponding embodiment. It should be noted that the specific way in which the application processor 8031 ​​executes each step is the same as that in the present application. Figures 1 to 4The corresponding method embodiments are based on the same concept, and the technical effects they bring are the same as those in this application. Figures 1 to 4 The corresponding method embodiments are the same. For specific contents, please refer to the description in the method embodiments shown above in this application, which will not be repeated here.

[0140] An embodiment of the present application provides a computer-readable storage medium, which includes computer instructions. When the computer instructions are executed by a processor, they are used to implement the technical solution of any game plug-in detection method in the embodiments of the present application.

[0141] The present application also provides a computer program product which, when executed on a computer, enables the computer to execute the aforementioned Figures 1 to 4 The illustrated embodiment describes the steps in a method for detecting game cheats.

[0142] Although the present application is disclosed as above with the preferred embodiments, it is not intended to limit the present application. Any person skilled in the art may make possible changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be based on the scope defined by the claims of the present application.

Claims

1. A method for detecting game cheats, characterized in that: The detection method comprises: Obtaining a historical game behavior sequence corresponding to a game account; the historical game behavior sequence is used to indicate the time sequence of game behaviors performed by the game account within a historical time period; Determine a state transition coordinate sequence corresponding to the historical game behavior sequence according to a Markov chain state transition space coordinate system; the state transition coordinate sequence includes at least one coordinate sequence sample; Determining the information entropy corresponding to the Markov chain state transition space according to the distribution of the at least one coordinate sequence sample in the Markov chain state transition space coordinate system; The cheating status of the game account is determined according to the information entropy corresponding to the Markov chain state transition space.

2. The method according to claim 1, characterized in that The step of obtaining a historical game behavior sequence corresponding to a game account includes: Obtaining a game log of the game account within the historical time period; the game log including the time when the game behavior occurred and the game behavior identifier corresponding to the time when the game behavior occurred; The historical game behavior sequence is established according to the game log; the historical game behavior sequence includes the game behavior identifier.

3. The method according to claim 2, characterized in that The method further comprises: Counting the frequency of each game behavior identifier appearing in the historical game behavior sequence; Sort all the game behavior identifiers according to the frequency, and obtain the order value corresponding to each game behavior identifier; Each game behavior identifier in the historical game behavior sequence is replaced with a sequence value corresponding to each game behavior identifier to obtain a digital sequence.

4. The method according to claim 3, characterized in that Determining the state transition coordinate sequence corresponding to the historical game behavior sequence based on the Markov chain state transition space coordinate system includes: Establishing the Markov chain state transition space coordinate system; Determining the coordinate dimension of the coordinate sequence sample according to the dimension of the Markov chain state transition space coordinate system; The state transition coordinate sequence is determined according to the dimension of the coordinate sequence samples and the digital sequence.

5. The method according to claim 4, characterized in that The step of establishing the Markov chain state transition space coordinate system includes: Constructing the Markov chain state transition space coordinate system according to a preset dimension; the preset dimension is used to indicate the number of state transitions of the game behavior; Determining the dimension of the coordinate sequence samples according to the dimension of the Markov chain state transition space coordinate system includes: The coordinate dimension of the coordinate sequence sample is determined according to the preset dimension; the coordinate dimension is the same as the dimension of the Markov chain state transition space coordinate system.

6. The method according to any one of claims 1 to 5, characterized in that Determining the information entropy corresponding to the Markov chain state transition space according to the distribution of the at least one coordinate sequence sample in the Markov chain state transition space coordinate system includes: Determine the frequency of each coordinate sequence sample appearing in the state transition coordinate sequence; Determine the total frequency based on the frequencies corresponding to all coordinate sequence samples; Determining a frequency value corresponding to each coordinate sequence sample according to the frequency at which each coordinate sequence sample appears in the state transition coordinate sequence and the total frequency; The information entropy corresponding to the Markov chain state transition space is determined according to the frequency values ​​corresponding to all coordinate sequence samples.

7. The method according to claim 6, characterized in that The determining the cheating status of the game account according to the information entropy corresponding to the Markov chain state transition space includes: When the information entropy corresponding to the Markov chain state transition space is less than a preset threshold, it is determined that the game account is a cheat game account.

8. The method according to any one of claims 1 to 5, characterized in that Determining the information entropy corresponding to the Markov chain state transition space according to the distribution of the at least one coordinate sequence sample in the Markov chain state transition space coordinate system includes: Dividing the Markov chain state transition space coordinate system according to preset parameters to obtain N state transition subspaces; wherein N is a positive integer greater than or equal to 1; Determining a target coordinate sequence sample in each state transition subspace of the N state transition subspaces according to the coordinate values ​​of the coordinate sequence sample; According to the distribution of the target coordinate sequence samples in each state transition subspace, the sub-information entropy corresponding to each state transition subspace is determined to obtain N sub-information entropies.

9. The method according to claim 8, characterized in that The determining, according to the distribution of the target coordinate sequence samples in each state transition subspace, the sub-information entropy corresponding to each state transition subspace includes: Determining a state transition subspace coordinate sequence corresponding to each state transition subspace according to target coordinate sequence samples in each state transition subspace; Determine the frequency of each target coordinate sequence sample appearing in the state transition subspace coordinate sequence; Determine the frequency corresponding to each state transition subspace according to the frequencies corresponding to all target coordinate sequence samples; Determining a target frequency value corresponding to each target coordinate sequence sample according to the frequency of each target coordinate sequence sample appearing in the state transition subspace coordinate sequence and the frequency corresponding to each state transition subspace; The sub-information entropy corresponding to the Markov chain state transition subspace is determined according to the target frequency values ​​corresponding to all target coordinate sequence samples.

10. The method according to claim 8, characterized in that The determining the cheating status of the game account according to the information entropy corresponding to the Markov chain state transition space includes: Determine a mean sub-information entropy value based on the N sub-information entropies; When the mean value of the sub-information entropy is less than a preset threshold, it is determined that the game account is a cheat game account.

11. The method according to claim 9, characterized in that The determining the cheating status of the game account according to the information entropy corresponding to the Markov chain state transition space includes: Determine a mean sub-information entropy value based on the N sub-information entropies; When the mean value of the sub-information entropy is less than a preset threshold, it is determined that the game account is a cheat game account.

12. The method according to claim 8, characterized in that The determining the cheating status of the game account according to the information entropy corresponding to the Markov chain state transition space includes: Combining the N sub-information entropies to obtain a combined feature; The combined feature is input into a cheat detection model, and the cheat status of the game account is determined according to the result of the cheat detection model; the cheat detection model is used to determine the randomness of the state transition of the game behavior according to the N sub-information entropies.

13. The method according to claim 9, characterized in that The determining the cheating status of the game account according to the information entropy corresponding to the Markov chain state transition space includes: Combining the N sub-information entropies to obtain a combined feature; The combined feature is input into a cheat detection model, and the cheat status of the game account is determined according to the result of the cheat detection model; the cheat detection model is used to determine the randomness of the state transition of the game behavior according to the N sub-information entropies.

14. A game cheat detection device, characterized in that: include: An acquisition unit, configured to acquire a historical game behavior sequence corresponding to a game account; the historical game behavior sequence is used to indicate a time sequence of game behaviors performed by the game account within a historical time period; a determining unit, configured to determine a state transition coordinate sequence corresponding to the historical game behavior sequence according to a Markov chain state transition space coordinate system; the state transition coordinate sequence includes at least one coordinate sequence sample; a calculation unit, configured to determine the information entropy corresponding to the Markov chain state transition space according to the distribution of the at least one coordinate sequence sample in the Markov chain state transition space coordinate system; The determining unit is further configured to determine the cheating status of the game account based on the information entropy corresponding to the Markov chain state transition space.

15. An electronic device, characterized in that: include: A memory and a processor, wherein the memory and the processor are coupled; The memory is used to store one or more computer instructions; The processor is used to execute the one or more computer instructions to implement the game cheat detection method as described in any one of claims 1-5, 7 or 9-13.

16. A computer storage medium having one or more computer instructions stored thereon, characterized in that: The instruction is executed by the processor to implement the game cheat detection method as described in any one of claims 1-5, 7 or 9-13.

Images