Cheating detection method and device based on H5 game

Abstract

The invention provides a cheating detection method and device based on an H5 game, and the method comprises the steps: determining to-be-detected data of a game user in a game match in response to a game behavior of the game user in the H5 game, inputting the to-be-detected data into a pre-trained cheating detection model, and carrying out the cheating detection through the cheating detection model. Determining abnormal data in the to-be-detected data, outputting the abnormal data, determining an abnormal coefficient corresponding to the abnormal data according to the abnormal data, and judging whether the game user has a cheating behavior or not according to a preset detection threshold value and the abnormal coefficient. Through the method, after the game data of the game user in the game match is obtained, whether the game user has cheating behaviors in the game match or not can be efficiently judged through the pre-trained cheating check model.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and more particularly to a cheating detection method and device based on H5 games. Background Art

[0002] At present, with the rapid development of Internet technology, various types of games are widely popular. Among them, H5 games are online games developed based on HTML5 technology. They can be run directly in modern web browsers that support HTML5 standards without installing any plug-ins or additional software. H5 games usually have a smaller file size and can run seamlessly on mobile devices and desktop platforms, so they have quickly become popular in recent years.

[0003] Furthermore, H5 games are generally run based on JavaScript scripts. Although H5 games enhance their interactivity and dynamic effects through JavaScript scripts, the readability and modifiability of JavaScript also make the game vulnerable to cheating. Some game users can easily cheat by modifying game data or using automated scripts to change the difficulty of the game. These cheating behaviors not only undermine the fairness of the game, but also affect the gaming experience of other game users.

[0004] Therefore, how to identify whether game users in H5 games are cheating has become an urgent problem to be solved. Summary of the invention

[0005] The embodiments of this specification provide a cheating detection method and device based on H5 games to partially solve the problems existing in the above-mentioned prior art.

[0006] The embodiments of this specification adopt the following technical solutions:

[0007] This specification provides a cheating detection method based on H5 games, the method comprising:

[0008] In response to a game user's game behavior in the H5 game, determine the data to be detected of the game user in the game match;

[0009] Inputting the data to be detected into a pre-trained cheating detection model, so as to determine abnormal data in the data to be detected through the cheating detection model, and outputting the abnormal data;

[0010] Determine, according to the abnormal data, an abnormal coefficient corresponding to the abnormal data;

[0011] According to the preset detection threshold and the abnormal coefficient, it is determined whether the game user has cheated.

[0012] Optionally, determining the data to be detected of the game user in the game game specifically includes:

[0013] Obtaining a game log of the H5 game, so as to determine the initial game data of the game user in the game match according to the game log;

[0014] The initial game data is standardized to determine the data to be detected.

[0015] Optionally, the data to be detected includes at least one of click frequency, game duration, input mode or game score.

[0016] Optionally, after determining the data to be detected of the game user in the game match, before inputting the data to be detected into a pre-trained cheating detection model, the method further includes:

[0017] According to a pre-established cheating example database, the data to be detected is detected to determine the matching degree between the data to be detected and each game cheating example in the cheating example database;

[0018] If the matching degree is greater than a preset threshold, the game match corresponding to the data to be detected is marked as cheating;

[0019] Otherwise, the data to be detected continues to be input into the pre-trained cheating detection model.

[0020] Optionally, determining an abnormality coefficient corresponding to the abnormal data according to the abnormal data specifically includes:

[0021] Obtaining game data corresponding to historical game matches without cheating as first historical data;

[0022] Determine benchmark data corresponding to each type of game data in the first historical data;

[0023] An abnormal coefficient corresponding to the abnormal data is determined according to benchmark data of a type corresponding to the abnormal data.

[0024] Optionally, the cheating detection model is trained in advance, specifically including:

[0025] Obtaining game data corresponding to historical game matches in which cheating behavior occurs as second historical data;

[0026] determining cheating data in the second historical data as a mark;

[0027] Inputting the second historical data as a sample into the cheating detection model, so as to determine predicted abnormal data in the second historical data through the cheating detection model, and output the predicted abnormal data;

[0028] The cheating detection model is trained according to the difference between the predicted abnormal data and the cheating data.

[0029] Optionally, training the cheating detection model according to the difference between the predicted abnormal data and the cheating data specifically includes:

[0030] Determine a loss value according to the predicted abnormal data and the annotation using a preset loss function;

[0031] Determining a parameter increment according to the loss value;

[0032] According to the parameter increment, the parameters of the cheating detection model are adjusted.

[0033] This specification provides a cheating detection device based on H5 games, the device comprising:

[0034] A response module, used to determine the data to be detected of the game user in the game match in response to the game behavior of the game user in the H5 game;

[0035] An input module, used for inputting the data to be detected into a pre-trained cheating detection model, so as to determine abnormal data in the data to be detected through the cheating detection model, and output the abnormal data;

[0036] A determination module, used to determine an abnormal coefficient corresponding to the abnormal data according to the abnormal data;

[0037] The judgment module is used to judge whether the game user has cheating behavior according to a preset detection threshold and the abnormality coefficient.

[0038] This specification provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned cheating detection method based on H5 games is implemented.

[0039] This specification provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned cheating detection method based on H5 games when executing the program.

[0040] At least one of the above technical solutions adopted in the embodiments of this specification can achieve the following beneficial effects:

[0041] The embodiment of this specification discloses a cheating detection method based on H5 games. In response to the game behavior of a game user in an H5 game, the data to be detected of the game user in the game match is determined, and the data to be detected is input into a pre-trained cheating detection model, so as to determine the abnormal data in the data to be detected through the cheating detection model, and output it, and determine the abnormal coefficient corresponding to the abnormal data according to the abnormal data, and judge whether the game user has cheating behavior according to the preset detection threshold and the abnormal coefficient. Through this method, after obtaining the game data of the game user in the game match, it is possible to efficiently judge whether the game user has cheating behavior in the game match through the pre-trained cheating detection model. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The drawings described herein are used to provide a further understanding of this specification and constitute a part of this specification. The illustrative embodiments and descriptions of this specification are used to explain this specification and do not constitute an improper limitation on this specification. In the drawings:

[0043] Figure 1 A flowchart of a cheating detection method based on H5 games provided in an embodiment of this specification;

[0044] Figure 2 A schematic diagram of a cheating detection device based on H5 games provided in an embodiment of this specification;

[0045] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this specification. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solutions and advantages of this specification more clear, the technical solutions of this specification will be clearly and completely described below in combination with the specific embodiments of this specification and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of this specification, not all of them. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this specification.

[0047] The technical solutions provided by the embodiments of this specification are described in detail below in conjunction with the accompanying drawings.

[0048] Figure 1 The following is a flow chart of a cheating detection method based on H5 games in this specification, which may specifically include the following steps:

[0049] S100: In response to a game user's game behavior in an H5 game, determine data to be detected of the game user in a game match.

[0050] Specifically, in the embodiments of this specification, game users play H5 games through devices such as computers or mobile phones. H5 games are based on HTML5 technology and JavaScript (JS) scripts. Due to the readability and modifiability of JS scripts, game users can easily tamper with data, use automated scripts, or exploit game loopholes to cheat. These cheating behaviors will not only undermine the fairness of the game, but also affect the gaming experience of other game users. Therefore, in response to the game behavior of game users in H5 games, relevant game performance data and interaction data generated by the game users' operations in the game can be obtained as data to be detected.

[0051] Specifically, the user's interaction data can be determined in response to various operations of the game user, including but not limited to the position of the touch screen, sliding trajectory, click frequency, key input, etc. At the same time, in response to the game user's operation, the game server will also provide the game user's score, ranking, game time, skill release frequency and other game performance data in real time. Based on this, the above-mentioned interaction data and game performance data can be used as data to be detected as a basis for subsequent judgment of whether the game user cheats. It should be noted that the specific content of the data to be detected is not limited by the embodiments of this specification.

[0052] Furthermore, in response to the game behavior of the game user in the H5 game, the terminal device that determines the data to be detected of the game user in the game match may be a laptop, a tablet computer, etc. Of course, it may also be a server connected to the H5 game. For the sake of ease of description, this specification only takes the server as an example of the execution subject to illustrate the cheating detection method based on the H5 game provided in this specification.

[0053] S102: Inputting the data to be detected into a pre-trained cheating detection model, so as to determine abnormal data in the data to be detected through the cheating detection model, and output the abnormal data.

[0054] Furthermore, after determining the data to be detected, the data to be detected may be input into a cheating detection model, so that the cheating detection model identifies whether there is abnormal data caused by cheating in the data to be detected, and outputs an identification result, i.e., the abnormal data in the data to be detected.

[0055] For example, in an H5 game, players need to control characters to move around the map and defeat enemies. The data to be tested obtained from game users in the game includes 30 minutes of game time, 500 shots, 490 hits, especially 480 consecutive hits with almost no errors, an average movement speed of 8 units per second, and 10 skill releases with 100% hits. When these data to be tested are input into the pre-trained cheating detection model, the model recognizes that the shooting hit rate and skill hit rate are much higher than the normal level, and the movement speed may also exceed the normal setting range of the game, and the number of hits, movement speed, and skill hit rate are judged as abnormal data.

[0056] It should be noted that in actual applications, cheating detection models will comprehensively consider multiple data points, as well as their correlation and time series, to more accurately determine whether there is cheating behavior. For example, if an abnormal shooting accuracy occurs at the same time as an abnormally high movement speed, and there is no reasonable in-game explanation, such as the use of cheating software or automated scripts, then the model will mark these data as abnormal data.

[0057] Among them, the cheating detection model is constructed based on machine learning or deep learning algorithms, and can be trained on a large amount of labeled data consisting of data corresponding to normal game behaviors and cheating behaviors, and learn and identify the characteristics of game cheating behaviors, including but not limited to abnormal operation modes, illogical game performance, and data patterns similar to historical cheating behaviors. It should be noted that the cheating detection model can be a cheating detection model obtained by training based on a preset neural network structure, or it can be a model obtained by training based on a pre-trained model, such as GPT-4, etc. The specific method of pre-training the cheating detection model is not limited in this specification embodiment.

[0058] Furthermore, after the data to be detected is input into the cheating detection model, the cheating detection model will evaluate the input data to be detected through the cheating behavior characteristics learned in the training phase, and determine whether there is abnormal data in the data to be detected based on the model parameters. In addition, in addition to being able to output abnormal data in the data to be detected, the cheating detection model can also determine the type of cheating behavior corresponding to the abnormal data, such as using automated scripts, data tampering, or exploiting game loopholes, etc. The specific content of the output results of the cheating detection model is not limited in the embodiments of this specification.

[0059] It should be noted that the abnormal data may be multiple data in the data to be detected, or it may be one data. The number of abnormal data is determined based on the data to be detected. If the cheating detection model does not output abnormal data, it means that the game game corresponding to the data to be detected does not contain cheating behavior.

[0060] S104: Determine an abnormal coefficient corresponding to the abnormal data according to the abnormal data.

[0061] Furthermore, after the cheating detection model outputs abnormal data, in order to ensure the accuracy of the detection, the game with abnormal data is not directly determined as a game with cheating behavior, but the abnormal coefficient corresponding to the abnormal data is determined based on the abnormal data. The abnormal coefficient is a quantitative indicator calculated based on the characteristics of the abnormal data, and different abnormal data corresponds to different abnormal coefficients.

[0062] Specifically, if there are multiple abnormal data, for each abnormal data, the abnormal coefficient corresponding to the abnormal data can be determined respectively, which is used for subsequent comparison with the detection threshold of the type to which the abnormal coefficient belongs. It is also possible to determine the abnormal coefficient corresponding to each abnormal data according to each abnormal coefficient and each coefficient weight after determining the coefficient weight corresponding to each abnormal coefficient in advance. It should be noted that the specific method of determining the abnormal coefficient corresponding to the abnormal data according to the abnormal data is not limited in the embodiments of this specification.

[0063] Furthermore, the greater the degree to which abnormal data deviates from the normal data range, the higher its abnormal coefficient is usually. For example, if a player's shooting hit rate is much higher than the average level of other players in the game and remains stable, then the abnormal coefficient of this abnormal data will be high. In addition, the frequency of abnormal data in the game is also an important influencing factor. If an abnormal data only appears once in a while, then its abnormal coefficient may be relatively low; but if the abnormal data appears frequently, then its abnormal coefficient will increase accordingly. At the same time, the time when the abnormal data appears may also affect the calculation of the abnormal coefficient. For example, the high movement speed that appears at the beginning of the game may just be the player getting familiar with the game, while the high movement speed that continues in the later stages of the game is more likely to be cheating.

[0064] S106: Determine whether the game user has cheated based on a preset detection threshold and the abnormality coefficient.

[0065] Further, after determining the abnormal coefficient corresponding to each abnormal data, the determined abnormal coefficient will be compared with the preset detection threshold to make a final judgment on whether the game user has cheated. The detection threshold is set in advance based on factors such as the characteristics of the game, historical cheating cases, and the performance of the cheating detection model. When the abnormal coefficient exceeds this value, it is considered that the user is highly suspected of cheating. The specific method of determining the detection threshold is not limited in the embodiments of this specification.

[0066] Furthermore, according to the preset detection threshold and the abnormal coefficient, it is determined whether the game user has cheated. For each abnormal coefficient, the detection threshold corresponding to the abnormal coefficient can be determined, and it can be determined whether the abnormal coefficient is lower than the detection threshold. If so, the task indicates that the game user has not cheated in the game match corresponding to the data to be detected. If not, that is, the abnormal coefficient is higher than the preset detection threshold, it means that the game user has cheated in the game match corresponding to the data to be detected. For example, if the abnormal coefficient corresponding to a single abnormal data is 1, and the preset detection threshold is 0.8, the game corresponding to the abnormal data can be regarded as a cheating game.

[0067] Of course, it is also possible to determine the global abnormality coefficient corresponding to each abnormality coefficient based on the preset weight corresponding to each abnormality coefficient, and then, based on the global abnormality coefficient and the detection threshold, determine whether the game user has cheated in the game corresponding to the data to be detected. The specific comparison method of the detection threshold and the abnormality coefficient is not limited in the embodiments of this specification.

[0068] based on Figure 1 A cheating detection method based on H5 games is provided. After obtaining the game data of game users in the game, the pre-trained cheating detection model can be used to efficiently determine whether the game users have cheated in the game.

[0069] Furthermore, determining the data to be detected of the game user in the game game can specifically include obtaining the game log of the H5 game, determining the initial game data of the game user in the game game according to the game log, and standardizing the initial game data to determine the data to be detected.

[0070] In other words, in response to the game behavior of the game user in the H5 game, the data to be detected of the game user in the game match can be determined, and the game log of the H5 game can be obtained first. Among them, the game log is a file that records various events and data when the game is running. It contains a variety of information such as the player's behavior in the game, changes in the game state, and system events. For H5 games, the game log may be stored on the server side, or it may be partially or completely generated on the client side and uploaded to the server. Obtaining these game logs is the basis for analyzing player behavior and detecting cheating behavior.

[0071] Then, based on the obtained game logs, the various operation behavior data of the game user during the game can be determined, such as movement trajectory, number of shots, number of hits, skill release, prop usage, etc., as well as game match information, such as match ID, match time, match type, map information, etc., and game performance data, such as wins and losses, scores, rankings, etc. These data are the initial game data of the game user in the game. In addition, the single-game game data of a single player can be organized into an independent information set, which is convenient for subsequent feature extraction and cheating detection by game.

[0072] Furthermore, since the data in the game log may come from different game versions, different servers or different time periods, their formats and units may differ. In order to uniformly process and analyze these data, the initial game data can be standardized. Standardization includes, but is not limited to, data cleaning: removing duplicate, missing, erroneous or irrelevant data. Data format unification: converting data in different formats into a unified format, such as converting timestamps into a readable date and time format. Data unit unification: converting data in different units into a unified unit, such as converting the speed unit from "pixels / second" to "units / second". Data normalization: converting data of different dimensions to the same dimension to facilitate subsequent data analysis and comparison.

[0073] It should be noted that the specific method of standardizing the initial game data is not limited in the embodiments of this specification. The initial game data after standardization becomes the data to be tested for subsequent analysis, which will be input into the cheating detection model for calculating abnormal coefficients, judging cheating behavior, etc.

[0074] The data to be detected includes at least one of click frequency, game duration, input mode or game score.

[0075] Furthermore, after determining the data to be detected of the game user in the game game, before inputting the data to be detected into the pre-trained cheating detection model, the data to be detected can also be detected according to a pre-established cheating example database to determine the matching degree between the data to be detected and each game cheating example in the cheating example database; if the matching degree is greater than a preset threshold, the game game corresponding to the data to be detected is marked as cheating; otherwise, the data to be detected continues to be input into the pre-trained cheating detection model.

[0076] Specifically, before inputting the data to be detected into the pre-trained cheating detection model, the data to be detected can also be preliminarily detected according to the pre-established cheating example database. Specifically, the data to be detected can be compared with the game data in the cheating example database, and the similarity or matching degree between the game data of each type and the corresponding data of each game cheating example can be determined according to the game data of each type in the data to be detected. The matching degree is calculated based on multiple factors, such as the similarity of operation modes, the similarity of score anomalies, the consistency of timestamps, etc.

[0077] If the match is greater than this threshold, the game match corresponding to the data to be detected contains cheating behavior, and it will be directly marked as cheating. This can quickly identify obvious cheating behaviors, reduce the burden of subsequent complex model processing, and speed up the processing of cheating behaviors. If the match between the data to be detected and the cheating example database is lower than the preset threshold, the system will consider that the data requires further analysis. At this point, the data to be detected will continue to be input into the pre-trained cheating detection model, which is usually based on more complex algorithms and more data features to evaluate cheating suspicions, and can identify more hidden or complex cheating behaviors.

[0078] Among them, the cheating example database is a database that stores the characteristics of known cheating behaviors. These characteristics come from the analysis of historical cheating cases, the experience summary of game developers or reports from the player community. Each cheating example records in detail the characteristics, patterns and related game data of the cheating behavior.

[0079] Furthermore, based on the abnormal data, an abnormal coefficient corresponding to the abnormal data is determined. Specifically, game data corresponding to historical game matches in which no cheating behavior occurs is obtained as first historical data, benchmark data corresponding to each type of game data in the first historical data is determined, and the abnormal coefficient corresponding to the abnormal data is determined based on the benchmark data of the type corresponding to the abnormal data.

[0080] First, a large amount of game data corresponding to historical game matches without cheating can be collected. This data is called the first historical data, which covers a wide range of game scenarios, player groups and game time periods corresponding to H5 games to ensure its representativeness and reliability. It is the basis for subsequent analysis and is used to establish benchmarks for various types of game data.

[0081] Next, the first historical data is analyzed in depth to determine the benchmark data of various types of game data, such as click frequency, operation time, score change, etc. Benchmark data refers to the normal range or distribution that these game data should present in the absence of cheating behavior. The central trend and dispersion of the data can be described by statistical methods, such as calculating the mean, standard deviation, median, quartiles, etc.

[0082] Furthermore, after determining the benchmarks for each type of game data, the new game data, i.e., the degree of deviation of the abnormal data, can be evaluated based on these benchmarks. For each abnormal data, the benchmark data range corresponding to the type of the abnormal data is determined, and the abnormal coefficient can be calculated by comparing the abnormal data with the corresponding benchmark data.

[0083] Furthermore, a Z-score can be used as an abnormality coefficient, which represents the standard deviation multiple between the abnormal data and the average value of the benchmark data. If the Z-score of the abnormal data is very high, it means that it is far away from the center of the benchmark data and is therefore more likely to be abnormal. Of course, other methods that can quantify the degree of deviation of the abnormal data relative to the benchmark data can also be used, such as using a percentage deviation method, and determining the specific method of the abnormality coefficient corresponding to the abnormal data according to the benchmark data of the type corresponding to the abnormal data. This specification embodiment does not limit it.

[0084] Furthermore, the cheating detection model is trained in advance, which can be specifically as follows: game data corresponding to historical game matches in which cheating behavior exists is obtained as the second historical data, cheating data in the second historical data is determined as a label, and the second historical data is used as a sample to input into the cheating detection model, so as to determine the predicted abnormal data in the second historical data through the cheating detection model and output it, and the cheating detection model is trained according to the difference between the predicted abnormal data and the cheating data.

[0085] Specifically, to train the cheating detection model, first, in the data preparation stage, obtain historical game data with cheating behavior as the second historical data. This data should contain obvious cheating features, such as abnormally high scores, unnatural operation modes, and quickly completed tasks, providing actual samples of cheating behavior. Then, through manual or automated means, determine the cheating data in the second historical data as the target label for supervised learning.

[0086] Furthermore, the second historical data containing cheating data and normal data is input as a sample into the cheating detection model, so that the cheating detection model predicts the input second historical data to determine which data is abnormal data caused by cheating. Specifically, the cheating detection model evaluates the cheating suspicion of each sample based on its internal algorithm and learned features, and outputs predicted abnormal data.

[0087] In order to evaluate the prediction performance of the model, the predicted abnormal data output by the model is compared with the actual cheating data. By calculating a series of performance indicators (such as accuracy, recall rate, F1 score, etc.), the accuracy and effectiveness of the model in identifying cheating behavior can be quantitatively evaluated. According to the evaluation results, the model can be adjusted and optimized as necessary, including modifying the model structure, adjusting algorithm parameters, adding regularization terms, etc., to improve the generalization ability and recognition accuracy of the model.

[0088] The above process usually requires multiple iterations to achieve the desired training effect. In each iteration, the model is fine-tuned based on the previous evaluation results and its performance is re-evaluated. Through continuous iterative training, the recognition ability of the model can be gradually improved, making it more adaptable to complex game environments and changing cheating methods.

[0089] Furthermore, the cheating detection model is trained according to the difference between the predicted abnormal data and the cheating data. Specifically, a loss value is determined according to the predicted abnormal data and the annotation using a preset loss function, a parameter increment is determined according to the loss value, and the parameters of the cheating detection model are adjusted according to the parameter increment.

[0090] Specifically, a suitable loss function can be defined based on the difference between the predicted abnormal data and the labeled cheating data to quantify the degree of mismatch between the model prediction and the actual situation. Then, the gradient of the loss function with respect to the model parameters is calculated by the back propagation algorithm, and the direction in which the loss value decreases fastest in the parameter space is identified, that is, the direction of parameter optimization. Then, the increment of the cheating detection model parameters is determined based on the calculated gradient and the learning rate (a pre-set hyperparameter). The learning rate controls the step size of the parameter update and affects the stability and convergence speed of the training process. Finally, the parameters of the model are updated using the calculated parameter increments. This process is carried out through iterations, and each iteration aims to reduce the loss value so that the model's prediction is closer to the actual cheating data, that is, the cheating detection model is trained with the goal of minimizing the difference between the predicted abnormal data and the cheating data. Through such a training process, the cheating detection model can gradually learn and adapt to the characteristics of cheating behavior, and improve its recognition accuracy and generalization ability on unknown data.

[0091] The above is a cheating detection method based on H5 games provided in the embodiments of this specification. Based on the same idea, this specification also provides corresponding devices, storage media and electronic devices.

[0092] Figure 2 A schematic diagram of a cheating detection device based on H5 games provided in an embodiment of this specification, the device includes:

[0093] A response module 200 is used to determine the to-be-detected data of the game user in the game match in response to the game behavior of the game user in the H5 game;

[0094] An input module 202, used to input the data to be detected into a pre-trained cheating detection model, so as to determine abnormal data in the data to be detected through the cheating detection model, and output the abnormal data;

[0095] A determination module 204, configured to determine an abnormal coefficient corresponding to the abnormal data according to the abnormal data;

[0096] The judgment module 206 is used to judge whether the game user has cheating behavior according to a preset detection threshold and the abnormality coefficient.

[0097] Optionally, the response module 200 is specifically used to obtain the game log of the H5 game to determine the initial game data of the game user in the game match based on the game log; and standardize the initial game data to determine the data to be detected.

[0098] Optionally, the data to be detected includes at least one of click frequency, game duration, input mode or game score.

[0099] Optionally, the response module 200 is also used to detect the data to be detected according to a pre-established cheating example database, and determine the matching degree between the data to be detected and each game cheating example in the cheating example database; if the matching degree is greater than a preset threshold, the game game corresponding to the data to be detected is marked as cheating; otherwise, the data to be detected continues to be input into the pre-trained cheating detection model.

[0100] Optionally, the determination module 204 is specifically used to obtain game data corresponding to historical game matches without cheating as the first historical data; determine benchmark data corresponding to each type of game data in the first historical data; and determine the abnormal coefficient corresponding to the abnormal data based on the benchmark data of the type corresponding to the abnormal data.

[0101] Optionally, the input module 202 is also used to obtain game data corresponding to historical game matches in which cheating behavior exists as second historical data; determine cheating data in the second historical data as a label; input the second historical data as a sample into the cheating detection model to determine and output predicted abnormal data in the second historical data through the cheating detection model; and train the cheating detection model based on the difference between the predicted abnormal data and the cheating data.

[0102] Optionally, the input module 202 is further specifically configured to determine a loss value according to the predicted abnormal data and the annotation using a preset loss function; determine a parameter increment according to the loss value; and adjust parameters of the cheating detection model according to the parameter increment.

[0103] This specification also provides a computer-readable storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, it can be used to perform the above Figure 1 Provides a cheating detection method based on H5 games.

[0104] based on Figure 1 The cheating detection method based on H5 game shown in the specification also provides Figure 3 The structural diagram of the electronic device shown in FIG. Figure 3 At the hardware level, the electronic device includes a processor, an internal bus, a network interface, a memory, and a non-volatile memory, and may also include other hardware required for the business. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to achieve the above Figure 1 The cheating detection method based on H5 games.

[0105] Of course, in addition to software implementation, this specification does not exclude other implementation methods, such as logic devices or a combination of software and hardware, etc., that is to say, the executor of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0106] In the 1990s, improvements to a technology could be clearly distinguished as hardware improvements (for example, improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the method flow). However, with the development of technology, many improvements to the method flow today can be regarded as direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that an improvement in a method flow cannot be implemented using a hardware entity module. For example, a programmable logic device (PLD) (such as a field programmable gate array (FPGA)) is such an integrated circuit whose logical function is determined by the user's programming of the device. Designers can "integrate" a digital system on a PLD by programming it themselves, without having to ask a chip manufacturer to design and produce a dedicated integrated circuit chip. Moreover, nowadays, instead of manually making integrated circuit chips, this kind of programming is mostly implemented by "logic compiler" software, which is similar to the software compiler used when developing and writing programs, and the original code before compilation must also be written in a specific programming language, which is called hardware description language (HDL). There is not only one HDL, but many kinds, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc. The most commonly used ones are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also know that it is only necessary to program the method flow slightly in the above-mentioned hardware description languages ​​and program it into the integrated circuit, and then it is easy to obtain the hardware circuit that implements the logic method flow.

[0107] The controller can be implemented in any appropriate manner, for example, the controller can take the form of a microprocessor or processor and a computer-readable medium storing a computer-readable program code (such as software or firmware) that can be executed by the (micro)processor, a logic gate, a switch, an application-specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320. The memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art also know that in addition to implementing the controller in a purely computer-readable program code manner, the controller can be implemented in the form of a logic gate, a switch, an application-specific integrated circuit, a programmable logic controller, and an embedded microcontroller by logically programming the method steps. Therefore, this controller can be considered as a hardware component, and the devices included therein for implementing various functions can also be regarded as structures within the hardware component. Or even, the devices for implementing various functions can be regarded as both software modules for implementing the method and structures within the hardware component.

[0108] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0109] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0110] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems, or computer program products. Therefore, this specification may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, this specification may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0111] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0112] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0113] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0114] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0115] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0116] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0117] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0118] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems or computer program products. Therefore, this specification may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, this specification may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0119] This specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0120] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0121] The above description is only an embodiment of the present specification and is not intended to limit the present specification. For those skilled in the art, the present specification may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present specification shall be included in the scope of the claims of the present specification.

Claims

1. A cheating detection method based on H5 games, characterized in that: The method comprises: In response to a game user's game behavior in the H5 game, determine the data to be detected of the game user in the game match; Inputting the data to be detected into a pre-trained cheating detection model, so as to determine abnormal data in the data to be detected through the cheating detection model, and outputting the abnormal data; Determine, according to the abnormal data, an abnormal coefficient corresponding to the abnormal data; According to the preset detection threshold and the abnormal coefficient, it is determined whether the game user has cheated.

2. The method according to claim 1, characterized in that Determining the data to be detected of the game user in the game game, specifically including: Obtaining a game log of the H5 game, so as to determine the initial game data of the game user in the game match according to the game log; The initial game data is standardized to determine the data to be detected.

3. The method according to claim 1, characterized in that The data to be detected includes at least one of click frequency, game duration, input mode or game score.

4. The method according to claim 1, characterized in that After determining the data to be detected of the game user in the game match, before inputting the data to be detected into a pre-trained cheating detection model, the method further includes: According to a pre-established cheating example database, the data to be detected is detected to determine the matching degree between the data to be detected and each game cheating example in the cheating example database; If the matching degree is greater than a preset threshold, the game match corresponding to the data to be detected is marked as cheating; Otherwise, the data to be detected continues to be input into the pre-trained cheating detection model.

5. The method according to claim 1, characterized in that Determining an abnormal coefficient corresponding to the abnormal data according to the abnormal data specifically includes: Obtaining game data corresponding to historical game matches without cheating as first historical data; Determine benchmark data corresponding to each type of game data in the first historical data; An abnormal coefficient corresponding to the abnormal data is determined according to benchmark data of a type corresponding to the abnormal data.

6. The method according to claim 1, characterized in that The cheating detection model is trained in advance, specifically including: Obtaining game data corresponding to historical game matches in which cheating behavior occurs as second historical data; determining cheating data in the second historical data as a mark; Inputting the second historical data as a sample into the cheating detection model, so as to determine predicted abnormal data in the second historical data through the cheating detection model, and output the predicted abnormal data; The cheating detection model is trained according to the difference between the predicted abnormal data and the cheating data.

7. The method according to claim 6, characterized in that Training the cheating detection model according to the difference between the predicted abnormal data and the cheating data specifically includes: Determine a loss value according to the predicted abnormal data and the annotation using a preset loss function; Determining a parameter increment according to the loss value; According to the parameter increment, the parameters of the cheating detection model are adjusted.

8. A cheating detection device based on H5 games, characterized in that: The device comprises: A response module, used to determine the data to be detected of the game user in the game match in response to the game behavior of the game user in the H5 game; An input module, used for inputting the data to be detected into a pre-trained cheating detection model, so as to determine abnormal data in the data to be detected through the cheating detection model, and output the abnormal data; A determination module, used to determine an abnormal coefficient corresponding to the abnormal data according to the abnormal data; The judgment module is used to judge whether the game user has cheating behavior according to a preset detection threshold and the abnormality coefficient.

9. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method described in any one of claims 1 to 7 is implemented.

Images