CN113318455B - Game data processing method and device - Google Patents
Game data processing method and device Download PDFInfo
- Publication number
- CN113318455B CN113318455B CN202110565399.8A CN202110565399A CN113318455B CN 113318455 B CN113318455 B CN 113318455B CN 202110565399 A CN202110565399 A CN 202110565399A CN 113318455 B CN113318455 B CN 113318455B
- Authority
- CN
- China
- Prior art keywords
- game data
- plug
- parameter
- identified
- identification model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 12
- 238000011156 evaluation Methods 0.000 claims abstract description 79
- 238000000034 method Methods 0.000 claims abstract description 42
- 230000008569 process Effects 0.000 claims abstract description 16
- 238000012545 processing Methods 0.000 claims description 25
- 238000010586 diagram Methods 0.000 claims description 21
- 239000000203 mixture Substances 0.000 claims description 10
- 238000012216 screening Methods 0.000 claims description 10
- 238000012549 training Methods 0.000 claims description 9
- 238000010606 normalization Methods 0.000 claims description 5
- 238000005516 engineering process Methods 0.000 abstract description 4
- 244000035744 Hura crepitans Species 0.000 description 7
- 230000006870 function Effects 0.000 description 5
- 230000002776 aggregation Effects 0.000 description 4
- 238000004220 aggregation Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 230000001960 triggered effect Effects 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63F—CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
- A63F13/00—Video games, i.e. games using an electronically generated display having two or more dimensions
- A63F13/70—Game security or game management aspects
- A63F13/75—Enforcing rules, e.g. detecting foul play or generating lists of cheating players
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63F—CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
- A63F13/00—Video games, i.e. games using an electronically generated display having two or more dimensions
- A63F13/80—Special adaptations for executing a specific game genre or game mode
- A63F13/822—Strategy games; Role-playing games
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63F—CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
- A63F2300/00—Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game
- A63F2300/50—Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game characterized by details of game servers
- A63F2300/55—Details of game data or player data management
- A63F2300/5586—Details of game data or player data management for enforcing rights or rules, e.g. to prevent foul play
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63F—CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
- A63F2300/00—Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game
- A63F2300/60—Methods for processing data by generating or executing the game program
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63F—CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
- A63F2300/00—Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game
- A63F2300/80—Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game specially adapted for executing a specific type of game
- A63F2300/807—Role playing or strategy games
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Business, Economics & Management (AREA)
- Computer Security & Cryptography (AREA)
- General Business, Economics & Management (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a game data processing method and device. Wherein the method comprises the following steps: acquiring game data generated by an object to be identified in a game process; inputting game data into an external identification model, wherein the external identification model is obtained by performing unsupervised learning on sample game data; acquiring an evaluation parameter corresponding to an object to be identified, which is output by the plug-in identification model, and comparing the evaluation parameter with a central parameter corresponding to the plug-in identification model, wherein the central parameter is used for representing the central position of the evaluation parameter of the plug-in identification model, which is output by aiming at a plurality of game data; and under the condition that the deviation between the evaluation parameter corresponding to the object to be identified and the central parameter is larger than a preset value, determining that the object to be identified introduces third-party auxiliary software in the game. The invention solves the technical problem of inaccurate identification of the plug-in players in the related technology.
Description
Technical Field
The invention relates to the field of games, in particular to a method and a device for processing game data.
Background
Sandbox Games (Sandbox Games) are a type of game that has evolved from a sand table game, consisting of one or more map areas, often containing a variety of game elements, including role playing, action, shooting, driving, etc. The virtual roles played by players in a sandbox game can change or affect and even create the world, which is also a feature of a sandbox game.
In order to obtain higher achievement in the game, some third party auxiliary software is introduced to cheat in the game, namely, the cheat player is provided with the plug-in, and the action has a great influence on the fairness of the game, so that the game experience of a normal player is influenced, and therefore, the identification of the cheat player provided with the plug-in the game is very important to the fairness of the game.
At present, when identifying cheating players in a sandbox game, relevant adaptation is usually performed according to the function of hanging, for example, the speed of the players is detected by identifying acceleration hanging. However, the sandbox game has multiple modes, multiple behavior data are provided in each mode, and the new plug-in layer exists endlessly, so that the plug-in player is identified by adopting a single judgment standard in the prior art, and the accuracy is low.
Aiming at the problem of inaccurate identification of the plug-in players in the related technology, no effective solution is proposed at present.
Disclosure of Invention
The embodiment of the invention provides a processing method and a processing device of game data, which at least solve the technical problem of inaccuracy in identifying an external player in the related technology.
According to an aspect of an embodiment of the present invention, there is provided a game data processing method including: acquiring game data generated by an object to be identified in a game process; inputting game data into an external identification model, wherein the external identification model is obtained by performing unsupervised learning on sample game data; acquiring an evaluation parameter corresponding to an object to be identified, which is output by the plug-in identification model, and comparing the evaluation parameter with a central parameter corresponding to the plug-in identification model, wherein the central parameter is used for representing the central position of the evaluation parameter of the plug-in identification model, which is output by aiming at a plurality of game data; and under the condition that the deviation between the evaluation parameter corresponding to the object to be identified and the central parameter is larger than a preset value, determining that the object to be identified introduces third-party auxiliary software in the game.
Further, inputting game data to the plug-in recognition model, comprising: screening target game data belonging to a target type from game data of an object to be identified; and inputting the target game data into the plug-in identification model.
Further, the method further comprises the steps of: obtaining an externally hung identification model, which comprises the following steps: obtaining sample game data, wherein the sample game data comprises game data which does not contain labels in a preset time period, and the labels are used for identifying whether the game data are normal or not; and performing unsupervised training through the initial model and the sample game data to obtain the plug-in identification model.
Further, the initial model is created by a gaussian mixture algorithm.
Further, obtaining sample game data includes: obtaining game data which does not contain sample labels in a preset time period; screening target type game data from game data not containing sample labels; and carrying out normalization processing on the sample game data of the screened target type to obtain sample game data.
Further, the method further comprises the steps of: determining a target type, the steps comprising: obtaining game data without labels, wherein the game data without labels comprises a plurality of parameters, and each parameter in the game data without labels is subjected to average processing to obtain a plurality of first averages corresponding to the game data without labels, and the labels are used for identifying whether the game data is normal or not; acquiring normal game data, wherein the normal game data comprises a plurality of parameters, each parameter in the normal game data is subjected to mean value processing respectively to obtain a plurality of second mean values corresponding to the normal game data, and the normal game data is game data corresponding to an object without introducing third party auxiliary software; and comparing the first average value and the second average value corresponding to each type of parameter in the game data which does not contain the labels with the normal game data, and acquiring a parameter type with the difference between the first average value and the second average value being larger than a preset value and taking the parameter type as a target type.
Further, the method further comprises the steps of: generating an evaluation parameter scatter diagram based on evaluation parameters of a plurality of objects to be identified output by the plug-in identification model, wherein the evaluation parameter of each object to be identified corresponds to one scatter point in the evaluation parameter scatter diagram; acquiring a first position with most scattered points and a second position with most scattered points in the evaluation parameter scattered points; and determining the evaluation parameter corresponding to the first position as a central parameter, and determining the difference between the evaluation parameter corresponding to the second position and the central parameter as a preset value.
According to an aspect of an embodiment of the present invention, there is provided a game data processing apparatus including: the first acquisition module is used for acquiring game data generated by the object to be identified in the game process; the input module is used for inputting game data into the plug-in identification model, wherein the plug-in identification model is obtained by performing unsupervised learning on sample game data; the second acquisition module is used for acquiring the evaluation parameters corresponding to the object to be identified output by the plug-in identification model, and comparing the evaluation parameters with the central parameters corresponding to the plug-in identification model, wherein the central parameters are used for representing the central positions of the evaluation parameters output by the plug-in identification model aiming at a plurality of game data; the determining module is used for determining that the object to be identified introduces third-party auxiliary software in the game under the condition that the deviation between the evaluation parameter corresponding to the object to be identified and the central parameter is larger than a preset value.
According to an aspect of an embodiment of the present invention, there is provided a storage medium including a stored program, wherein the device on which the storage medium is controlled to execute the above-described game data processing method when the program runs.
According to an aspect of an embodiment of the present invention, there is provided a processor for running a program, wherein the program executes the above-described game data processing method.
In the embodiment of the invention, game data generated by an object to be identified in a game process is obtained; inputting game data into an external identification model, wherein the external identification model is obtained by performing unsupervised learning on sample game data; acquiring an evaluation parameter corresponding to an object to be identified, which is output by the plug-in identification model, and comparing the evaluation parameter with a central parameter corresponding to the plug-in identification model, wherein the central parameter is used for representing the central position of the evaluation parameter of the plug-in identification model, which is output by aiming at a plurality of game data; and under the condition that the deviation between the evaluation parameter corresponding to the object to be identified and the central parameter is larger than a preset value, determining that the object to be identified introduces third-party auxiliary software in the game. According to the scheme, the external hanging player is not identified from the external hanging principle of the external hanging software, but the external hanging identification model is created based on the game data, and the external hanging identification model is analyzed from the whole game data, so that the external hanging of unknown types can be effectively identified, the accuracy and the real-time performance of external hanging identification in the game are improved, and the technical problem that the external hanging player identification in the related art is inaccurate is solved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute a limitation on the application. In the drawings:
FIG. 1 is a flow chart of a method of processing game data according to an embodiment of the present invention;
FIG. 2 is a scatter diagram of game data output after prediction using the plug-in recognition model provided by the present application;
FIG. 3 is a schematic diagram of player numbers and scores obtained after predicting game data using the plug-in recognition model provided by the present application;
FIG. 4 is a scatter diagram of game data output after prediction in different modes using the plug-in recognition model provided by the present application;
fig. 5 is a schematic diagram of a processing apparatus for game data according to an embodiment of the present invention.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
According to an embodiment of the present invention, there is provided an embodiment of a method of processing game data, it being noted that the steps shown in the flowcharts of the drawings may be performed in a computer system such as a set of computer executable instructions, and that although a logical order is shown in the flowcharts, in some cases the steps shown or described may be performed in an order different from that herein.
Fig. 1 is a flowchart of a game data processing method according to an embodiment of the present invention, as shown in fig. 1, the method including the steps of:
Step S102, game data generated by the object to be identified in the game process is obtained.
Specifically, the object to be identified may be a virtual character of a player of the target game in the target game. The target game herein may be a sandbox game (e.g., my world, mini world, etc.), or other game that includes virtual characters. The game data may be behavior data of the virtual character in the game, for example: parameters such as the moving speed, the stay time, the attack frequency and the like of the virtual character.
The steps described above may be performed by an anti-cheating server of the target game. The anti-plug-in server acquires game data of all players in the game in real time so as to identify the players who open the plug-in the game process.
Step S104, inputting the game data into an external identification model, wherein the external identification model is obtained by performing unsupervised learning on the sample game data.
Specifically, the external hanging recognition model may be a neural network model, and the sample game data is used for training the external hanging recognition model. The sample game data is unlabeled data, so that the learning is not supervised by the plug-in recognition model.
In an alternative embodiment, the generation of the plug-in recognition model may be performed at a certain period. And each time of the generated plug-in recognition model, the plug-in player in the target game can be recognized from the current moment to the next moment of generating the plug-in recognition model.
In the scheme, the external recognition model is trained in an unsupervised learning mode, so that the neural network model for external recognition is obtained under the condition of lack of priori knowledge.
Step S106, obtaining an evaluation parameter corresponding to the object to be identified output by the plug-in identification model, and comparing the evaluation parameter with a central parameter corresponding to the plug-in identification model, wherein the central parameter is used for representing the central position of the evaluation parameter output by the plug-in identification model aiming at a plurality of game data
The external hanging recognition model performs unsupervised learning, and the unlabeled sample game data comprises normal game data and cheating game data of the external hanging. Because the characteristic information of the normal game data is relatively close, certain similarity can appear, so that the evaluation parameter of the normal game data has a center, namely, the center is used for reading books, and the closer to the center, the lower the possibility of being the externally hung game data. On the basis, the deviation degree of game data of the object to be identified and the center can be determined through the evaluation parameters, so that whether the object to be identified is hung or not is confirmed.
Step S108, under the condition that the deviation between the evaluation parameter corresponding to the object to be identified and the central parameter is larger than a preset value, the object to be identified is determined to introduce third party auxiliary software into the game.
Specifically, the object to be identified introduced with the third party auxiliary software is the object to be identified which is hung outside. The evaluation parameter may be a scoring of the input game data by the plug-in recognition model.
In an alternative embodiment, in the process of training the plug-in recognition model, the plug-in recognition model learns the score distribution of the normal game data in the sample game data, so that the center score corresponding to the normal game data can be determined. In the identification process, game data are input into the plug-in identification model to obtain the score of the game data output by the plug-in identification model, the score of the game data of the object to be identified is compared with the center score, if the difference between the score and the center score is larger than a preset value, it is determined that the object to be identified is introduced with third party auxiliary software in the game process, namely the object to be identified is determined to be hung.
The difference between the normal game data of the common player and the cheating game data of the external player is not very clear, and the difference in the number of the normal game data and the cheating game data of the external player easily makes the judgment of the Gaussian mixture algorithm error. In an alternative embodiment, the plug-in identification model is a gaussian mixture model created by a gaussian mixture algorithm, and a cost function (score for determining a deviation degree of a sample from a center) of game data obtained by the gaussian mixture model may be used as an evaluation parameter of the game data.
The object to be identified can be all current players, and through the scheme, whether each player in the game is hung on the outer side or not can be judged, so that the player hung on the outer side is identified from all players. After identifying the out-of-hook player, the out-of-hook player may be alerted in the game or the player may be directly removed from the current game.
It should be noted that, the anti-plug-in server for implementing the method steps in the embodiment may be disposed in the data center, and the game data is directly sent to the data center by the game server, which further reduces the delay of plug-in recognition compared with the case that the game log (log) needs to be transmitted into the data center for plug-in recognition in the related art.
Under the condition that a new plug-in layer exists endlessly, a single judgment standard is adopted to identify the plug-in players, the types of the plug-in players which can be identified and processed are limited, and the accuracy is low. The embodiment of the application acquires game data generated by the object to be identified in the game process; inputting game data into an external identification model, wherein the external identification model is obtained by performing unsupervised learning on sample game data; acquiring an evaluation parameter corresponding to an object to be identified, which is output by the plug-in identification model, and comparing the evaluation parameter with a central parameter corresponding to the plug-in identification model, wherein the central parameter is used for representing the central position of the evaluation parameter of the plug-in identification model, which is output by aiming at a plurality of game data; and under the condition that the deviation between the evaluation parameter corresponding to the object to be identified and the central parameter is larger than a preset value, determining that the object to be identified introduces third-party auxiliary software in the game. According to the scheme, the external hanging player is not identified from the external hanging principle of the external hanging software, but the external hanging identification model is created based on the game data, and the external hanging identification model is analyzed from the whole game data, so that the external hanging of unknown types can be effectively identified, the accuracy and the real-time performance of external hanging identification in the game are improved, the technical problem that the external hanging player is not accurately identified in the related art is solved, the external hanging player can be effectively identified, and the method has universality of cross playing methods.
As an alternative embodiment, inputting game data into the plug-in recognition model includes: screening target game data belonging to a target type from game data of an object to be identified; and inputting the target game data into the plug-in identification model.
Specifically, the target type is used for indicating the parameter types with certain influence when the plug-in is opened, and the game data of the parameter types have larger difference when the plug-in is opened and when the plug-in is not opened, so that the identification of the plug-in player can be performed based on the game data of the types, and useless game data can be removed.
In an alternative embodiment, game data of a player to be identified is obtained in real time, the game data is screened to obtain game data of a preset parameter type, and then the game data of a target type is input into the plug-in identification model for identification.
It should be noted that, game parameters affected by different plug-in software are different, so the target type may be changed along with different plug-in software, and in particular may be performed according to a certain period. The period may be the same as or different from the period of generating the plug-in recognition model,
As an alternative embodiment, the method further comprises: obtaining an externally hung identification model, which comprises the following steps: obtaining sample game data, wherein the sample game data comprises game data which does not contain labels in a preset time period, and the labels are used for identifying whether the game data are normal or not; and performing unsupervised training through the initial model and the sample game data to obtain the plug-in identification model.
Specifically, the game data not including the label is used for indicating that whether the plug-in is open or not is not labeled. In an alternative embodiment, the game data for the most recent preset period of time (28 days) may be obtained as sample game data while training the plug-in recognition model. The sample game data comprises normal game data and cheating game data of open plug-in.
In an alternative embodiment, as much game data for the player as possible may be collected, in units of a single field. For example, the number of times the virtual character is turned up in mid-air speed (double-jump), the time the virtual character stays at a certain specific height, the number of times the virtual character collides with a square, the number of times the square under the virtual character is air, the distance from the other party when the virtual character attacks, the dot product (dot) of the visual field direction and the relative coordinate vector when the virtual character attacks, the long-distance movement (which may include the time of long-distance movement and the ratio of the highest speed to the normal speed when the virtual character moves long-distance) of the virtual character beyond the normal movement speed, the number of abnormal game mode changes of the client, the long-time line-of-sight locking to other players, and the like may be affected by cheating.
As an alternative embodiment, the initial model is created by a gaussian mixture algorithm.
The Gaussian distribution, namely normal distribution, can estimate the distribution probability of the game data of the object to be identified according to the deviation degree between the evaluation of the game data of the object to be identified of the plug-in identification model and the center position, so as to judge whether the object to be identified is cheated.
It should be noted that, some types of game data cannot be accurately determined, for example, detecting the game data related to the flight may generate some false positives when the player logs in the card, but since there are macroscopically many players with the same game data, if a large amount of game data deviates by the same value in the gaussian distribution, the center of the determination may deviate accordingly. These cases do not need to be handled when acquiring sample game data.
It should be further noted that the game data of various parameter types are not isolated, for example, a plurality of relevant game data can be detected about the flying of a virtual character in the game, and once one of the relevant game data is triggered, other pieces are easily triggered together, so that the data is abnormal. Based on the characteristics of the Gaussian mixture algorithm, the association relation among the game data of different parameter types can be not considered, so that overall operation is carried out on the game data of all the parameter types.
According to the scheme, the initial model is built through the Gaussian mixture algorithm, so that the plug-in identification model obtained through training can have certain stability, the influence of external factors such as network blocking on identification can be avoided, the mutual influence between related game data can be avoided, and accurate determination during plug-in identification is further provided.
As an alternative embodiment, obtaining sample game data includes: obtaining game data which does not contain sample labels in a preset time period; screening sample game data of a target type from game data not containing sample labels; and respectively carrying out normalization processing on the screened game data of the target type to obtain sample game data.
Specifically, the normalization process is used for scaling the game data to a close dimension according to a certain rule, namely converting the game data into a specified interval, so that errors caused by the influence of some game parameters with overlarge numerical values in the operation process can be avoided.
As an alternative embodiment, the method further comprises: determining a target type, the steps comprising: obtaining game data without labels, wherein the game data without labels comprises a plurality of parameters, and each parameter in the game data without labels is subjected to average processing to obtain a plurality of first averages corresponding to the game data without labels, and the labels are used for identifying whether the game data is normal or not; acquiring normal game data, wherein the normal game data comprises a plurality of parameters, each parameter in the normal game data is subjected to mean value processing respectively to obtain a plurality of second mean values corresponding to the normal game data, and the normal game data is game data corresponding to an object without introducing third party auxiliary software; and comparing the first average value and the second average value corresponding to each type of parameter in the game data which does not contain the labels with the normal game data, and acquiring a parameter type with the difference between the first average value and the second average value being larger than a preset value and taking the parameter type as a target type.
The above steps are used to select the parameter types that can be used to determine whether the player is cheating, which are more susceptible to the effects of the plug-in software. The target type obtained by the scheme is the target type which is screened from game data of the object to be identified when the object to be identified is identified.
The Android system and the iOS system have great difference in the convenience degree of using the plug-ins, so most plug-ins are concentrated in the Android system, and based on this, in an alternative embodiment, the game data obtained from the Android system can be directly used as the game data which does not include labels, the game data obtained from the iOS system is used as the normal game data, and each parameter in the game data which does not include labels and the normal game data is respectively averaged to obtain a first average value and a second average value corresponding to the parameters. And if the difference between the first average values and the second average values is large, the parameters corresponding to the average values are influenced by the plug-in software, so that the parameter types are used as the target types.
According to the scheme, the sample game data of the target type is selected from the plurality of sample game data to carry out the plug-in recognition model, so that irrelevant parameters can be removed, the data processing amount is reduced, and the accuracy of the plug-in recognition model recognition is improved.
As an optional embodiment, generating an evaluation parameter scatter diagram based on evaluation parameters of a plurality of objects to be identified output by the plug-in identification model, wherein the evaluation parameter of each object to be identified corresponds to one scatter point in the evaluation parameter scatter diagram; acquiring a first position with most scattered points and a second position with most scattered points in the evaluation parameter scattered points; and determining the evaluation parameter corresponding to the first position as a central parameter, and determining the difference between the evaluation parameter corresponding to the second position and the central parameter as a preset value.
Specifically, the second position with the scattered points gathered many times can be used as a boundary between a normal player and an externally hung player.
The preset value for comparing with the evaluation parameter output by the plug-in identification model can be obtained in such a way. The cheating game data of the external hanging has certain commonality, so that the evaluation parameters of the cheating game data also have certain aggregation, and therefore, another aggregation center except the aggregation caused by the normal game data, namely, the second position with the scattered points aggregated a plurality of times can be selected as the boundary between the normal player and the external hanging player, so that the difference between the second position and the first position with the most scattered points aggregated forms the preset value. FIG. 2 is a scatter diagram of game data output after prediction by using the plug-in recognition model provided by the application, wherein the abscissa of the scatter diagram is used for representing the number of players, the ordinate is used for representing the evaluation parameters output by the plug-in recognition model, according to the scatter diagram, most of the scores of players are concentrated at a part above 0, and a relatively obvious aggregation layering exists in the score-48, and the cheating rate of the players with the score less than-48 is very high as determined by cross comparison from other channels. Thus, it can be basically determined that a score below-48 is a cheating player.
FIG. 3 is a schematic diagram of player numbers and scores obtained after predicting game data using the plug-in recognition model provided by the present application; referring to fig. 3, the horizontal axis represents scoring of game data by the plug-in recognition model, the vertical axis represents the total number of players below a certain score, and it can be seen that the number of players is mutated near-48, so that it is possible to verify that it is feasible to distinguish plug-in players with-48 as the standard.
Fig. 4 is a scatter diagram of game data output after prediction in different modes by using the plug-in recognition model provided by the application, and it can be seen that the scatter diagrams in different game modes have similar images, so that it can be determined that the scheme provided by the embodiment is suitable for various game modes.
And removing the game from the identified external player, and closing the game, so as to obtain a table-one result, wherein the table-one result is exemplified by different modes of the game A, A-1 indicates that the game A is in the 1 mode, and the like. The numbers 1, 2 and 4 represent the results of the function of removing the plug-in player after being started, and the numbers 3, 5 and 6 represent the results of the function of removing the plug-in player after not being started, so that compared with the results, the accuracy of the plug-in player identified by the scheme in the embodiment is high, and the cheating rate of the player can be effectively reduced.
List one
Sequence number | Game machine | Total number of samples | Predicting plug-in rate |
1 | A-1 | 2269161 | 1.781% |
2 | A-2 | 972540 | 0.764% |
3 | A-3 | 432830 | 7.303% |
4 | A-4 | 387209 | 2.040% |
5 | A-5 | 338204 | 4.409% |
6 | A-6 | 15238 | 10.736% |
According to an embodiment of the present invention, there is provided an embodiment of a game data processing method, and fig. 5 is a schematic diagram of a game data processing apparatus according to an embodiment of the present invention, as shown in fig. 5, the apparatus includes:
a first obtaining module 50, configured to obtain game data generated during a game by an object to be identified;
The input module 52 is configured to input game data into an external recognition model, where the external recognition model is obtained by performing unsupervised learning on sample game data;
The second obtaining module 54 is configured to obtain an evaluation parameter corresponding to the object to be identified output by the plug-in identification model, and compare the evaluation parameter with a central parameter corresponding to the plug-in identification model, where the central parameter is used to characterize a central position of the evaluation parameter output by the plug-in identification model for a plurality of game data;
The determining module 56 is configured to determine that the object to be identified introduces third party auxiliary software in the game if a deviation between the evaluation parameter corresponding to the object to be identified and the center parameter is greater than a preset value.
As an alternative embodiment, the input module includes: the screening sub-module is used for screening target game data belonging to a target type from game data of an object to be identified; and the input sub-module is used for inputting the target game data into the plug-in identification model.
As an alternative embodiment, the above device further comprises: the third acquisition module is used for acquiring the plug-in identification model, and the third acquisition module comprises: the second acquisition submodule is used for acquiring sample game data, wherein the sample game data comprises game data which does not contain labels in a preset time period, and the labels are used for identifying whether the game data are normal or not; and the training sub-module is used for performing unsupervised training through the initial model and the sample game data to obtain an externally hung identification model.
As an alternative embodiment, the initial model is created by a gaussian mixture algorithm.
As an alternative embodiment, the second acquisition sub-module includes: the first acquisition unit is used for acquiring game data which does not contain sample labels in a preset time period; the screening unit is used for screening out the game data of the target type from the game data which does not contain the sample mark; and the processing unit is used for carrying out normalization processing on the screened game data of the target type to obtain sample game data.
As an alternative embodiment, the above device further comprises: the determining module is used for obtaining game data without labels, wherein the game data without labels comprises a plurality of parameters, each parameter in the game data without labels is subjected to mean value processing respectively to obtain a plurality of first mean values corresponding to the game data without labels, and the labels are used for identifying whether the game data is normal or not; the third acquisition unit is used for acquiring normal game data, wherein the normal game data comprises a plurality of parameters, each parameter in the normal game data is subjected to mean value processing respectively to obtain a plurality of second mean values corresponding to the normal game data, and the normal game data is game data corresponding to an object without introducing third party auxiliary software; and the comparison unit is used for comparing the first average value and the second average value corresponding to each type of parameters in the game data which do not contain the labels with the normal game data, and acquiring the parameter type with the difference between the first average value and the second average value being larger than a preset value and taking the parameter type as the target type.
As an alternative embodiment, the above device further comprises: the generation module is used for generating an evaluation parameter scatter diagram based on the evaluation parameters of the plurality of objects to be identified output by the plug-in identification model, wherein the evaluation parameter of each object to be identified corresponds to one scatter point in the evaluation parameter scatter diagram; the third acquisition module is used for acquiring a first position with the most scattered points and a second position with the most scattered points in the evaluation parameter scattered points; the second determining module is used for determining the evaluation parameter corresponding to the first position as a central parameter and determining the difference between the evaluation parameter corresponding to the second position and the central parameter as a preset value.
According to an embodiment of the present invention, there is provided a storage medium including a stored program, wherein the program is controlled to execute the above-described game data processing method by a device in which the storage medium is located when the program runs.
According to an embodiment of the present invention, there is provided a processor for running a program, wherein the program executes the above-described game data processing method.
The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.
In the foregoing embodiments of the present invention, the descriptions of the embodiments are emphasized, and for a portion of this disclosure that is not described in detail in this embodiment, reference is made to the related descriptions of other embodiments.
In the several embodiments provided in the present application, it should be understood that the disclosed technology may be implemented in other manners. The above-described embodiments of the apparatus are merely exemplary, and the division of the units, for example, may be a logic function division, and may be implemented in another manner, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some interfaces, units or modules, or may be in electrical or other forms.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a usb disk, a read-only memory (ROM), a random access memory (RAM, random Access Memory), a removable hard disk, a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The foregoing is merely a preferred embodiment of the present invention and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present invention, which are intended to be comprehended within the scope of the present invention.
Claims (9)
1. A method of processing game data, comprising:
Acquiring game data generated by an object to be identified in a game process;
inputting the game data into an external identification model, wherein the external identification model is obtained by performing unsupervised learning on sample game data, the sample game data comprises game data which does not contain labels in a preset time period, and the labels are used for identifying whether the game data are normal or not;
Acquiring an evaluation parameter corresponding to the object to be identified, which is output by the plug-in identification model, and comparing the evaluation parameter with a central parameter corresponding to the plug-in identification model, wherein the central parameter is used for representing the central position of the evaluation parameter output by the plug-in identification model aiming at a plurality of game data;
Under the condition that the deviation between the evaluation parameter corresponding to the object to be identified and the central parameter is larger than a preset value, determining that the object to be identified introduces third-party auxiliary software in the game;
Wherein the method further comprises: generating an evaluation parameter scatter diagram based on evaluation parameters respectively corresponding to a plurality of objects to be identified output by the plug-in identification model, wherein the evaluation parameter of each object to be identified corresponds to one scatter point in the evaluation parameter scatter diagram; acquiring a first position with most scattered points and a second position with most scattered points in the evaluation parameter scattered points; and determining the evaluation parameter corresponding to the first position as the central parameter, and determining the difference between the evaluation parameter corresponding to the second position and the central parameter as the preset value.
2. The method of claim 1, wherein inputting the game data into a plug-in recognition model comprises:
screening target game data belonging to a target type from the game data of the object to be identified;
and inputting the target game data into the plug-in identification model.
3. The method according to claim 1, wherein the method further comprises: obtaining an externally hung identification model, which comprises the following steps:
acquiring the sample game data;
And performing unsupervised training through the initial model and the sample game data to obtain the plug-in identification model.
4. A method according to claim 3, wherein the initial model is created by a gaussian mixture algorithm.
5. A method according to claim 3, wherein obtaining the sample game data comprises:
Obtaining game data which does not contain sample labels in the preset time period;
Screening out the game data of the target type from the game data which does not contain the sample label;
and carrying out normalization processing on the screened game data of the target type to obtain the sample game data.
6. The method according to claim 2 or 5, characterized in that the method further comprises: determining the target type, which comprises:
Obtaining game data without labels, wherein the game data without labels comprises a plurality of parameters, each parameter in the game data without labels is subjected to mean value processing respectively to obtain a plurality of first mean values corresponding to the game data without labels, and the labels are used for identifying whether the game data is normal or not;
Obtaining normal game data, wherein the normal game data comprises a plurality of parameters, each parameter in the normal game data is subjected to average processing to obtain a plurality of second averages corresponding to the normal game data, and the normal game data is game data corresponding to objects without third party auxiliary software;
And comparing the first average value and the second average value corresponding to each type of parameter in the game data which does not contain the labels with the normal game data, and acquiring a parameter type with the difference between the first average value and the second average value being larger than a preset value and taking the parameter type as a target type.
7. A game data processing apparatus, comprising:
the first acquisition module is used for acquiring game data generated by the object to be identified in the game process;
The input module is used for inputting the game data into an external identification model, wherein the external identification model is obtained by performing unsupervised learning on sample game data, the sample game data comprises game data which does not contain labels in a preset time period, and the labels are used for identifying whether the game data are normal or not;
The second acquisition module is used for acquiring the evaluation parameters corresponding to the object to be identified, which are output by the plug-in identification model, and comparing the evaluation parameters with the central parameters corresponding to the plug-in identification model, wherein the central parameters are used for representing the central positions of the evaluation parameters output by the plug-in identification model aiming at a plurality of game data;
The determining module is used for determining that the object to be identified introduces third-party auxiliary software in the game under the condition that the deviation between the evaluation parameter corresponding to the object to be identified and the central parameter is larger than a preset value;
Wherein the device is further for: generating an evaluation parameter scatter diagram based on evaluation parameters respectively corresponding to a plurality of objects to be identified output by the plug-in identification model, wherein the evaluation parameter of each object to be identified corresponds to one scatter point in the evaluation parameter scatter diagram; acquiring a first position with most scattered points and a second position with most scattered points in the evaluation parameter scattered points; and determining the evaluation parameter corresponding to the first position as the central parameter, and determining the difference between the evaluation parameter corresponding to the second position and the central parameter as the preset value.
8. A storage medium comprising a stored program, wherein the program, when run, controls a device in which the storage medium is located to perform the method of processing game data according to any one of claims 1 to 6.
9. A processor for running a program, wherein the program runs to execute the game data processing method according to any one of claims 1 to 6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110565399.8A CN113318455B (en) | 2021-05-24 | 2021-05-24 | Game data processing method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110565399.8A CN113318455B (en) | 2021-05-24 | 2021-05-24 | Game data processing method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113318455A CN113318455A (en) | 2021-08-31 |
CN113318455B true CN113318455B (en) | 2024-04-19 |
Family
ID=77416471
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110565399.8A Active CN113318455B (en) | 2021-05-24 | 2021-05-24 | Game data processing method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113318455B (en) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107158707A (en) * | 2017-04-27 | 2017-09-15 | 浙江大学 | A kind of method for detecting abnormality and device played for MMORPGs |
CN107614075A (en) * | 2015-05-20 | 2018-01-19 | Cy游戏公司 | Information processing system, server and program and terminal and program |
CN108905209A (en) * | 2018-06-11 | 2018-11-30 | 网易(杭州)网络有限公司 | Plug-in detection method and system, electronic equipment and storage medium in game |
CN109464807A (en) * | 2018-11-06 | 2019-03-15 | 网易(杭州)网络有限公司 | Detect game plug-in method, apparatus and terminal |
KR20200039536A (en) * | 2018-10-05 | 2020-04-16 | 넷마블 주식회사 | Method and apparatus for detecting abnormal case |
KR20200133555A (en) * | 2019-05-20 | 2020-11-30 | 넷마블 주식회사 | Method for collecting data for machine learning |
CN112791414A (en) * | 2021-04-15 | 2021-05-14 | 腾讯科技(深圳)有限公司 | Plug-in recognition model training method and device, electronic equipment and storage medium |
-
2021
- 2021-05-24 CN CN202110565399.8A patent/CN113318455B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107614075A (en) * | 2015-05-20 | 2018-01-19 | Cy游戏公司 | Information processing system, server and program and terminal and program |
CN107158707A (en) * | 2017-04-27 | 2017-09-15 | 浙江大学 | A kind of method for detecting abnormality and device played for MMORPGs |
CN108905209A (en) * | 2018-06-11 | 2018-11-30 | 网易(杭州)网络有限公司 | Plug-in detection method and system, electronic equipment and storage medium in game |
KR20200039536A (en) * | 2018-10-05 | 2020-04-16 | 넷마블 주식회사 | Method and apparatus for detecting abnormal case |
CN109464807A (en) * | 2018-11-06 | 2019-03-15 | 网易(杭州)网络有限公司 | Detect game plug-in method, apparatus and terminal |
KR20200133555A (en) * | 2019-05-20 | 2020-11-30 | 넷마블 주식회사 | Method for collecting data for machine learning |
CN112791414A (en) * | 2021-04-15 | 2021-05-14 | 腾讯科技(深圳)有限公司 | Plug-in recognition model training method and device, electronic equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN113318455A (en) | 2021-08-31 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111773732B (en) | Target game user detection method, device and equipment | |
US20070232398A1 (en) | System and method for detecting collusion in online gaming via conditional behavior | |
CN107335220B (en) | Negative user identification method and device and server | |
CN113179263A (en) | Network intrusion detection method, device and equipment | |
CN104753949B (en) | Game data packet validity detection method and device | |
CN104463201A (en) | Method and device for recognizing driving state and driver | |
KR20180025692A (en) | Method, apparatus, and computer program for detecting bot program based on group using location information in online game | |
CN113318455B (en) | Game data processing method and device | |
CN112221162B (en) | Network game interaction method based on artificial intelligence recognition and intelligent game platform | |
CN111803956B (en) | Method and device for determining game plug-in behavior, electronic equipment and storage medium | |
KR20150100993A (en) | Computer system and computer implemented method for detecting bot based on self similarity | |
Vihanga et al. | Weekly seasonal player population patterns in online games: A time series clustering approach | |
KR101872345B1 (en) | System and method for detecting abnormal character by analyzing location characterlistic in on-line game | |
Dinh et al. | An empirical study of anomaly detection in online games | |
KR20130099786A (en) | Illegal user detecting apparatus and method | |
CN111209567B (en) | Method and device for judging perceptibility of improving robustness of detection model | |
CN113095672A (en) | Method and system for evaluating face image comparison algorithm | |
Zhang et al. | Detection of illegal players in massively multiplayer online role playing game by classification algorithms | |
CN110033295B (en) | Device and method for detecting rewarding importance of event in game | |
CN112422505A (en) | Network malicious traffic identification method based on high-dimensional extended key feature vector | |
KR20210061693A (en) | Methods and systems for recommending quests and items to users | |
Syed et al. | DeepGamble: Towards unlocking real-time player intelligence using multi-layer instance segmentation and attribute detection | |
CN112827176B (en) | Game level generating method and device, electronic equipment and storage medium | |
Zutis et al. | Who’s counting? real-time blackjack monitoring for card counting detection | |
Kummer et al. | Identifying niche stage in MMORPGs using ensemble classifier |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |