Method and system for detecting boosting gamer

Abstract

The present invention relates to a method performed on a computing device comprising processors, and a memory storing one or more programs executed by one or more of the processors, and the method may comprise the steps of: receiving a boosting gamer inquiry request from a user terminal; acquiring game-related information of a gamer included in the boosting gamer inquiry request; detecting whether the gamer is a boosting gamer by inputting the game-related information of the gamer to a pre-trained machine learning model; and generating a boosting-related report on the basis of whether the gamer is a boosting gamer and the game-related information of the gamer.

Description

Boosting Gamer Detection Method and System

[0001] The present invention relates to a method and system for detecting a boosting gamer capable of detecting a boosting gamer.

[0002] The present invention was derived from research conducted as part of the Ministry of Science and ICT's Information and Communication Broadcasting Innovation Talent Training Project (Project Unique Number: 1711179353, Project Number: 00156360, Research Project Name: Regional Intelligence Innovation Talent Training, Project Management Agency: Information and Communication Planning Evaluation Institute, Project Execution Agency: Soongsil University Industry-Academic Cooperation Foundation, Research Period: January 1, 2023 - December 31, 2023). Meanwhile, the Korean government has no property interest in any aspect of the present invention.

[0003] Recently, with the advancement of the Internet, online games, which allow players to connect and play online, have become increasingly popular. In online games, multiple players log in and compete. In competitive games, proxy play (or boosting play), where players play on behalf of other players to gain unfair advantages, often occurs.

[0004] This type of proxy play disrupts fair competition among users and diminishes the inherent fun and value of the game. However, the current detection of proxy players relies entirely on a user reporting system. When users report cheating, the game operating company manually verifies the reports. This makes it difficult to detect boosting players in real time, and its effectiveness is limited.

[0005] The purpose of the present invention is to provide a method and system for detecting a boosting gamer capable of detecting a boosting gamer during a game.

[0006] In addition, the present invention aims to provide a method and system for detecting a boosting gamer, which can inform the user terminal of the boosting gamer's probability of the gamer and send a report related to boosting in response to a request for inquiry about the boosting gamer.

[0007] In addition, the present invention aims to provide a method and system for detecting boosting gamers, which can check in real time at a glance whether other users are engaged in boosting activities within a game, thereby maintaining the soundness of the game community and securing reliability for the game.

[0008] In addition, the present invention aims to provide a method and system for detecting a boosting gamer, which can improve performance, such as increasing data processing speed or reducing required memory capacity, compared to a conventional system, by not having to separately run multiple programs when detecting a boosting gamer.

[0009] Meanwhile, the technical tasks to be achieved in the present invention are not limited to the technical tasks mentioned above, and other technical tasks not mentioned can be clearly understood by a person having ordinary knowledge in the technical field to which the present invention belongs from the description below.

[0010] A method for detecting a boosting gamer according to one embodiment of the present disclosure is a method performed in a computing device having one or more processors and a memory storing one or more programs executed by the one or more processors, the method comprising: receiving a request for inquiry about a boosting gamer from a user terminal; obtaining game-related information of a gamer included in the request for inquiry about a boosting gamer; inputting the game-related information of the gamer into a pre-trained machine learning model to detect whether the gamer is a boosting gamer; and generating a boosting-related report based on whether the gamer is a boosting gamer and the game-related information of the gamer.

[0011] In addition, the step of receiving the boosting gamer inquiry request may include the steps of: obtaining an authentication code from the user terminal; obtaining an access token based on the authentication code, and receiving personal information of the user based on the access token; issuing a JWT (Jason Web Token) to the user terminal based on the personal information of the user; and receiving a boosting gamer inquiry request from the user terminal to which the JWT is issued.

[0012] Additionally, the game-related information of the gamer may include the gamer's tier information and game execution indicator information for the gamer's previous games.

[0013] In addition, the step of detecting whether the gamer is a boosting gamer may include the step of inputting game execution indicator information for the gamer's previous games into the machine learning model to classify the gamer's tier; and the step of comparing the gamer's tier classified by the machine learning model with the tier based on the gamer's tier information to output a probability value of whether the gamer is a boosting gamer.

[0014] In addition, when a request for inquiry about a boosting gamer is received from the user terminal, the method further includes: determining a game that is a target of the request for inquiry about a boosting gamer as a game to be analyzed; and determining the number of preset tiers corresponding to the game to be analyzed as the total number of tiers of the game to be analyzed, and the step of detecting whether the gamer is a boosting gamer may include: inputting game execution indicator information for the game to be analyzed of the gamer into the machine learning model to classify the tier of the gamer based on the total number of tiers; and comparing the tier of the gamer classified by the machine learning model and the tier by the tier information of the gamer based on the total number of tiers, and outputting a probability value of whether the gamer is a boosting gamer.

[0015] Additionally, the boosting-related report may display a probability value as to whether a gamer responding to the boosting gamer inquiry request is a boosting gamer, and may include grounds for the probability value.

[0016] Additionally, the boosting-related report may include a visualization comparing the average game execution indices of other gamers with the same tier as the gamer and the game execution indices of the gamer's previous games.

[0017] The above boosting gamer detection method may further include the steps of: receiving a gamer record search request from a user terminal; obtaining game execution indicator information for previous games of the gamer corresponding to the gamer record search request; and processing the game execution indicator information for the previous games of the gamer to form gamer record data and transmitting the processed game execution indicator information to the user terminal.

[0018] The above boosting gamer detection method may further include a step of training the machine learning model, and the step of training the machine learning model may include a step of obtaining learning data for a preset number of recent games for each tier of each gamer and labeling the data based on tier information of each gamer; a step of excluding invalid data from the obtained learning data; a step of normalizing the learning data by dividing the learning data by the time the game was played; and a step of inputting the normalized learning data into a machine learning model so that the machine learning model classifies the tier of the gamer who played the game.

[0019] A boosting gamer detection system according to one embodiment disclosed includes: a user terminal transmitting a boosting gamer inquiry request; and a boosting management server receiving the boosting gamer inquiry request from the user terminal, obtaining game-related information of the gamer included in the boosting gamer inquiry request, inputting the game-related information of the gamer into a pre-trained machine learning model to detect whether the gamer is a boosting gamer, and reporting a result of whether the gamer is a boosting gamer to the user terminal.

[0020] The above user terminal can register gamers suspected of being boosting gamers as gamers of interest based on user input, and track and manage whether the gamers registered as gamers of interest are playing boosting games through a boosting gamer inquiry request.

[0021] According to one aspect of the disclosed invention, a boosting gamer can be detected during a game.

[0022] In addition, according to an embodiment of the present invention, in response to a request for inquiry about a boosting gamer from a user terminal, the boosting gamer probability of the corresponding gamer can be notified and a report related to boosting can be sent.

[0023] In addition, according to an embodiment of the present invention, it is possible to check at a glance in real time whether other users are engaged in boosting activities within the game, thereby maintaining the health of the game community and securing reliability of the game.

[0024] In addition, according to an embodiment of the present invention, when detecting a boosting gamer, it is not necessary to run multiple programs separately, so that performance such as an increase in data processing speed or a decrease in required memory capacity can be improved compared to a conventional system.

[0025] Meanwhile, the effects that can be obtained from the present invention are not limited to the effects mentioned above, and other effects that are not mentioned can be clearly understood by a person having ordinary skill in the technical field to which the present invention belongs from the description below.

[0026] FIG. 1 is a diagram illustrating a boosting gamer detection system according to one embodiment of the present invention.

[0027] FIG. 2 is a block diagram showing the configuration of a boosting management server according to one embodiment of the present invention.

[0028] FIG. 3 is a diagram showing game execution indicator information when the game is League of Legends in one embodiment of the present invention.

[0029] FIG. 4 is a diagram illustrating a gamer record data screen according to one embodiment of the present invention.

[0030] FIG. 5 is a diagram showing a predicted value that belongs to a specific tier of gamers output by a machine learning model according to one embodiment of the present invention.

[0031] FIG. 6 is a diagram showing a screen of indicator visualization analysis according to one embodiment of the present invention.

[0032] FIGS. 7 to 10 are diagrams showing boosting-related reports according to one embodiment of the present invention.

[0033] Figure 11 is a diagram showing a state in which learning data is labeled according to tiers in one embodiment of the present invention.

[0034] Figure 12 is a diagram schematically illustrating a preprocessing process of learning data according to one embodiment of the present invention.

[0035] Figure 13 is a diagram schematically illustrating the structure of a machine learning model according to one embodiment of the present invention.

[0036] FIG. 14 is a block diagram illustrating a computing environment including a computing device suitable for use in exemplary embodiments.

[0037] Like reference numerals refer to like components throughout the specification. This specification does not describe all elements of the embodiments, and general contents in the technical field to which the disclosed invention belongs or contents overlapping between embodiments are omitted. The term '~module' used in the specification may be implemented in software or hardware, and depending on the embodiments, multiple '~modules' may be implemented as a single component, or a single '~module' may include multiple components. The function provided by the '~module' may be performed separately by multiple components, or may be integrated with other additional components. The '~module' in this specification is not limited to software or hardware, and may be configured to be in an addressable storage medium, or may be configured to play one or more processors.

[0038] Furthermore, when a part is said to "include" a component, this does not exclude other components unless otherwise specifically stated, but rather means that other components may be included. Singular expressions include plural expressions unless the context clearly indicates otherwise. Identification numbers in each step are used for convenience of explanation and do not describe the order of each step, and each step may be performed in a different order than stated unless the context clearly indicates a specific order. The following describes the operating principles and embodiments of the disclosed invention with reference to the attached drawings.

[0039] Hereinafter, embodiments of the present invention will be described in more detail with reference to the attached drawings. The embodiments of the present invention may be modified in various ways, and the scope of the present invention should not be construed as being limited to the embodiments described below. These embodiments are provided to more fully explain the present invention to those of ordinary skill in the art. Accordingly, the shapes of elements in the drawings have been exaggerated for clarity.

[0040] In order to clearly solve the problem to be solved by the present invention, the composition of the invention is described in detail with reference to the attached drawings based on a preferred embodiment of the present invention, and when assigning reference numbers to components in the drawings, the same reference numbers are assigned to the same components even if they are in different drawings, and it is made clear in advance that components in other drawings may be cited when necessary when describing the drawings.

[0041] FIG. 1 is a diagram illustrating a boosting gamer detection system according to one embodiment of the present invention.

[0042] Referring to FIG. 1, the boosting gamer detection system (100) may include a user terminal (102), a boosting management server (104), and a game operation server (106). The user terminal (102), the boosting management server (104), and the game operation server (106) are each communicatively connected to each other through a communication network (150).

[0043] In one embodiment, the communication network (150) may include the Internet, one or more local area networks, wide area networks, a cellular network, a mobile network, other types of networks, or a combination of these networks.

[0044] A user terminal (102) is a terminal of a user who connects online and plays a game. The user terminal (102) may be a variety of electronic devices (e.g., mobile phones, smart phones, tablet PCs, notebook PCs, desktop PCs, PDAs (Personal Digital Assistants), etc.) capable of communicating with the boosting management server (104) and the game operation server (106).

[0045] A user terminal (102) can access a game operation server (106) and execute an online game provided by the game operation server (106). The game operation server (106) provides an online game to each user terminal (102) and may be a server computing device that performs functions related to execution and operation of the online game. The game operation server (106) can store various information (e.g., game log information, game execution index information, etc.) regarding the game being played by the user terminal (102).

[0046] In addition, the user terminal (102) can access the boosting management server (104) to perform member authentication. When member authentication is completed, the user terminal (102) can transmit a boosting gamer inquiry request to the boosting management server (104). At this time, the user terminal (102) can transmit a boosting gamer inquiry request including identification information (e.g., nickname) of a gamer input by the user (a gamer suspected of being a boosting gamer by the user) to the boosting management server (104). In one embodiment, if a user encounters a gamer suspected of being a boosting gamer while playing a game, the user can input the nickname of the gamer and transmit the boosting gamer inquiry request.

[0047] At this time, the user terminal (102) can register gamers suspected of being boosters as "Gamers of Interest" based on user input. In this case, the user can continuously track and manage whether the gamers registered as gamers of interest engage in boosting play. That is, by repeatedly sending a boosting gamer inquiry request containing the nickname of the gamer of interest while playing the game, the user can continuously check whether the gamer of interest engages in boosting play.

[0048] Additionally, the user terminal (102) may transmit a gamer record search request to the boosting management server (104). The gamer record search request may include identification information (e.g., nickname, etc.) of the gamer seeking to check the record of previous games. The user can check the latest record of other gamers through the gamer record search request.

[0049] The boosting management server (104) can confirm whether the gamer is a boosting gamer in response to a boosting gamer inquiry request from a user terminal (102). The boosting management server (104) can confirm whether the gamer is a boosting gamer based on machine learning technology. In response to a gamer record search request from a user terminal (102), the boosting management server (104) can transmit the gamer's record data to the user terminal (102).

[0050] FIG. 2 is a block diagram showing the configuration of a boosting management server (104) according to one embodiment of the present invention.

[0051] Referring to FIG. 2, the boosting management server (104) may include a member management module (111), a gamer record acquisition module (113), a boosting detection module (115), and an analysis module (117).

[0052] The member management module (111) can perform member management to provide a boosting gamer detection service. The member management module (111) can store user names, user email addresses, user log information, etc. for each user. The member management module (111) can perform member authentication for user terminals (102) accessing the boosting management server (104).

[0053] In one embodiment, the user terminal (102) may input member identification information such as an ID and password to obtain an authentication code (for example, the authentication code may be obtained from a Google API server) and then transmit the authentication code to the boosting management server (104). Then, the member management module (111) may transmit the authentication code to a Google API server (not shown) and then obtain an access token and receive personal information of the corresponding user (such as a user name and user email) based on the access token. The member management module (111) may issue a JWT (Jason Web Token) to the user terminal (102) based on the personal information of the corresponding user.

[0054] The member management module (111) can issue a JWT based on a user name, user email, preset secret key, and expiration date. The user terminal (102) to which the JWT has been issued can send a boosting gamer inquiry request to the boosting management server (104).

[0055] The gamer record acquisition module (113) can acquire game-related information of the gamer included in the boosting gamer inquiry request. The gamer record acquisition module (113) can transmit the game-related information of the gamer to the boosting detection module (115). Here, the game-related information of the gamer can include the gamer's tier information and game execution indicator information of the gamer's previous games. The gamer's tier information can refer to ranking or level information that serves as an indicator of the gamer's gaming prowess.

[0056] Additionally, game execution indicator information is data generated when a gamer runs a game, and may include various information depending on the type of game. For example, if the game is League of Legends, it may include game wins and losses, damage per minute, assists per game, vision score, gold earned per minute, percentage of damage taken by the team, number of kills per minute, kill contribution, vision score per minute, average wards per minute, average number of wards cleared, and number of control wards per minute. However, it is not limited to this, and game execution indicator information may of course vary depending on the game.

[0057] FIG. 3 is a diagram showing game execution indicator information when the game is League of Legends in one embodiment of the present invention.

[0058] The gamer record acquisition module (113) can extract the identification information of the gamer (e.g., the gamer's nickname, etc.) from the boosting gamer inquiry request, and then request and receive the game-related information of the gamer from the game operation server (106). At this time, the game operation server (106) can check whether the gamer has a game account, and if the game account exists, can extract the game-related information of the gamer and then transmit it to the gamer record acquisition module (113). When extracting game execution index information for the gamer's previous games, the game operation server (106) can extract game execution index information for a previously set number of games. For example, the game operation server (106) can extract game execution index information for the gamer's 10 most recent games.

[0059] In addition, the gamer record acquisition module (113) can acquire game execution indicator information for the gamer's previous games in response to a gamer record search request. The gamer record acquisition module (113) can process the game execution indicator information for the gamer's previous games to form gamer record data and then transmit the processed game execution indicator information to the user terminal (102).

[0060] FIG. 4 is a diagram illustrating a gamer record data screen according to one embodiment of the present invention.

[0061] The boosting detection module (115) can detect whether a gamer responding to a boosting gamer inquiry request is a boosting gamer (i.e., a proxy gamer) based on game-related information. The boosting detection module (115) may include a machine learning model (115a).

[0062] The boosting detection module (115) can input game-related information of a gamer suspected of boosting into a machine learning model (115a) to detect whether the gamer is a boosting gamer. Here, the machine learning model (115a) can be trained to predict whether the gamer is a boosting gamer by inputting game execution indicator information for the gamer's previous games and the gamer's tier information.

[0063] The machine learning model (115a) can classify the tier of a gamer by inputting game execution indicator information about the gamer's previous games. The boosting detection module (115) can detect whether the gamer is a boosting gamer by comparing the tier of the gamer classified by the machine learning model (115a) with the actual tier of the gamer (the tier based on the gamer's tier information). The boosting detection module (115) can output a probability value indicating whether the gamer is a boosting gamer.

[0064] FIG. 5 is a diagram showing a predicted value that belongs to a specific tier of a gamer output from a machine learning model (115a) according to one embodiment of the present invention.

[0065] Referring to FIG. 5, the boosting detection module (115) can detect whether the gamer is a boosting gamer based on the difference between the tier of the gamer (e.g., the probability of belonging to each tier) classified by the machine learning model (115a) and the actual tier of the gamer.

[0066] In one embodiment, if the game is League of Legends, the gamer's tier can be divided into 7 grades (grades 0 to 6) as described below, and if the predicted value of the gamer's specific tier output from the machine learning model (115a) does not fall within the red box shown in FIG. 5, the gamer can be determined to be a boosting gamer. However, this is not limited thereto, and the number of grades of tiers set for each game may vary depending on the game, and the size of the red box may be appropriately adjusted accordingly. For example, if the game is League of Legends, the boosting detection module (115) may determine that the gamer is a boosting gamer if there are 7 tiers and the difference between the gamer's tier (e.g., the probability of belonging to each tier) and the actual gamer's tier is 2 or more, and the gamer does not fall within the red box. However, if the game is a completely different game, there may be 10 tiers and if the difference between the gamer's tier and the actual gamer's tier is 3 or more, the gamer may be determined to be a boosting gamer if the gamer does not fall within the red box.

[0067] At this time, the number of tiers for a specific game may be preset for each game. In addition, the difference between the tier of the gamer (e.g., the probability of belonging to each tier) that serves as the standard for the boosting detection module (115) to determine that the gamer is a boosting gamer and the actual tier of the gamer may also be preset for each game. Of course, the difference between the tiers that serves as the standard for the boosting detection module (115) to determine that the gamer is a boosting gamer is not preset, but may be automatically determined by the boosting detection module (115) based on the number of tiers of the game being analyzed. For example, the boosting detection module (115) may determine the tier difference that serves as the standard for determining that a gamer of League of Legends is a boosting gamer to be 2 based on the fact that the total number of tiers in League of Legends is 7.

[0068] When the boosting detection module (115) receives a boosting gamer inquiry request from a user terminal (102), it can determine the game that is the target of the boosting gamer inquiry request as the game to be analyzed. For example, if the gamer's game that is the target of the boosting gamer inquiry request received from the user terminal (102) is League of Legends, the boosting detection module (115) can determine League of Legends as the game to be analyzed.

[0069] The boosting detection module (115) can determine the number of preset tiers corresponding to the game to be analyzed as the total number of tiers of the game to be analyzed. For example, if the gamer's game that is the target of the boosting gamer inquiry request received from the user terminal (102) is League of Legends, the boosting detection module (115) can determine the total number of tiers of the game to be analyzed as 7, since the number of tiers of the game called League of Legends is 7.

[0070] The boosting detection module (115) inputs game execution indicator information for the game being analyzed into a machine learning model, and can classify the gamer's tier based on the total number of tiers. In addition, the boosting detection module (115) can compare the gamer's tier classified by the machine learning model based on the total number of tiers with the tier based on the gamer's tier information, and output a probability value indicating whether the gamer is a boosting gamer.

[0071] For example, if the game to be analyzed is League of Legends, and the total number of tiers is 7, the boosting detection module (115) may determine 2, which is a preset reference value corresponding to the total number of tiers of 7, as a value of a tier difference suitable for detecting a boosting gamer. Alternatively, the boosting detection module (115) may determine 2, which is a preset reference value for detecting a boosting gamer for the game League of Legends, as a value of a tier difference suitable for detecting a boosting gamer.

[0072] In this way, the boosting gamer detection system (100) according to one embodiment can not only detect boosting gamers for a specific type of game, but can also separately receive information on tier differences suitable for detecting boosting gamers, regardless of the game of the gamer who wants to be analyzed transmitted by the user terminal (102), or automatically determine information on tier differences suitable for detecting boosting gamers based on the number of tiers of the game, and detect boosting gamers for the game that is the subject of analysis based on this.

[0073] Meanwhile, a detailed description of the structure and learning method of the machine learning model (115a) will be described later.

[0074] The analysis module (117) can analyze game execution indicator information for the gamer's previous games based on the gamer's tier information. The analysis module (117) can compare and visualize the average of each game execution indicator of other gamers who have the same tier as the gamer with the game execution indicators for the gamer's previous games. Here, comparing and visualizing the game execution indicators of other gamers with the game execution indicators for the gamer's previous games can be referred to as indicator visualization analysis. The analysis module (117) can transmit the screen of the indicator visualization analysis to the user terminal (102).

[0075] Figure 6 is a diagram illustrating a screen for indicator visualization analysis according to one embodiment of the present invention. Figure 6 shows game execution indicators higher than the average "GRANDMASER" and game execution indicators lower than the average "GRANDMASER" when the tier of the gamer (nickname: Hide on bush) is "GRANDMASER."

[0076] The analysis module (117) can generate a boosting-related report based on the probability value of whether a player is a boosting gamer from the boosting detection module (115) and the game-related information of the player. In one embodiment, the analysis module (117) can generate the boosting-related report in a document format (e.g., PDF, etc.). The analysis module (117) can transmit the boosting-related report to the user terminal (102). Upon receiving the boosting-related report, the user can report the boosting play to the game operator based on the boosting-related report.

[0077] Here, the boosting-related report may include the basis for the probability value of whether a player is a boosting gamer output from the boosting detection module (115). That is, when the probability value of whether a player is a boosting gamer is output from the boosting detection module (115), the basis for such probability value may be included in the boosting-related report.

[0078] Figures 7 through 10 illustrate a boosting-related report according to one embodiment of the present invention. While the boosting-related report can be generated as a single, continuous document, for convenience of explanation, the items included in the boosting-related report are presented separately in multiple drawings.

[0079] First, referring to Figure 7, a boosting report may display the probability that the gamer (nickname: "Eun Chae") who requested a boosting inquiry is a boosting gamer (proxy player) (A). Furthermore, the boosting report may include evidence based on indicator visualization analysis (B). Specifically, the boosting report may include a visualization comparing the average game performance indicators of other players in the same tier as the player in question with the game performance indicators of the player's previous games.

[0080] Also, referring to FIGS. 8 and 9, the boosting-related report may include as a basis the probabilities that the gamer belongs to each tier (e.g., Iron, Bronze, Silver, Gold, Platinum, Diamond, etc.) among the tiers classified into multiple tiers in the game (C). The analysis module (117) may analyze the game execution indicator information for the gamer's previous games to calculate the probability that the gamer belongs to each tier among the tiers classified into multiple tiers. At this time, the analysis module (117) may infer the actual tier of the gamer (D). That is, the analysis module (117) may check the probability that the gamer belongs to each tier, and infer that the tier with the highest probability is the gamer's actual tier.

[0081] Additionally, referring to FIG. 10, the boosting-related report may include final analysis text (E) explaining whether the gamer is a boosting gamer and the basis for this determination. The analysis module (117) may analyze the gamer's tier information and game performance indicators for the gamer's previous games to generate the final analysis text. For example, the final analysis text may be generated based on machine learning, but is not limited thereto.

[0082] According to the disclosed embodiment, in response to a request for inquiry about a boosting gamer from a user terminal, the boosting gamer probability of the corresponding gamer is notified and a report related to boosting is sent, thereby allowing the user to check at a glance in real time whether other users are engaged in boosting activities within the game, thereby maintaining the soundness of the game community and ensuring reliability of the game.

[0083] Meanwhile, the process of training a machine learning model (115a) in the boosting detection module (115) will be described. For convenience, the game League of Legends will be used as an example. However, the game to which the present invention is applied is not limited to this game, and it is clear that the present invention can be applied to various other online games.

[0084] The boosting detection module (115) can obtain learning data to be used for training the machine learning model (115a). The boosting detection module (115) can obtain learning data from the game operation server (106). Here, the learning data can include information on the tier of each gamer and game execution indicator information generated while each gamer is playing the game. In this case, the boosting detection module (115) can obtain learning data for the last 10 games for each gamer's tier.

[0085] The boosting detection module (115) can exclude invalid data from the acquired learning data from the data to be trained by the machine learning model (115a). For example, the boosting detection module (115) can exclude games that intentionally induce defeat, that violate the rules set in the game and have non-corresponding lines among the preset lines, or that have changed the API version of the game in the middle of the acquired learning data.

[0086] In League of Legends, players' tiers can be broadly categorized into general tiers (IRON, BRONZE, SILVER, GOLD, PLATINUM, DIAMOND) and top tiers (MASTER, GRANDMASTER, CHALLENGER). Because the top tier has a narrow data distribution, three tiers can be combined into a single tier, for a total of seven tiers.

[0087] The boosting detection module (115) can simplify the data by preprocessing the learning data acquired from the game operation server (106). That is, the learning data acquired from the game operation server (106) is in JSON format, so if it is used directly in the form of raw data, not only will the amount of data be large, but there will also be a lot of auxiliary information, so the data can be simplified through the preprocessing process.

[0088] The boosting detection module (115) can label each learning data according to its tier. That is, the boosting detection module (115) can classify each learning data into one of seven tiers based on the tier information of the corresponding gamer.

[0089] Figure 11 is a diagram showing a state in which learning data is labeled according to tiers in one embodiment of the present invention.

[0090] Referring to FIG. 11, the boosting detection module (115) can label tiers IRON with 0, BRONZE with 1, SILVER with 2, GOLD with 3, PLATINUM with 4, DIAMOND with 5, and MASTER, GRANDMASTER, and CHALLENGER with 6.

[0091] The boosting detection module (115) can normalize each learning data by dividing it by the time the game was played (i.e., game time). That is, since data values ​​vary significantly depending on the time the game was played, regardless of the gamer's tier, each learning data can be normalized by dividing it by game time.

[0092] Figure 12 is a diagram schematically illustrating a preprocessing process of learning data according to one embodiment of the present invention.

[0093] Referring to Fig. 12, 100,000 game match data can be acquired from the game operation server (106) via the Riot API, and data on the most recent 10 games can be extracted from among them for each tier. Then, a total of 67 data, including win / loss data, line data, and 65 data for classifying tiers, are extracted from each game, and a learning data set of a total of 670 columns is created by concatenating 10 game data, which can be classified by each tier and saved as a CSV file.

[0094] The boosting detection module (115) can input learning data for which the preprocessing process has been completed into the machine learning model (115a) to train the machine learning model (115a).

[0095] FIG. 13 is a diagram schematically showing the structure of a machine learning model (115a) according to one embodiment of the present invention.

[0096] Referring to FIG. 13, the machine learning model (115a) may include an embedding layer (121), an encoder (123), and a classifier (125).

[0097] The embedding layer (121) can embed the input training data. The input training data is 670-dimensional data, but can be converted to the model's dimension of 256 through embedding. At this time, the embedding layer (121) can assign order to each embedding through position embedding.

[0098] The encoder (123) can extract feature vectors from embedded training data. In one embodiment, the encoder (123) may be a transformer encoder. The transformer encoder learns the interactions between different parts of the input sequence through an attention mechanism. Since the transformer encoder is a well-known technology, a detailed description thereof will be omitted.

[0099] The classifier (125) can classify the tier of a gamer who played the game based on the feature vector extracted from the encoder (123). The feature vector extracted from the encoder (123) passes through a global average pooling layer, a fully connected layer, and a softmax layer, and the softmax layer can output a classification value for the tier of the gamer. At this time, the softmax layer can output the probability that the gamer belongs to each tier among a total of seven tiers.

[0100] FIG. 14 is a block diagram illustrating a computing environment (10) including a computing device suitable for use in exemplary embodiments. In the illustrated embodiment, each component may have different functions and capabilities other than those described below, and may include additional components other than those described below.

[0101] The illustrated computing environment (10) includes a computing device (12). In one embodiment, the computing device (12) may be a user terminal (102). In addition, the computing device (12) may be a boosting management server (104). In addition, the computing device (12) may be a game operation server (106).

[0102] A computing device (12) includes at least one processor (14), a computer-readable storage medium (16), and a communication bus (18). The processor (14) may cause the computing device (12) to operate according to the exemplary embodiments mentioned above. For example, the processor (14) may execute one or more programs stored in the computer-readable storage medium (16). The one or more programs may include one or more computer-executable instructions, which, when executed by the processor (14), may be configured to cause the computing device (12) to perform operations according to the exemplary embodiments.

[0103] A computer-readable storage medium (16) is configured to store computer-executable instructions or program code, program data, and / or other suitable forms of information. A program (20) stored in the computer-readable storage medium (16) includes a set of instructions executable by the processor (14). In one embodiment, the computer-readable storage medium (16) may be a memory (volatile memory such as random access memory, non-volatile memory, or a suitable combination thereof), one or more magnetic disk storage devices, optical disk storage devices, flash memory devices, any other form of storage medium that can be accessed by the computing device (12) and store desired information, or a suitable combination thereof.

[0104] A communication bus (18) interconnects various other components of the computing device (12), including the processor (14) and computer-readable storage media (16).

[0105] The computing device (12) may also include one or more input / output interfaces (22) that provide interfaces for one or more input / output devices (24) and one or more network communication interfaces (26). The input / output interfaces (22) and the network communication interfaces (26) are connected to the communication bus (18). The input / output devices (24) may be connected to other components of the computing device (12) via the input / output interfaces (22). Exemplary input / output devices (24) may include input devices such as pointing devices (such as a mouse or a trackpad), a keyboard, a touch input device (such as a touchpad or a touchscreen), a voice or sound input device, various types of sensor devices and / or photographing devices, and / or output devices such as display devices, printers, speakers and / or network cards. The exemplary input / output devices (24) may be included within the computing device (12) as a component constituting the computing device (12), or may be connected to the computing device (12) as a separate device distinct from the computing device (12).

[0106] The detailed description above is illustrative of the present invention. Furthermore, the foregoing description illustrates preferred embodiments of the present invention, and the present invention can be used in various other combinations, modifications, and environments. In other words, changes or modifications may be made within the scope of the inventive concepts disclosed herein, the scope equivalent to the written disclosure, and / or the scope of technology or knowledge in the art. The written embodiments illustrate the best possible state for implementing the technical idea of ​​the present invention, and various modifications required for specific applications and uses of the present invention are also possible. Therefore, the detailed description of the invention above is not intended to limit the present invention to the disclosed embodiments. Furthermore, the appended claims should be construed to include other embodiments.

Claims

1. One or more processors, and A method performed on a computing device having a memory storing one or more programs executed by one or more processors, A step of receiving a boosting gamer inquiry request from a user terminal; A step of obtaining game-related information of a gamer included in the above boosting gamer inquiry request; A step of inputting the game-related information of the gamer into a pre-trained machine learning model to detect whether the gamer is a boosting gamer; and A method for detecting a boosting gamer, comprising the step of generating a boosting-related report based on whether the gamer is a boosting gamer and the game-related information of the gamer.

2. In claim 1, The step of receiving the above boosting gamer inquiry request is: A step of obtaining an authentication code from the user terminal; A step of obtaining an access token based on the above authentication code and receiving personal information of the user based on the access token; A step of issuing a JWT (Jason Web Token) to the user terminal based on the personal information of the user; and A method for detecting a boosting gamer, comprising the step of receiving a boosting gamer inquiry request from a user terminal to which the JWT is issued.

3. In claim 1, The above gamer's game-related information is: A method for detecting boosting gamers, comprising information on the tier of the gamer and information on game execution indicators for the gamer's previous games.

4. In claim 3, The step of detecting whether the above gamer is a boosting gamer is as follows: A step of inputting game execution indicator information for the gamer's previous games into the machine learning model to classify the gamer's tier; and A method for detecting a boosting gamer, comprising a step of comparing the tier of the gamer classified by the machine learning model with the tier based on the tier information of the gamer and outputting a probability value as to whether the gamer is a boosting gamer.

5. In claim 4, When a boosting gamer inquiry request is received from the user terminal, a step of determining the game that is the target of the boosting gamer inquiry request as the analysis target game; and Further comprising a step of determining the number of preset tiers corresponding to the game to be analyzed as the total number of tiers of the game to be analyzed, The step of detecting whether the above gamer is a boosting gamer is as follows: A step of inputting game execution indicator information for the game to be analyzed of the gamer into the machine learning model to classify the tier of the gamer based on the total number of tiers; and A method for detecting a boosting gamer, comprising the step of comparing the tier of the gamer classified by the machine learning model and the tier based on the tier information of the gamer based on the total number of tiers, and outputting a probability value as to whether the gamer is a boosting gamer.

6. In claim 4, The above boosting related report is: A method for detecting a boosting gamer, wherein a probability value is displayed as to whether a gamer responding to the above-mentioned boosting gamer inquiry request is a boosting gamer, and the basis for the probability value is included.

7. In claim 6, The above boosting related report is: A method for detecting boosting gamers, comprising visualizing the average game execution indices of other gamers having the same tier as the gamer and the game execution indices of the gamer's previous games.

8. In claim 1, The above boosting gamer detection method is, A step of receiving a request for gamer record search from a user terminal; A step of obtaining game execution indicator information for previous games of a gamer corresponding to the above gamer record search request; and A boosting gamer detection method further comprising a step of processing game execution indicator information for the gamer's previous games to form gamer record data and then transmitting the data to the user terminal.

9. In claim 1, The above boosting gamer detection method is, Further comprising a step of training the machine learning model, The step of training the above machine learning model is: A step of acquiring learning data for a preset number of recent games for each gamer's tier and labeling the data based on each gamer's tier information; A step of excluding invalid data from the acquired learning data; A step of normalizing the above learning data by dividing it by the time the game was played; and A boosting gamer detection method, comprising a step of inputting the normalized learning data into a machine learning model and training the machine learning model to classify the tier of gamers who played the game.

10. User terminal sending a boosting gamer inquiry request; and A boosting gamer detection system, comprising a boosting management server that receives a boosting gamer inquiry request from the user terminal, obtains game-related information of the gamer included in the boosting gamer inquiry request, inputs the gamer's game-related information into a pre-trained machine learning model to detect whether the gamer is a boosting gamer, and reports the result of whether the gamer is a boosting gamer to the user terminal.

11. In claim 10, The above user terminal, A boosting gamer detection system that registers gamers suspected of being boosting gamers as gamers of interest based on user input, and tracks and manages whether the gamers registered as gamers of interest are playing boosting games through boosting gamer inquiry requests.

Images