KR20230060286A - System for providing prediction information and method thereof - Google Patents

Abstract

Disclosed in the present invention are a system for providing game outcome prediction information, and a method thereof. That is, according to the present invention, in an in-game state, a learning function is performed using game information of a plurality of players in a team including the user of a terminal and game information of a plurality of other players in an opposing team as input values of a statistical prediction model, and by providing prediction information about a match between the teams based on the results of the learning function, whether the game is won or lost is checked depending on the combination of multiple players forming the same team, and it is possible to increase the player's concentration on the game by arousing interest and attention.

Description

Game prediction information providing system and method

The present invention relates to a game prediction information providing system and method, and more particularly, in an in-game state, game information of a plurality of players in a team including a user of a terminal and game information of a plurality of other players included in an opposing team are converted into a statistical prediction model. An object of the present invention is to provide a match prediction information providing system and method for performing a learning function as an input value of and providing match prediction information between teams according to a result of performing the learning function.

This sport (e-Sports) is a sport that competes online using a computer network or the Internet.

Among these sports, strategic simulation games in which victory or defeat is determined according to various strategies of players include Starcraft and League of Legends.

In this strategy simulation game, various experiences, strategies, tactics, win-loss information, etc. of players constituting the same team affect the match against the opponent team, but the comparison between the players constituting the same team and the players constituting the opponent team Match predictions are not provided properly.

Korean Registered Patent No. 10-2021-0096831 [Title: Online game helper service system targeting users who use online PC and mobile games, and business method using the service]

An object of the present invention is to perform a learning function by using game information of a plurality of players in a team including a user of a terminal and game information of a plurality of other players included in an opposing team as input values of a statistical prediction model in an in-game state, An object of the present invention is to provide a match prediction information providing system and method for providing match prediction information between teams according to a result of performing a learning function.

Another object of the present invention is to provide a game prediction information providing system and method for providing game prediction information based on multi-search when a player forms a team with a plurality of players selected by the player.

A match prediction information providing system according to an embodiment of the present invention, when a player corresponding to a game nickname transmitted from a match prediction information providing server is playing a game in real time, checks a game in which the player corresponding to the game nickname is participating , Game nicknames of a plurality of players who are on the same team as the game nicknames participating in the identified game and game nicknames of a plurality of other players who are the opposing team, respectively, and game for each of the plurality of players participating in the game a game server that checks the character type, the selected item type, and the selected position, respectively, and provides them to the match prediction information providing server; and game nicknames of a plurality of players of the same team as the game nickname and game nicknames of a plurality of other players of the opposing team among the overall performance information and statistical information for each of the plurality of game nicknames managed by the match prediction information providing server. Comprehensive record information and statistical information are retrieved, respectively, and input including comprehensive record information and statistical information corresponding to the searched game nickname for each player participating in the game, game character type for each player, selected item type, and selected position By performing a learning function based on the value, the contribution to victory of each player for each position is calculated based on the result of performing the learning function, and match prediction information is generated based on the contribution to victory of multiple players for each team calculated for each position. , the match prediction information providing server for transmitting the player's overall record information, statistical information, and the generated match prediction information corresponding to the searched game nickname to the terminal.

As an example related to the present invention, the match prediction information includes expected victory probability per team, average KDA (kill/death/assist number) per team, average damage per team, average gold per team, average ward installation per team, and game information per player who is a team member may contain at least one.

In the method for providing match prediction information according to an embodiment of the present invention, when a player corresponding to a game nickname transmitted from a terminal is playing a game in real time by a match prediction information providing server in conjunction with a game server, the game nickname checking the game in which the corresponding player is participating; Game nicknames of a plurality of players who are on the same team as the game nickname participating in the confirmed game together with the game server by the match prediction information providing server, and game nicknames of a plurality of other players who are the opposing team checking each, and checking the game character type, the selected item type, and the selected position for each of the plurality of players participating in the game; Game nicknames of a plurality of players who are on the same team as the game nickname and other plurality of opponent teams among the overall performance information and statistical information for each of a plurality of game nicknames managed by the game prediction information providing server by the match prediction information providing server Retrieving comprehensive overall information and statistical information corresponding to the game nickname of the player of the above; By the match prediction information providing server, an input value including overall performance information and statistical information corresponding to the searched game nickname for each player participating in the game, game character type for each of a plurality of players, selected item type, and selected position Performing a learning function based on the basis, and calculating the victory contribution of each player for each position based on the result of performing the learning function; generating, by the match prediction information providing server, match prediction information based on contributions to victory of a plurality of players for each team calculated for each position; and transmitting, by the match prediction information providing server, comprehensive game record information, statistical information, and generated match prediction information of the player corresponding to the searched game nickname to the terminal.

As an example related to the present invention, the step of calculating each player's contribution to victory for each position includes the overall record information of players for each position according to the positions of a plurality of players included in our team and the opposing team participating in the same game. And information including statistical information, game character type, and selected item type is used as an input value for a statistical prediction model for each tier-rank-position to perform a learning function, and according to the result of performing the learning function, the player's contribution to victory for each position each can be calculated.

As an example related to the present invention, the step of calculating each player's contribution to victory for each position includes a preset number of variables set for each position, a preset weight corresponding to the variables, and the searched player-specific overall record information. A learning function is performed using information including values corresponding to the variables as input values of a statistical prediction model for each tier, rank, and position, and the player's contribution to victory can be calculated for each position according to the result of performing the learning function.

As an example related to the present invention, the step of receiving, by the terminal, comprehensive overall information, statistical information, and the match prediction information of a player corresponding to the game nickname transmitted from the match prediction information providing server; displaying, by the terminal, a game nickname detailed information screen including the received comprehensive information by default; and displaying, by the terminal, statistical information related to the received game nickname corresponding to the selected statistical tap when a statistics tab in a game nickname detailed information screen displayed on the terminal is selected. .

As an example related to the present invention, the step of receiving, by the terminal, comprehensive overall information, statistical information, and the match prediction information of a player corresponding to the game nickname transmitted from the match prediction information providing server; and displaying match prediction information related to a game in which the game nickname is participating, when an in-game information tab in the game nickname detailed information screen displayed on the terminal is selected.

The system for providing match prediction information according to an embodiment of the present invention receives a plurality of game nicknames and identification information of the terminal transmitted from a terminal, and among the comprehensive overall information and statistical information for each game nickname managed by a match prediction information providing server Searching for overall comprehensive information and statistical information corresponding to the received plurality of game nicknames, and performing a learning function based on an input value including comprehensive comprehensive information and statistical information for each of the retrieved plurality of game nicknames, Based on the result of performing the function, the contribution to victory of each player for each position is calculated, and the expected win probability is calculated based on the contribution to victory of multiple players calculated for each position, and the calculated expected win probability and multiple game nicknames the match prediction information providing server for transmitting each recommended position information and summary information for each of the searched plurality of game nicknames to the terminal; and the terminal displaying an expected win probability transmitted from the match prediction information providing server, recommended position information for each of a plurality of game nicknames, and summary information for each of the plurality of game nicknames.

As an example related to the present invention, the summary information includes at least one of a game nickname, recent winning streak/loss streak information, level, pick rate and win rate and IQ for each preferred game character, pick rate and win rate and IQ for each preferred position, and recent play results. can do.

A method for providing match prediction information according to an embodiment of the present invention includes receiving, by a match prediction information providing server, a plurality of game nicknames transmitted from a terminal and identification information of the terminal; By the match prediction information providing server, the match prediction information providing server retrieves comprehensive overall information and statistical information corresponding to the received plurality of game nicknames from among comprehensive overall information and statistical information for each of the plurality of game nicknames managed by the match prediction information providing server. doing; The match prediction information providing server performs a learning function based on input values including overall overall information and statistical information for each of the retrieved game nicknames, and the player's contribution to victory for each position based on the result of performing the learning function Calculating each; calculating, by the game prediction information providing server, an expected victory probability based on the contributions to victory of a plurality of players calculated for each position; transmitting, by the match prediction information providing server, the calculated expected winning probability, recommended position information for each of a plurality of game nicknames, and summary information for each of the plurality of game nicknames retrieved to the terminal; and displaying, by the terminal, the expected win probability transmitted from the game prediction information providing server, recommended position information for each of a plurality of game nicknames, and summary information for each of the plurality of game nicknames.

The present invention performs a learning function by taking game information of a plurality of players in a team including a user of a terminal and game information of a plurality of other players included in an opponent team as input values of a statistical prediction model in an in-game state, and learning function By providing match prediction information between teams according to the results of the performance, it is possible to check whether the game is won or lost according to the combination of multiple players constituting the same team together, and to increase the concentration on the game by arousing interest and interest of the players. There is an effect.

In addition, the present invention provides match prediction information according to multi-search when forming a team with a plurality of players selected by a player, thereby constructing an optimal team member according to a virtual team combination between a plurality of players before game play. It has the effect of increasing the interest of the player.

1 is a block diagram showing the configuration of a match prediction information providing system according to an embodiment of the present invention.
2 and 3 are flowcharts illustrating a method of providing match prediction information according to a first embodiment of the present invention.
4 to 14 are diagrams showing examples of screens of a terminal according to an embodiment of the present invention.
15 is a flowchart illustrating a method of providing match prediction information according to a second embodiment of the present invention.
16 to 18 are diagrams illustrating examples of screens of a terminal according to an embodiment of the present invention.

It should be noted that technical terms used in the present invention are only used to describe specific embodiments and are not intended to limit the present invention. In addition, technical terms used in the present invention should be interpreted in terms commonly understood by those of ordinary skill in the art to which the present invention belongs, unless specifically defined otherwise in the present invention, and are excessively inclusive. It should not be interpreted in a positive sense or in an excessively reduced sense. In addition, when the technical terms used in the present invention are erroneous technical terms that do not accurately express the spirit of the present invention, they should be replaced with technical terms that those skilled in the art can correctly understand. In addition, general terms used in the present invention should be interpreted as defined in advance or according to context, and should not be interpreted in an excessively reduced sense.

Also, singular expressions used in the present invention include plural expressions unless the context clearly dictates otherwise. Terms such as "consisting of" or "comprising" in the present invention should not be construed as necessarily including all of the various elements or steps described in the invention, and some of the elements or steps may not be included. It should be construed that it may, or may further include additional components or steps.

In addition, terms including ordinal numbers such as first and second used in the present invention may be used to describe components, but components should not be limited by the terms. Terms are used only to distinguish one component from another. For example, a first element may be termed a second element, and similarly, a second element may be termed a first element, without departing from the scope of the present invention.

Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the accompanying drawings, but the same or similar components are given the same reference numerals regardless of reference numerals, and redundant description thereof will be omitted.

In addition, in describing the present invention, if it is determined that a detailed description of a related known technology may obscure the gist of the present invention, the detailed description will be omitted. In addition, it should be noted that the accompanying drawings are only for easily understanding the spirit of the present invention, and should not be construed as limiting the spirit of the present invention by the accompanying drawings.

1 is a block diagram showing the configuration of a match prediction information providing system 10 according to an embodiment of the present invention.

As shown in FIG. 1 , the game prediction information providing system 10 is composed of a terminal 100 , a game server 200 and a game prediction information providing server 300 . Not all of the components of the match prediction information providing system 10 shown in FIG. 1 are essential components, and the match prediction information providing system 10 may be implemented by more components than those shown in FIG. There is, and the game prediction information providing system 10 may be implemented with fewer components.

The terminal 100 includes a smart phone, a portable terminal, a mobile terminal, a foldable terminal, a personal digital assistant (PDA), and a portable multimedia device (PMP). Player terminal, telematics terminal, navigation terminal, personal computer, notebook computer, slate PC, tablet PC, ultrabook, wearable device Device, for example, watch type terminal (Smartwatch), glass type terminal (Smart Glass), HMD (Head Mounted Display), etc.), Wibro terminal, IPTV (Internet Protocol Television) terminal, smart TV, digital broadcasting It can be applied to various terminals such as terminals, audio video navigation (AVN) terminals, audio/video (A/V) systems, flexible terminals, digital signage devices, and the like.

The game server 200 and the match prediction information providing server 300 may be implemented in the form of a web server, database server, proxy server, or the like. In addition, in the game server 200 and the match prediction information providing server 300, a network load balancing mechanism, or the corresponding game server 200 and the match prediction information providing server 300 may operate on the Internet or other networks. One or more of a variety of software may be installed, and through this, it may be implemented as a computerized system. Also, the network can be an http network, a private line, an intranet or any other network. Furthermore, the connection between the terminal 100, the game server 200, and the match prediction information providing server 300 may be connected through a secure network to prevent data from being attacked by any hacker or other third party. In addition, the game server 200 and the match prediction information providing server 300 may include a plurality of database servers, and these database servers may connect to the game server through any type of network connection including a distributed database server architecture. 200 and the match prediction information providing server 300 may be separately connected.

Each of the terminal 100, the game server 200, and the match prediction information providing server 300 includes a communication unit (not shown) and various information and programs (or applications) for performing a communication function with other terminals. A storage unit (not shown) for storing, a display unit (not shown) for displaying various information and program execution results, an audio output unit (not shown) for outputting audio information corresponding to the various information and program execution results, A control unit (not shown) for controlling various components and functions of each terminal may be included.

The terminal 100 communicates with the game server 200, the match prediction information providing server 300, and the like. At this time, the terminal 100 is a terminal possessed by a player (or user) who plays a game provided by the game server 200 and wants to receive match prediction information provided by the match prediction information providing server 300. can be

In addition, the terminal 100 is provided with a game play function, an item purchase function, etc. through a dedicated app and/or website provided by the game server 200 by interworking with the game server 200. Sign up as a user (or player) and register personal information in the game server 200 . At this time, the personal information includes ID, e-mail address, password (or password), name, gender, date of birth, contact information, address (or address information/region of residence), and the like.

In addition, the terminal 100 may register as a user in the game server 200 using SNS account information or other site account information to which the user of the terminal 100 has joined. Here, the SNS account may be information related to Facebook, Twitter, Instagram, Kakao Story, Naver Blog, and the like. In addition, the account of the other site may be information related to YouTube, Kakao, Naver, and the like.

In addition, when performing the membership sign-up process, the terminal 100 must complete the authentication function through a user authentication means (eg, mobile phone, credit card, i-PIN, etc.) to complete the membership sign-up process for the game server 200. can be completed normally.

In addition, after membership registration is complete, the terminal 100 uses a dedicated game app (or application/application program/specific app) provided from the game server 200 in order to use the service provided by the game server 200. ) is installed in the terminal 100. In this case, the dedicated game app may be an app for performing a game play function, an item purchase function, and the like.

In addition, after membership registration is completed, the terminal 100 may display a discount coupon provided from the game server 200 through a corresponding dedicated game app. In this case, the discount coupon may be a discount coupon that includes discount information of a certain percentage when performing a game play function, item purchase function, etc. provided by the corresponding game server 200 .

In addition, in order to play a game provided by the game server 200, the terminal 100 interworks with the game server 200 and a payment server (not shown) to perform a payment function according to a subscription function. At this time, the game server 200 performs card payment, automatic transfer through linkage with a bank payment account, payment using cash equivalent points or cash remaining in the account of the terminal 100 subscribed to the game server 200, A payment function can be performed through a simple payment system including Kakao Pay, Naver Pay, and the like.

When the payment fails, the terminal 100 receives information indicating that the payment has failed transmitted from the game server 200 (or the payment server) (eg, balance shortage, limit exceeded, etc.), and Outputs (or displays) information indicating that the received payment has failed.

In addition, the terminal 100 receives a payment function execution result transmitted from the game server 200 after the payment function is normally performed. Here, the result of performing the payment function includes a subscription period, payment amount, payment date and time information, and the like.

In addition, the terminal 100 has a function to check comprehensive overall information for each player through a dedicated app and/or website provided by the match prediction information providing server 300 by interworking with the match prediction information providing server 300. , Sign up as a general user (or player) to receive statistical information check function for each player, in-game information check function for each player, multi-search check function for each player, etc., and personal information, etc. Register in the provision server 300. At this time, the personal information includes ID, e-mail address, password (or password), name, gender, date of birth, contact information, address (or address information/region of residence), and the like.

In addition, the terminal 100 may register as a user in the match prediction information providing server 300 using SNS account information or other site account information to which the user of the corresponding terminal 100 has joined. Here, the SNS account may be information related to Facebook, Twitter, Instagram, Kakao Story, Naver Blog, and the like. In addition, the account of the other site may be information related to YouTube, Kakao, Naver, and the like.

In addition, when performing a membership sign-up procedure, the terminal 100 must complete an authentication function through a user authentication means (including, for example, a mobile phone, credit card, i-PIN, etc.) to become a member of the match prediction information providing server 300. You can complete the registration process normally.

In addition, after membership registration is completed, the terminal 100 uses a dedicated app (or application/ An application program/specific app) is installed in the corresponding terminal 100 . In this case, the dedicated app may be an app for performing a function for checking total overall information for each player, a function for checking statistical information for each player, a function for checking in-game information for each player, and a function for checking multi-search for each player.

In addition, after membership registration is completed, the terminal 100 may display a discount coupon provided from the match prediction information providing server 300 through a corresponding dedicated game app. At this time, the discount coupon is used when the game prediction information providing server 300 performs a function to check overall information for each player, a function to check statistical information for each player, a function to check in-game information for each player, and a multi-search check function for each player. It may be a discount coupon including discount information of a certain percentage.

In addition, the terminal 100 is provided by the match prediction information providing server 300, such as a function to check the overall record information for each player, a function to check statistical information for each player, a function to check in-game information for each player, and a multi-search check function for each player. To this end, different payment functions are performed according to different subscription functions in conjunction with the match prediction information providing server 300 and the payment server. At this time, the match prediction information providing server 300 performs card payment, automatic transfer through linkage with a bank payment account, and cash points remaining in the account of the terminal 100 subscribed to the match prediction information providing server 300. The payment function can be performed through cash payment, simple payment including Kakao Pay, Naver Pay, and the like.

If the payment fails, the terminal 100 receives information indicating that the payment has failed (eg, including balance shortage, limit exceeding, etc.) transmitted from the match prediction information providing server 300 (or the payment server). and outputs (or displays) information indicating that the received payment has failed.

Also, after the payment function is normally performed, the terminal 100 receives another payment function performance result transmitted from the match prediction information providing server 300 . Here, the result of performing the other payment function includes other subscription period, other payment amount, and other payment date and time information.

In addition, the terminal 100 executes a dedicated game app pre-installed in the corresponding terminal 100 and displays a game app execution result screen according to the execution of the dedicated game app. At this time, the game app execution result screen is a game start menu (or button/item) for playing a game according to the game mode, a home menu for moving to a home screen (or standby screen), and a strategic strategy for playing a game in an individual game. Team battle menu, Battle menu to play a tournament type game between teams, My information menu to check information such as proficiency, rank, and honor, information related to game characters (e.g. champions in LOL, skins) , emotes, runes, spells/spells, etc.) to check game character information menus, workshop menus to check information related to items you own, and your own to trade items according to the game character's level (or grade). Store menu, store menu for trading general items, cache menu for displaying information about a player's in-game cash (e.g. including Role Play at LOL, blue essence, etc.), player's in-game Level menu to display levels, friend list menu to display a list of one or more players with whom the player has friends, chat window menu to chat with other players, information related to missions (or quests) It includes a mission menu to check, a setting menu to set dedicated game app functions, and so on. Here, the terminal 100 is registered as a member of the game server 200 that provides the dedicated game app, and uses an ID and password according to membership registration, a barcode or QR code including the ID, and the like. When the game app is executed, a log-in procedure is performed to perform one or more functions (eg, game play function, item purchase function, etc.) of the dedicated game app.

In addition, when a preset game start menu is selected from the game app execution result screen displayed on the terminal 100, the terminal 100 is configured to play a game corresponding to a game mode selected by a player (or user). Displays the game selection screen. Here, the game selection screen includes a plurality of game modes provided by the game server 200 that manages the corresponding dedicated game app. At this time, the plurality of game modes include a play with a person provided by LOL or a play selection menu by artificial intelligence, a summoner's gorge mode (or menu / button / item) to play a game between teams, and all players Random total battle mode to play the game using randomly selected game characters (or champions), strategic team battle mode to play the game in individual battles, and other types of games provided as events (e.g., team members' votes) and an event mode for playing a game using a game character that has obtained the most votes by , etc.).

In addition, after one game mode is selected from among a plurality of game modes included in the game selection screen displayed on the terminal 100, a preset match search menu displayed on one side of the screen related to the selected game mode is selected. In this case, the terminal 100 matches a player to play with through the mediation of the game server 200 .

In addition, when a plurality of players to play with the player of the terminal 100 are matched, the terminal 100 matches the player according to the player selection (or player/user touch/input/control) of the terminal 100 Determines whether or not to proceed with a game with multiple players.

As a result of the determination, when the game with the matched plurality of players is rejected, the terminal 100 performs a new plurality of player matching process through the mediation of the game server 200 .

In addition, as a result of the determination, when the game progress with the plurality of matched players is accepted (or when all game progress acceptance is determined for each matched plurality of players), the terminal 100 is the terminal 100 According to player selection (or player/user touch/input/control), a player of the corresponding terminal 100 selects a game character to be used. Here, the game character to be used by the player may correspond to a champion in the LOL. At this time, the terminal 100 interlocks with the game server 200 to prohibit the use of a specific game character that is difficult to play against in correspondence with the game character used by the player of the terminal 100, which is set in advance in conjunction with the game server 200. Among the plurality of game characters, one or more game characters that cannot be played by the players of the opposing team may be selected, and the one or more game characters that the players of the selected opposing team cannot play may be configured so that the players of the opposing team cannot select them. there is.

In addition, when all players participating in the game complete game character selection, the terminal 100 interworks with the game server 200 and displays a loading window. Here, the loading window targets all players who participated in the game, and includes information related to game characters selected by a plurality of players belonging to our team including the player of the terminal 100 (for example, in LOL). tier, rank, spell, rune, skin, level, etc.), and information related to game characters selected by multiple players belonging to the opposing team.

As such, the terminal 100 (or a player of the terminal 100) executes a dedicated game app installed in the terminal 100, and performs preliminary work for game play using a game character that the player wants to play. can complete.

In addition, the terminal 100 accesses a website provided by the match prediction information providing server 300 in order to check match prediction information related to a game in which the terminal 100 is participating, and provides information provided by the website. Display the initial screen. At this time, the initial screen includes a record search menu for checking the player's record, a multi-search menu for providing match prediction information of a team composed of a plurality of players, a champion analysis menu for providing information by game character, and game ranking information. It includes a ranking menu to provide, a community menu to share information between players, an e-sports menu to provide match prediction events for each major game, and a setting menu to set user information. Here, the terminal 100 is registered as a member of the match prediction information providing server 300 that provides the website, and uses an ID and password according to membership registration, a barcode or QR code including the ID, and the like. When accessing the above website, a login procedure is performed to perform one or more functions of the website (e.g., a function to check overall record information for each player, a function to check statistical information for each player, a function to check in-game information for each player, and a function to check multi-search for each player). etc.) can be performed.

In addition, when a preset record search menu is selected on the initial screen displayed on the terminal 100, the terminal 100 checks (or searches) the record of a specific player that the player of the corresponding terminal 100 wants to check. Displays the total search screen for Here, the record search screen includes a game nickname input item for inputting the game nickname (or summoner name) of the player to check the record of a specific player (or match prediction information of a specific player currently participating in the game). . At this time, when accessing the initial screen, the terminal 100 may display a total search screen corresponding to the total search menu by default.

In addition, the terminal 100 is a game nickname corresponding to the game nickname input item according to the player input (or player / user selection / touch / control) of the terminal 100 on the overall search screen displayed on the terminal 100 receive

In addition, the terminal 100 transmits the received game nickname and identification information of the terminal 100 to the match prediction information providing server 300 . Here, the identification information of the terminal 100 includes MDN (Mobile Directory Number), mobile IP, mobile MAC, Sim (subscriber identity module: subscriber identity module) card-specific information, serial number, and the like.

In addition, the terminal 100 receives information indicating that the corresponding game nickname transmitted from the game prediction information providing server 300 is not a member registered in the game server 200 in response to the previously transmitted game nickname, etc. and displays (or outputs) information indicating that the received corresponding game nickname is not a member registered in the game server 200 .

In addition, the terminal 100 receives comprehensive game record information, statistical information, match prediction information, etc. transmitted from the game prediction information providing server 300 in response to a previously transmitted game nickname.

In addition, the terminal 100 displays the game nickname detailed information screen including the received overall score information by default. At this time, when the match prediction information is received by the terminal 100, the terminal 100 may display a preset on state icon on one side of the in-game information tab of the game nickname detailed information screen. Here, the game nickname detailed information screen includes a game nickname item of the corresponding player, a ranking item, a record update menu item, a recent update information item, a level and record information item in solo rank mode, a level in free 5:5 rank mode and Stats information item, Stats tab to check overall stats information, Stats tab to check statistical information, In-game information to check match predictions for the same team and the opposing team while participating in the game in real time including tabs, etc. At this time, the in-game information tab may be a menu for providing a screen (or page) for checking the strength of each team and the expected victory probability.

In addition, when a statistics tab (or menu/item) in the game nickname detailed information screen displayed on the terminal 100 is selected, the terminal 100 determines information related to the received corresponding game nickname corresponding to the selected statistics tab. Display statistical information.

In addition, when an in-game information tab in the game nickname detailed information screen displayed on the terminal 100 is selected, the terminal 100 displays a player corresponding to the received game nickname corresponding to the selected in-game information tab. Displays match prediction information related to the game being played. At this time, the in-game information tab may be activated only when the match prediction information is received from the terminal 100 .

In addition, when a preset multi-search menu is selected on the initial screen displayed on the terminal 100, the terminal 100 displays match prediction information of a team composed of a plurality of players selected by the player of the terminal 100. Displays the multi-search screen for confirmation. Here, the multi-search screen includes a plurality of game nickname input items for inputting game nicknames (or summoner names) of a plurality of players who will form the same team. At this time, when accessing the initial screen, the terminal 100 may display a total search screen corresponding to the total search menu by default.

In addition, the terminal 100 displays a plurality of game nickname input items corresponding to a plurality of game nickname input items according to player input (or player/user selection/touch/control) of the terminal 100 on the multi-search screen displayed on the terminal 100. Receives the game nickname of

In addition, the terminal 100 transmits the received plurality of game nicknames and identification information of the terminal 100 to the match prediction information providing server 300 . Here, the identification information of the terminal 100 includes MDN, mobile IP, mobile MAC, Sim (subscriber identification module) card-specific information, serial number, and the like.

In addition, the terminal 100 receives the predicted winning probability transmitted from the match prediction information providing server 300 in response to the plurality of game nicknames transmitted previously, the recommended position information for each of the plurality of game nicknames, and the plurality of game nicknames. receive summary information, etc.

In addition, the terminal 100 displays (or outputs) the received expected win probability, recommended position information for each of the plurality of game nicknames, and summary information for each of the plurality of game nicknames.

In an embodiment of the present invention, various functions (eg, game play function, item purchase function, overall overall information check function for each player, statistical information check function for each player, in-game information check function for each player, multi-search check function for each player, etc.) can be performed.

The game server 200 communicates with the terminal 100 and the match prediction information providing server 300 and the like.

The game server 200 described in the present invention mainly describes a server that provides a role (league of legends: LOL, League of Legends, hereinafter referred to as LOL), but is not limited thereto, and the terminal It may be a server that provides various types of games that players of (100) can play.

In addition, the main game terms described in the present invention may be game terms used in the corresponding LOL.

In addition, the game server 200 performs a membership registration procedure for users of the terminal 100 and the like.

In addition, the game server 200 registers personal information related to the user of the terminal 100 and the like. At this time, the game server 200 may register (or manage) corresponding personal information in a DB server (not shown).

In addition, the game server 200 performs a member management function for users such as the terminal 100 .

In addition, the game server 200 provides a dedicated game app and/or game website that provides a game play function, item purchase function, etc. to the terminal 100 and the like.

In addition, the game server 200 provides a bulletin board function for announcements, events, and the like.

In addition, the game server 200 interworks with the terminal 100 and the payment server to provide a payment function according to the execution of the subscription function in the terminal 100 for a specific game provided by the corresponding game server 200. carry out

If the payment function fails, the game server 200 provides payment failure information (including, for example, payment date, payment amount, failure information (including, for example, insufficient balance, exceeding limit, etc.)) (or indicates that the payment has failed). information) is provided to the terminal 100.

Also, after the payment function with the terminal 100 is normally performed, the game server 200 transmits the payment function execution result provided from the payment server to the terminal 100 . Here, the result of performing the payment function includes a subscription period, payment amount, payment date and time information, and the like.

In addition, the game server 200 manages (or stores/registers) the result of performing the payment function by mapping (or matching/linking) the corresponding terminal 100 (or account information related to the terminal 100). .

In addition, the game server 200 provides various information, functions, etc. for playing the game provided by the game server 200 through the dedicated game app on the terminal 100 according to the subscription function. .

In addition, when trying to use a game service provided by the game server 200 by executing a dedicated game app pre-installed in the terminal 100 on the terminal 100, the game server 200 is the terminal ( 100), an authentication process according to the log-in procedure is performed, and it is confirmed whether the game service can be normally used in the corresponding terminal 100.

As a result of the check, when the game service cannot be normally used in the corresponding terminal 100, the game server 200 blocks (or stops/stops) the execution of the game function in the corresponding terminal 100.

In addition, as a result of the confirmation, when the game service can be normally used in the terminal 100, the game server 200 allows the terminal 100 to perform the game function in single mode or multi mode. Various functions (or services) related to games are provided to the terminal 100 .

In addition, after one game mode is selected from among a plurality of game modes included in the game selection screen displayed on the terminal 100, a preset match search menu displayed on one side of the screen related to the selected game mode is selected. In this case, the game server 200 matches (or finds) a player to play the game with the corresponding terminal 100 among a plurality of other terminals (not shown) connected to the corresponding game server 200 .

In addition, when all players participating in the game complete game character selection, the game server 200 provides a loading window to the terminal 100 .

In addition, the game server 200 interworks with the game prediction information providing server 300 to determine (or check) whether a player corresponding to a corresponding game nickname is playing a game in real time. At this time, the game server 200 interworks with the game prediction information providing server 300 to first check whether the received game nickname is a member registered in the game server 200, and determines whether the game nickname is Only in the case of a member registered in the game server 200, it is possible to determine whether a player corresponding to a corresponding game nickname is playing a game in real time.

As a result of the determination (or as a result of the confirmation), if the corresponding game nickname is not a member registered in the game server 200, the match prediction information providing server 300 registers the corresponding game nickname in the game server 200 Information indicating that the user is not a member is transmitted to the terminal 100 .

In addition, as a result of the determination (or the confirmation result), if the corresponding game nickname is a member registered in the game server 200, the game server 200 interworks with the match prediction information providing server 300 to obtain the corresponding game nickname. It is determined (or confirmed) whether the player corresponding to the game nickname is playing a game in real time.

In addition, the game server 200 receives a game nickname transmitted from the match prediction information providing server 300 and real-time game state confirmation request information of the corresponding game nickname.

Also, the game server 200 determines whether the player corresponding to the received game nickname is currently playing a game.

As a result of the determination (or as a result of the confirmation), if the player corresponding to the corresponding game nickname is not playing the game in real time, the game server 200 sends information indicating that the player corresponding to the corresponding game nickname is not currently playing the game. It is transmitted to the match prediction information providing server 300 .

In addition, as a result of the determination (or the result of the confirmation), when the player corresponding to the corresponding game nickname is playing a game in real time, the game server 200 determines the game that the corresponding game nickname (or the player corresponding to the corresponding game nickname) is participating in. Check the

In addition, the game server 200 provides game nicknames of a plurality of players who are the same team as the corresponding game nicknames participating in the game in which the player corresponding to the identified game nickname is participating, and a plurality of other players who are the opposing team. Check the game nickname of each.

In addition, the game server 200 provides game character types, selected item types, and selected positions for each of a plurality of players participating in the same game to the match prediction information providing server 300 .

In this way, the game server 200 provides various types of games according to the subscription service to the terminal 100, manages comprehensive overall information, statistical information, etc. of a plurality of registered players, For the prediction information providing server 300, the match prediction information providing server 300 is provided with comprehensive overall information and statistical information for each member (or each player).

The match prediction information providing server 300 communicates with the terminal 100 and the game server 200 .

In addition, the match prediction information providing server 300 performs a membership sign-up procedure for users such as the terminal 100 and the like.

In addition, the match prediction information providing server 300 registers personal information related to the user of the terminal 100 and the like. At this time, the game prediction information providing server 300 may register (or manage) the corresponding personal information in a DB server (not shown).

In addition, the match prediction information providing server 300 performs a member management function for users such as the terminal 100 .

In addition, the match prediction information providing server 300 is a dedicated app and/or web application that provides a function to check overall overall information for each player, a function to check statistical information for each player, a function to check in-game information for each player, and a multi-search check function for each player. The site is provided to the terminal 100 and the like.

In addition, the match prediction information providing server 300 provides a bulletin board function for announcements, events, and the like.

In addition, the match prediction information providing server 300 interworks with the terminal 100 and the payment server, and checks the overall record information for each player provided by the match prediction information providing server 300 and the statistical information for each player. It performs other payment functions according to the execution of the subscription function, including functions, in-game information check function for each player, and multi-search check function for each player.

If the payment function fails, the match prediction information providing server 300 includes payment failure information (eg, payment date, payment amount, failure information (eg, including insufficient balance, limit exceeded, etc.)) (or information indicating a failed state) is provided to the terminal 100 .

In addition, the match prediction information providing server 300 transmits another payment function performance result provided from the payment server to the terminal 100 after another payment function with the terminal 100 is normally performed. Here, the result of performing the payment function includes other subscription period, other payment amount, and other payment date and time information.

In addition, the match prediction information providing server 300 maps (or matches/interlocks) the result of performing the payment function with the terminal 100 (or account information related to the terminal 100) and manages (or stores/ register).

In addition, according to the performance of the other subscription function, the match prediction information providing server 300 provides any information provided by the match prediction information providing server 300 through the dedicated app and/or website in the terminal 100. It provides various information and functions related to the player.

In addition, the match prediction information providing server 300 utilizes a plurality of training datasets collected in advance as continuous machine learning (or deep learning) data. Here, the input dataset for machine learning (or training dataset) is divided into a training set and a test set at a preset ratio (eg 7:3), and training and testing functions are performed. can be performed. In addition, the input data set for machine learning includes comprehensive overall information and statistical information for each player collected later, game character types, selected item types, and the like. In addition, the output dataset for machine learning is a part to be predicted, and includes a player's contribution to victory.

That is, the match prediction information providing server 300 sets (or determines / generates) the winning contribution of the corresponding player in relation to specific raw data for the statistical prediction model for each tier-rank-position through preset learning data. perform a learning function. At this time, the match prediction information providing server 300 stores raw data (including, for example, learning data, etc.) in parallel and distributedly, and unstructured data included in the stored raw data (or including learning data, etc.), Refine structured data and semi-structured data, perform preprocessing including classification as meta data, perform analysis including data mining on the preprocessed data, and at least one type Big data can be built by learning, training, and testing based on machine learning. At this time, at least one type of machine learning includes supervised learning, semi-supervised learning, unsupervised learning, reinforcement learning, and deep reinforcement learning. It may consist of any one or a combination of at least one of them.

In this way, the match prediction information providing server 300 performs a learning function on the capability model in the form of a neural network through the learning data and the like.

At this time, the match prediction information providing server 300 uses a decision tree through Xgboost for a plurality of variables used in the game in relation to the game provided by the corresponding game server 200. Set (or decide) the main variables that affect the pre-set number of wins/losses.

In this way, the match prediction information providing server 300 collects a plurality of variables used in the corresponding game, and classifies key variables affecting win/loss through machine learning based on the collected variables (or can be set/determined).

In addition, the game prediction information providing server 300 generates a plurality of decision trees for the plurality of variables and creates a plurality of models using the plurality of generated decision trees. At this time, the match prediction information providing server 300 does not select one model with the best predictive performance from among the plurality of models generated, but rather combines several models having poor performance to obtain a final model. generate Here, the statistical prediction model for each tier-rank-position is a tier-rank-top position statistical prediction model, a tier-rank-jungle position statistical prediction model, and a tier-rank-mid position statistical prediction model in order to calculate the win contribution for each position. model, tier-rank-wondeal position statistical prediction model, tier-rank-supporter position statistical prediction model, etc.

In addition, the match prediction information providing server 300 uses logistic regression to generate weights for each statistical prediction model for each tier-rank-position for the main variables selected for each statistical prediction model for each tier-rank-position using logistic regression. set (or decide) At this time, the logistic regression is a statistical prediction model that can be used when the dependent variable is categorical data, and since there are only two cases of win or loss in the case of the LOL game, it is appropriate to use the logistic regression. can

In this way, the game prediction information providing server 300 learns a plurality of game data in which whether or not the game is won or lost is learned, and then uses the game data that does not inform the win or loss as an input value and uses each variable to determine which team is in the wonder. By repeating the process (or learning) of guessing victory or defeat, variable selection and variable weight selection can be performed for each statistical prediction model for each tier-rank-position.

That is, the match prediction information providing server 300 puts variables selected from the Xgboost into the logistic regression model to perform a model performance test, and confirms which variables are statistically significant and accuracy.

In addition, the match prediction information providing server 300 performs this process N times to calculate (or select/determine) variables and weights suitable for match prediction. Here, N may be a natural number.

In addition, in the case of the LOL game, considering that the winning patterns of players are different for each tier and rank, which are preset game levels, and the importance of variables varies according to the positions of each player, the match prediction information providing server (300) reflects the characteristics of the corresponding role (LOL) game, a tier-rank-top position statistical prediction model, a tier-rank-top position statistical prediction model, a tier-rank-top position statistical prediction model, a tier-rank- It manages the mid-position statistical prediction model, tier-rank-adult position statistical prediction model, and tier-rank-supporter position statistical prediction model, respectively. Here, the tier is Iron, Bronze, Silver, Gold, Platinum, Diamond, Master, Grandmaster, and Challenger. (Challenger), etc., and may be set according to the administrator setting of the corresponding game server 200. In addition, the rank (Rank) includes steps 1 to 4, and may be set according to the administrator settings of the game server 200. In addition, in the case of the role (LOL), it is composed of five positions (including, for example, top, jungle, mid, one deal, support, etc.) according to the administrator settings of the corresponding game server 200. Here, the tower represents a position of a game character for attacking an opponent's champion and destroying a turret, and the jungle represents a position of a game character for catching preset monsters that are regenerated every specific time, helping allies fight opponents, and destroying turrets. , The mid represents a position of a game character for attacking an opponent champion and destroying a turret, and the one deal represents a position of a game character for attacking an opponent champion together with a supporter and destroying a turret, and the supporter represents a position of a game character for attacking an opponent champion together with a supporter and destroying a turret. indicates the position of the game character to attack and destroy the turret.

In addition, when the terminal 100 accesses a website provided by the match prediction information providing server 300 in order to check match prediction information related to a game in which the terminal 100 is participating, the match prediction information The providing server 300 provides the terminal 100 with an initial screen provided by a corresponding website. At this time, the initial screen includes a record search menu for checking the player's record, a multi-search menu for providing match prediction information of a team composed of a plurality of players, a champion analysis menu for providing information by game character, and game ranking information. It includes a ranking menu to provide, a community menu to share information between players, an e-sports menu to provide match prediction events for each major game, and a setting menu to set user information.

In addition, the match prediction information providing server 300 receives a game nickname, identification information of the terminal 100, etc. transmitted from the terminal 100.

In addition, the match prediction information providing server 300 interworks with the game server 200 to determine (or check) whether the corresponding game nickname is being played in real time. At this time, the match prediction information providing server 300 interworks with the game server 200, first checks whether the received game nickname is a member registered in the game server 200, and determines whether the game nickname is Only in the case of a member registered in the game server 200, it is possible to determine whether the corresponding game nickname is playing a game in real time.

As a result of the check, if the corresponding game nickname is not a member registered in the game server 200, the match prediction information providing server 300 determines that the corresponding game nickname is not a member registered in the game server 200 The indicated information is transmitted to the terminal 100.

As a result of the check, if the corresponding game nickname is a member registered in the game server 200, the game prediction information providing server 300 displays the received game nickname, real-time game status confirmation request information of the corresponding game nickname, etc. It is transmitted to the game server 200.

As a result of the determination (or as a result of the confirmation), when the corresponding game nickname is not in the game in real time, the match prediction information providing server 300 aggregates the results for each of the plurality of game nicknames managed by the match prediction information providing server 300 Comprehensive overall information and statistical information corresponding to the received game nickname among information and statistical information (or comprehensive overall information and statistical information for each player) (or comprehensive overall information and statistical information corresponding to a player related to the game nickname) Search (or check) the .

In addition, the match prediction information providing server 300 responds to request information for real-time game status confirmation of the previously transmitted game nickname, and the player corresponding to the corresponding game nickname transmitted from the game server 200 is not currently playing the game. Receives information indicating that, and searches (or confirms) comprehensive overall information and statistical information corresponding to the received game nickname among the comprehensive overall information and statistical information for each game nickname managed by the match prediction information providing server 300 )do.

In addition, the match prediction information providing server 300 transmits the searched (or confirmed) game nickname to the terminal 100 , such as overall performance information and statistical information. Here, the overall overall information includes information on players (or summoners) who played games with the same team (eg, including game nickname/summoner name, number of games, win/loss information, win rate, etc.), rank information (eg For example, including overall rank, solo rank, free 5:5 rank, etc.), recent rank win rate information (for example, including win/loss information for each game character, win rate, etc.), preferred position information (for example, including game win rate for each position, etc.) ), recent game win rate information (eg, record and win/loss information, win rate, etc.), information on the results of recently played games (eg game mode, game date, win/loss information, game time, game character type, used items) Type, rating, kill involvement, earned gold, CS (creep slain: number of minions and jungle monsters hunted), role IQ (Intelligence Quotient) (or an indicator calculated by player's understanding of the game + physical ability), by team information about the player, etc.), etc. In addition, the statistical information includes the number of games, win/loss information, rating, gold, CS, maximum number of kills, maximum number of deaths, average amount of damage inflicted, average amount of damage received, number of double kills, number of triple kills, number of quadra kills, and number of penta kills for each game character. Include etc.

In addition, as a result of the determination (or as a result of the confirmation), when the corresponding game nickname is playing in real time, the game prediction information providing server 300 interworks with the game server 200 to obtain the corresponding game nickname (or corresponding game nickname). The player corresponding to ) checks the game in which he is participating.

In addition, the game prediction information providing server 300 interworks with the game server 200 to provide a plurality of teams that are the same team as the game nickname in which the player corresponding to the identified game nickname is participating. The player's game nickname and the game nicknames of a plurality of other players who are opposing teams are respectively checked.

In addition, the game prediction information providing server 300 checks the game character type, selected item type, and selected position for each of a plurality of players participating in the same game.

In addition, the match prediction information providing server 300 is a match prediction information providing server 300, among the overall performance information and statistical information for each of a plurality of game nicknames managed by the match prediction information providing server 300, together with the same game in which the player corresponding to the game nickname is participating. Comprehensive record information and statistical information corresponding to game nicknames of a plurality of players who are participating in the same team and game nicknames of a plurality of other players who are the opposing team are retrieved (or checked), respectively.

In addition, the game prediction information providing server 300 provides the game nickname, the retrieved comprehensive overall information and statistical information for each player participating in the same game, the game character type for each player participating in the same game, the selected item type, and the selected item type. The learning function (or machine learning/artificial intelligence/deep learning) is performed by using input values including positions as input values of a pre-learned statistical prediction model for each tier-rank-position, and each position according to the result of performing the learning function Calculate (or calculate/generate) each player's contribution to victory, respectively. Here, the statistical prediction model for each tier-rank-position is a tier-rank-top position statistical prediction model, a tier-rank-jungle position statistical prediction model, and a tier-rank-mid position statistical prediction model in order to calculate the win contribution for each position. model, tier-rank-wondeal position statistical prediction model, tier-rank-supporter position statistical prediction model, etc.

That is, the match prediction information providing server 300 is configured according to the positions of a plurality of players included in our team and the opposing team participating in the same game, the overall record information and statistical information of the player for each position, the type of game character, A learning function is performed using information including the selected item type as input values of the statistical prediction model for each tier-rank-position, and the player's contribution to victory is calculated for each position according to the result of performing the learning function. Here, the statistical prediction model for each tier-rank-position is a preset number of variables that greatly affect win/loss among a plurality of variables that affect win/loss of the game provided by the corresponding game server 200 for each position. Variables may be set differently.

At this time, the match prediction information providing server 300 includes a preset number of variables set for each position, a preset weight corresponding to the corresponding variable, and a value corresponding to the variable among the searched player-specific overall record information (or information/data) as input values of the statistical prediction model for each tier-rank-position to perform a learning function, and calculate the player's contribution to victory for each position according to the result of performing the learning function.

In addition, the match prediction information providing server 300 generates match prediction information, which is in-game information, based on the contribution to victory of a plurality of players for each team calculated for each position. Here, the match prediction information (or match prediction information for each team) is expected win probability for each team, average number of KDA (kill/death/assist) per team, average damage per team, average gold per team, average ward installation per team, game information for each player who is a team member (For example, including game nickname, recent winning streak/loss streak information, game character type, win rate, score, level, record, rank, preferred position, recent play result, etc.) and the like. At this time, the ward may be an item that illuminates the field of view of the game character in LOL.

That is, since the match prediction information providing server 300 has different levels (or tiers) or ranks between a plurality of players constituting the same team, in order to standardize them, the levels and ranks of the plurality of players belonging to the same team Calculate (or calculate) the average value of (or the average value of the level for each team and the average value of the rank for each team) respectively.

In addition, the match prediction information providing server 300 applies the calculated average value of level for each team and average value of rank for each team to a plurality of players included in each team.

In addition, the match prediction information providing server 300 calculates the average value of the level for each team and the average value of the rank for each team and statistics according to the position among the statistical values for each level-rank-position managed by the match prediction information providing server 300 Check the values (including, for example, mean, standard deviation, etc.).

In addition, the match prediction information providing server 300 standardizes the player's contribution to victory for each position in the corresponding team based on the calculated contribution to victory of the player and the statistical value according to the position of the player (or standardized the contribution to victory). calculate/calculate).

In addition, the match prediction information providing server 300 calculates an average of standardized victory contributions (or an average of standardized victory contributions for each team) for each of a plurality of players within the same team.

In addition, the game prediction information providing server 300 checks a point where the average of the calculated team-specific standardized contributions to victory is located in a preset standard normal distribution table, and confirms the confirmed point as the expected win probability of the corresponding team. (or calculate/calculate)

In addition, the match prediction information providing server 300 determines the predicted victory probability for each team, average KDA (kill/death/assist number) for each team, average damage for each team, average gold for each team, average ward installation for each team, and games for each player who is a team member. The match prediction information including information and the like is generated.

In addition, the game prediction information providing server 300 transmits comprehensive game record information, statistical information, and generated game prediction information corresponding to the searched (or confirmed) corresponding game nickname to the terminal 100.

In addition, the match prediction information providing server 300 receives a plurality of game nicknames and identification information of the terminal 100 transmitted from the terminal 100 .

In addition, the match prediction information providing server 300 is the match prediction information providing server 300, among the plurality of game nickname-specific overall score information and statistical information, respectively corresponding to the plurality of game nicknames received. Comprehensive score information and search (or check) statistical information, respectively. Here, the overall overall information includes information on players (or summoners) who played games with the same team (eg, including game nickname/summoner name, number of games, win/loss information, win rate, etc.), rank information (eg For example, including overall rank, solo rank, free 5:5 rank, etc.), recent rank win rate information (for example, including win/loss information for each game character, win rate, etc.), preferred position information (for example, including game win rate for each position, etc.) ), recent game win rate information (eg, record and win/loss information, win rate, etc.), information on the results of recently played games (eg game mode, game date, win/loss information, game time, game character type, used items) Type, rating, kill involvement, earned gold, CS (creep slain: number of minions and jungle monsters hunted), role IQ (or an indicator calculated by player's understanding of the game + physical ability), information about players by team including etc.) and the like. In addition, the statistical information includes the number of games, win/loss information, rating, gold, CS, maximum number of kills, maximum number of deaths, average amount of damage inflicted, average amount of damage received, number of double kills, number of triple kills, number of quadra kills, and number of penta kills for each game character. Include etc.

In addition, the game prediction information providing server 300 performs a learning function (or machine learning/artificial intelligence/deep learning), and calculates (or calculates/generates) the winning contribution of each player for each position according to the result of the learning function. At this time, the match prediction information providing server 300 calculates the victory contribution for each position for the plurality of game nicknames, and calculates the highest value among the victory contributions for each position of the specific game nickname of the specific game nickname. You can also set it as a win contribution at that position.

That is, the match prediction information providing server 300 provides statistical prediction of tier-rank-top positions by providing information including overall player record information and statistical information, game character type, etc. according to a plurality of positions for each of the plurality of game nicknames. Model, Tier-Rank-Jungle Position Statistical Prediction Model, Tier-Rank-Mid Position Statistical Prediction Model, Tier-Rank-One Deal Position Statistical Prediction Model, Tier-Rank-Supporter Position Statistical Prediction Model, etc. Each is performed, and the player's contribution to victory is calculated for each position according to the learning function execution result. Here, the statistical prediction model for each tier-rank-position is a preset number of variables that greatly affect win/loss among a plurality of variables that affect win/loss of the game provided by the corresponding game server 200 for each position. Variables may be set differently.

In addition, the game prediction information providing server 300 matches and manages a specific position showing the highest contribution to victory among the contribution to victory of the player calculated for each position with the contribution to victory of the corresponding player.

At this time, the match prediction information providing server 300 includes a preset number of variables set for each position, a preset weight corresponding to the corresponding variable, and a value corresponding to the variable among the searched player-specific overall record information (or information/data) as input values of the statistical prediction model for each tier-rank-position to perform a learning function, and calculate the player's contribution to victory for each position according to the result of performing the learning function.

In addition, the game prediction information providing server 300 calculates (or calculates) an expected victory probability based on the victory contributions of a plurality of players calculated for each position.

That is, since the match prediction information providing server 300 is in a state where the levels (or tiers) or ranks of a plurality of players constituting the same team are different, in order to standardize them, the average value of the levels and ranks of the plurality of players (or The average value of the level of each team and the average value of the rank of each team) are calculated (or calculated) respectively.

In addition, the match prediction information providing server 300 applies the calculated average value of level for each team and average value of rank for each team to a plurality of players, respectively.

In addition, the match prediction information providing server 300 calculates the average value of the level for each team and the average value of the rank for each team and statistics according to the position among the statistical values for each level-rank-position managed by the match prediction information providing server 300 Check the values (including, for example, mean, standard deviation, etc.).

In addition, the match prediction information providing server 300 standardizes the player's contribution to victory for each position in the corresponding team based on the calculated contribution to victory of the player and the statistical value according to the position of the player (or standardized the contribution to victory). calculate/calculate).

In addition, the match prediction information providing server 300 calculates an average of standardized victory contributions (or an average of standardized victory contributions for each team) for each of a plurality of players within the same team.

In addition, the game prediction information providing server 300 checks a point where the average of the calculated team-specific standardized contributions to victory is located in a preset standard normal distribution table, and confirms the confirmed point as the expected win probability of the corresponding team. (or calculate/calculate)

In addition, the match prediction information providing server 300 transmits the calculated predicted winning probability, the recommended position information for each of the identified plurality of game nicknames, and the searched (or confirmed) summary information for each of the plurality of game nicknames to the terminal Send to (100). Here, the summary information (or game information) is a game nickname, recent winning streak/loss streak information, level, pick rate and win rate and IQ for each preferred game character, pick rate and win rate and IQ for each preferred position, and recent play results (eg For example, it includes win/loss information, KDA, game date information, and the like.

In addition, the recommended position information (or position distribution map) for each of the plurality of game nicknames uses overall performance information corresponding to the preset number of games recently played by each player (for example, 20 games), The probability of which position has been played the most among the three positions (eg, top, jungle, mid, remote, and support) is calculated, respectively, and the position with the highest probability is indicated.

In addition, the preferred position in the summary information represents the most selected position by the corresponding player. In addition, the preferred game character (or preferred champion) in the summary information corresponds to the number of other preset games (eg, 100 games) recently played by the corresponding player among a plurality of game characters provided by the corresponding game server 200. Using the overall overall information, it is displayed in order of game characters that have been played the most among the number of other games. At this time, the pick rate represents the probability that the corresponding player picks (or selects) the corresponding game character.

In this way, in the in-game state, the game information of a plurality of players in the team including the user of the terminal and the game information of the plurality of other players included in the opposing team are used as input values of the statistical prediction model to perform the learning function, and the learning function Match prediction information between teams may be provided according to the performance results.

In addition, in this way, match prediction information according to multi-search in the case of forming a team with a plurality of players selected by the player can be provided.

Hereinafter, a method for providing match prediction information according to the present invention will be described in detail with reference to FIGS. 1 to 18 .

2 and 3 are flowcharts illustrating a method of providing match prediction information according to a first embodiment of the present invention.

First, the terminal 100 executes a dedicated game app pre-installed in the corresponding terminal 100 and displays a game app execution result screen according to the execution of the dedicated game app. At this time, the game app execution result screen is a game start menu (or button/item) for playing a game according to the game mode, a home menu for moving to a home screen (or standby screen), and a strategic strategy for playing a game in an individual game. Team battle menu, Battle menu to play a tournament type game between teams, My information menu to check information such as proficiency, rank, and honor, information related to game characters (e.g. champions in LOL, skins) , emotes, runes, spells/spells, etc.) to check game character information menus, workshop menus to check information related to items you own, and your own to trade items according to the game character's level (or grade). Store menu, store menu for trading general items, cache menu for displaying information about a player's in-game cash (e.g. including Role Play at LOL, blue essence, etc.), player's in-game Level menu to display levels, friend list menu to display a list of one or more players with whom the player has friends, chat window menu to chat with other players, information related to missions (or quests) It includes a mission menu to check, a setting menu to set dedicated game app functions, and so on. Here, the terminal 100 is registered as a member of the game server 200 that provides the dedicated game app, and uses an ID and password according to membership registration, a barcode or QR code including the ID, and the like. When the game app is executed, a log-in procedure is performed to perform one or more functions (eg, game play function, item purchase function, etc.) of the dedicated game app.

In addition, when a preset game start menu is selected from the game app execution result screen displayed on the terminal 100, the terminal 100 is configured to play a game corresponding to a game mode selected by a player (or user). Displays the game selection screen. Here, the game selection screen includes a plurality of game modes provided by the game server 200 that manages the corresponding dedicated game app. At this time, the plurality of game modes include a play with a person provided by LOL or a play selection menu by artificial intelligence, a summoner's gorge mode (or menu / button / item) to play a game between teams, and all players Random total battle mode to play the game using randomly selected game characters (or champions), strategic team battle mode to play the game in individual battles, and other types of games provided as events (e.g., team members' votes) and an event mode for playing a game using a game character that has obtained the most votes by , etc.).

For example, the first terminal 100 executes a LOL app pre-installed in the first terminal, and displays a roll app execution result screen 400 as shown in FIG. 4 . At this time, the first terminal may be in a state in which the login procedure has been performed after executing the corresponding role app using the account information (eg, 1-1 ID and 1-1 password) of the first player of the first terminal. there is.

In addition, when the game start menu 410 is selected from the roll app execution result screen 400, as shown in FIG. 5, the first terminal displays a game selection screen corresponding to the selected game start menu 410. (500) is displayed (S210).

Thereafter, after one game mode is selected from among a plurality of game modes included in the game selection screen displayed on the terminal 100, a preset match search menu displayed on one side of the screen related to the selected game mode is selected. In this case, the terminal 100 matches a player to play with through the mediation of the game server 200 .

For example, when a summoner's gorge mode 510 is selected from among a plurality of game modes included in the game selection screen 500 displayed on the first terminal, as shown in FIG. 6, the first terminal The matchmaking screen 600 corresponding to the selected summoner's gorge mode 510 is displayed.

In addition, when the match match menu 610 is selected from the match match screen 600 displayed on the first terminal, as shown in FIG. In conjunction with, a player to play with the corresponding first player is matched, and a matching waiting screen 700 according to the matching is displayed (S220).

Then, when a plurality of players to play with the player of the terminal 100 are matched, the terminal 100 matches the player according to the player selection (or player/user touch/input/control) of the terminal 100 Determines whether or not to proceed with a game with multiple players.

In addition, as a result of the determination, when the game progress with the plurality of matched players is accepted (or when all game progress acceptance is determined for each matched plurality of players), the terminal 100 is the terminal 100 According to player selection (or player/user touch/input/control), a player of the corresponding terminal 100 selects a game character to be used. Here, the game character to be used by the player may correspond to a champion in the LOL. At this time, the terminal 100 interlocks with the game server 200 to prohibit the use of a specific game character that is difficult to play against in correspondence with the game character used by the player of the terminal 100, which is set in advance in conjunction with the game server 200. Among the plurality of game characters, one or more game characters that cannot be played by the players of the opposing team may be selected, and the one or more game characters that the players of the selected opposing team cannot play may be configured so that the players of the opposing team cannot select them. there is.

For example, when the 2nd to 5th players to play on the same team as the first player (or our team) and the 6th to 10th players to play on the opposing team are matched, as shown in FIG. , The first terminal determines (for example, accepts) whether to proceed with a game with the plurality of matched players according to the selection of the first player (810).

Also, as shown in FIG. 9 , the first terminal displays a game character selection screen 900 for the first player to select a game character (or champion) to play.

Also, the first terminal selects a first game character (for example, Karma) according to the selection of the first player from the game character selection screen 900 displayed on the first terminal (S230).

Thereafter, when all players participating in the game complete game character selection, the terminal 100 interworks with the game server 200 and displays a loading window. Here, the loading window targets all players who participated in the game, and includes information related to game characters selected by a plurality of players belonging to our team including the player of the terminal 100 (for example, in LOL). tier, rank, spell, rune, skin, level, etc.), and information related to game characters selected by multiple players belonging to the opposing team.

As such, the terminal 100 (or a player of the terminal 100) executes a dedicated game app installed in the terminal 100, and performs preliminary work for game play using a game character that the player wants to play. can complete.

For example, when the first to tenth players participating in the same game select the first to tenth game characters, respectively, as shown in FIG. A loading window 1000 including information on the 10th game character selected from the 1st game character to the 10th player is displayed (S240).

Thereafter, the terminal 100 accesses the website provided by the match prediction information providing server 300 in order to check the match prediction information related to the game in which the terminal 100 is participating, and the website provides Display the initial screen. At this time, the initial screen includes a record search menu for checking the player's record, a multi-search menu for providing match prediction information of a team composed of a plurality of players, a champion analysis menu for providing information by game character, and game ranking information. It includes a ranking menu to provide, a community menu to share information between players, an e-sports menu to provide match prediction events for each major game, and a setting menu to set user information. Here, the terminal 100 is registered as a member of the match prediction information providing server 300 that provides the website, and uses an ID and password according to membership registration, a barcode or QR code including the ID, and the like. When accessing the above website, a login procedure is performed to perform one or more functions of the website (e.g., a function to check overall record information for each player, a function to check statistical information for each player, a function to check in-game information for each player, and a function to check multi-search for each player). etc.) can be performed.

In addition, when a preset record search menu is selected on the initial screen displayed on the terminal 100, the terminal 100 checks (or searches) the record of a specific player that the player of the corresponding terminal 100 wants to check. Displays the total search screen for Here, the record search screen includes a game nickname input item for inputting the game nickname (or summoner name) of the player to check the record of a specific player (or match prediction information of a specific player currently participating in the game). . At this time, when accessing the initial screen, the terminal 100 may display a total search screen corresponding to the total search menu by default.

For example, the first terminal accesses a first website (eg https://wa.gg) that provides match prediction information, and as shown in FIG. 11, a default search screen (1100) ) is displayed. At this time, the first terminal performs a log-in procedure after accessing the first website using the account information (for example, ID 1-2 and password 1-2) of the first player of the first terminal. It may be (S250).

Thereafter, the terminal 100 displays a game nickname corresponding to the game nickname input item according to the player input (or player/user selection/touch/control) of the terminal 100 on the overall search screen displayed on the terminal 100 receive

In addition, the terminal 100 transmits the received game nickname and identification information of the terminal 100 to the match prediction information providing server 300 . Here, the identification information of the terminal 100 includes MDN, mobile IP, mobile MAC, Sim (subscriber identification module) card-specific information, serial number, and the like.

For example, the first terminal receives the first game nickname (eg, El Lim) of the first player according to the first player input on the total search screen 1100 displayed on the first terminal.

In addition, the first terminal transmits the received first game nickname (eg, El Lim) and identification information of the first terminal to the match prediction information providing server 300 (S260).

Thereafter, the match prediction information providing server 300 receives a game nickname, identification information of the terminal 100, etc. transmitted from the terminal 100.

In addition, the match prediction information providing server 300 interworks with the game server 200 to determine (or check) whether a player corresponding to a corresponding game nickname is playing a game in real time. At this time, the match prediction information providing server 300 interworks with the game server 200, first checks whether the received game nickname is a member registered in the game server 200, and determines whether the game nickname is Only in the case of a member registered in the game server 200, it is possible to determine whether a player corresponding to a corresponding game nickname is playing a game in real time.

That is, the match prediction information providing server 300 transmits the received game nickname and real-time game state confirmation request information of the corresponding game nickname to the game server 200 .

In addition, the game server 200 receives a game nickname transmitted from the match prediction information providing server 300 and real-time game state confirmation request information of the corresponding game nickname.

Also, the game server 200 determines whether the player corresponding to the received game nickname is currently playing a game.

For example, the match prediction information providing server 300 receives a first game nickname (eg, El Lim) transmitted from the first terminal, identification information of the first terminal, and the like.

In addition, the match prediction information providing server 300 transmits the received first game nickname (eg, El Lim), real-time game status check request information of the first game nickname, etc. to the LOL game server do.

In addition, the LOL game server receives a first game nickname (eg, El Lim) transmitted from the match prediction information providing server 300, real-time game status check request information of the first game nickname, and the like .

In addition, the LOL game server checks a game in which the first player corresponding to the received first game nickname is participating among a plurality of real-time games managed by the corresponding LOL game server (S270).

As a result of the determination (or as a result of the confirmation), when the player corresponding to the corresponding game nickname is not playing the game in real time, the match prediction information providing server 300 provides a plurality of games managed by the match prediction information providing server 300 Among the comprehensive overall information and statistical information for each nickname, comprehensive overall information and statistical information corresponding to the received game nickname are searched for (or checked).

That is, when the player corresponding to the game nickname is not currently playing a game, the game server 200 transmits information indicating that the player corresponding to the corresponding game nickname is not currently playing a game to the match prediction information providing server 300 ) is sent to

In addition, the match prediction information providing server 300 responds to request information for real-time game status confirmation of the previously transmitted game nickname, and the player corresponding to the corresponding game nickname transmitted from the game server 200 is not currently playing the game. Receives information indicating that, and searches (or confirms) comprehensive overall information and statistical information corresponding to the received game nickname among the comprehensive overall information and statistical information for each game nickname managed by the match prediction information providing server 300 )do.

In addition, the match prediction information providing server 300 transmits the searched (or confirmed) game nickname to the terminal 100 , such as overall performance information and statistical information. Here, the overall overall information includes information on players (or summoners) who played games with the same team (eg, including game nickname/summoner name, number of games, win/loss information, win rate, etc.), rank information (eg For example, including overall rank, solo rank, free 5:5 rank, etc.), recent rank win rate information (for example, including win/loss information for each game character, win rate, etc.), preferred position information (for example, including game win rate for each position, etc.) ), recent game win rate information (eg, record and win/loss information, win rate, etc.), information on the results of recently played games (eg game mode, game date, win/loss information, game time, game character type, used items) Type, rating, kill involvement, earned gold, CS (creep slain: number of minions and jungle monsters hunted), role IQ (or an indicator calculated by player's understanding of the game + physical ability), information about players by team including etc.) and the like. In addition, the statistical information includes the number of games, win/loss information, rating, gold, CS, maximum number of kills, maximum number of deaths, average amount of damage inflicted, average amount of damage received, number of double kills, number of triple kills, number of quadra kills, and number of penta kills for each game character. Include etc.

As an example, as a result of the check, when the first game nickname is not in a game in real time, the match prediction information providing server 300 displays the results for each of the plurality of game nicknames managed by the match prediction information providing server 300 Among the comprehensive information and statistical information, first overall overall information and first statistical information corresponding to the received first game nickname are searched.

In addition, the match prediction information providing server 300 transmits first overall overall information and first statistical information corresponding to the retrieved first game nickname to the first terminal (S280).

In addition, as a result of the determination (or the result of the confirmation), when a player corresponding to the corresponding game nickname is playing a game in real time, the match prediction information providing server 300 interworks with the game server 200 to obtain the corresponding game nickname ( or the player corresponding to the game nickname) is participating in the game.

In addition, the game prediction information providing server 300 interworks with the game server 200 to provide a plurality of teams that are the same team as the game nickname in which the player corresponding to the identified game nickname is participating. The player's game nickname and the game nicknames of a plurality of other players who are opposing teams are respectively checked.

In addition, the game prediction information providing server 300 checks the game character type, selected item type, and selected position for each of a plurality of players participating in the same game.

As an example, as a result of the check, when the first game nickname is in a game in real time, the match prediction information providing server 300 interworks with the LOL game server to provide a game corresponding to the first game nickname. The first game in which the first player is participating is checked.

In addition, the match prediction information providing server 300 interworks with the LOL game server, and the first game nickname of the first player participating in the first game to the fifth game nickname of the fifth player ( For example, Elim, cxzqweasd, Dabong Rammus, Gen G Winner, Summer Summer Fall Winter), the 6th game nickname of the opposing team player 6 or the 10th game nickname of the 10th player (for example, this was not lol, Gen G Peyz, Butterfly Seeing Flowers, Don’t Live Roughly, and Sonkywak Park) respectively.

In addition, the match prediction information providing server 300 interworks with the LOL game server to determine the first game character type to the tenth game character type of each of the first to tenth players participating in the first game. , the 1st item type to the 10th item type, the 1st position to the 10th position (for example, the positions of supporter, AD, mid, jungle and top in the order of the 1st to 5th players of the same team, the opponent team 1st player) In the order of the 6th player to the 10th player, supporter, AD, mid, jungle and top positions, etc.) are respectively confirmed (S290).

Thereafter, the match prediction information providing server 300 provides the match prediction information providing server 300 to play the same game in which the player corresponding to the corresponding game nickname is participating among the comprehensive overall information and statistical information for each game nickname managed by the match prediction information providing server 300 Comprehensive record information and statistical information corresponding to game nicknames of a plurality of players who are participating in the same team and game nicknames of a plurality of other players who are the opposing team are retrieved (or checked), respectively.

As an example, the game prediction information providing server 300 selects the first game nickname to the th game participating in the first game from among the overall performance information and statistical information for each of a plurality of game nicknames managed by the match prediction information providing server 300. The first to tenth overall overall information and the first to tenth statistical information respectively corresponding to the 10 game nicknames are searched (S300).

Thereafter, the match prediction information providing server 300 provides the game nickname, the retrieved comprehensive overall information and statistical information for each player participating in the same game, the game character type for each player participating in the same game, the selected item type, and the selected item type. The learning function (or machine learning/artificial intelligence/deep learning) is performed by using input values including positions as input values of a pre-learned statistical prediction model for each tier-rank-position, and each position according to the result of performing the learning function Calculate (or calculate/generate) each player's contribution to victory, respectively. Here, the statistical prediction model for each tier-rank-position is a tier-rank-top position statistical prediction model, a tier-rank-jungle position statistical prediction model, and a tier-rank-mid position statistical prediction model in order to calculate the win contribution for each position. model, tier-rank-wondeal position statistical prediction model, tier-rank-supporter position statistical prediction model, etc.

That is, the match prediction information providing server 300 is configured according to the positions of a plurality of players included in our team and the opposing team participating in the same game, the overall record information and statistical information of the player for each position, the type of game character, A learning function is performed using information including the selected item type as input values of the statistical prediction model for each tier-rank-position, and the player's contribution to victory is calculated for each position according to the result of performing the learning function. Here, the statistical prediction model for each tier-rank-position is a preset number of variables that greatly affect win/loss among a plurality of variables that affect win/loss of the game provided by the corresponding game server 200 for each position. Variables may be set differently.

At this time, the match prediction information providing server 300 includes a preset number of variables set for each position, a preset weight corresponding to the corresponding variable, and a value corresponding to the variable among the searched player-specific overall record information (or information/data) as input values of the statistical prediction model for each tier-rank-position to perform a learning function, and calculate the player's contribution to victory for each position according to the result of performing the learning function.

As an example, the match prediction information providing server 300 may provide first overall overall information, first statistical information, and first game character type related to a first player corresponding to a supporter position among a plurality of players participating in the first game. (eg, Karma), the first input value including the first item, the first position (eg, supporter), and the corresponding tier-rank-supporter position 5 variables (eg, DPM) preset in the statistical prediction model. , Gold Earned, CS, Total Assists, Total Deaths, etc.) and the preset weights (including, for example, 3.2156, 5.6367, 1.7259, 0.6174, -4.2591, etc.) in response to the five variables, the tier-rank-supporter position The learning function is performed using the input value of the statistical prediction model, and according to the learning function performance result, the first player's first win contribution (eg (3.2156 * first player's DPM) + (5.6367 * first player's Gold Earned) + (1.7259 * Player 1's CS) + (0.6174 * Player 1's Total Assists) + (-4.2591 * Player 1's Total Deaths) = 2.34).

In addition, the match prediction information providing server 300 may provide second overall performance information, second statistical information, and second game character type related to a second player corresponding to a one-deal position among a plurality of players participating in the first game ( For example, Barus), the second input value including the second item, the second position (for example, one deal), and the other five variables (for example, DPM) preset in the corresponding tier-rank-one deal position statistical prediction model. , Gold Earned, Total Kills, Total Assists, Total Deaths, etc.) and other preset weights corresponding to the other five variables (including, for example, 3.2156, 5.6367, 1.9654, 0.6174, -4.2591, etc.) - The learning function is performed using the input value of the one-deal position statistical prediction model, and the second player's second victory contribution (eg, 3.26 points) is calculated according to the learning function performance result.

In addition, the match prediction information providing server 300 may provide third overall overall information, third statistical information, and third game character type related to a third player corresponding to a mid position among a plurality of players participating in the first game ( eg rise), 3rd item, 3rd position (eg mid), etc. and another 5 variables preset in the corresponding tier-rank-mid position statistical prediction model and the corresponding another A learning function is performed using another preset weight corresponding to the five variables as an input value of the tier-rank-mid position statistical prediction model, and according to the result of performing the learning function, the third player's third victory contribution (eg For example, 1.19 points) is calculated.

In addition, the match prediction information providing server 300 provides fourth overall overall information, fourth statistical information, and fourth game character type related to a fourth player corresponding to a jungle position among a plurality of players participating in the first game ( eg Viego), the 4th input value including the 4th item, the 4th position (eg jungle), etc. and another 5 variables preset in the corresponding tier-rank-jungle position statistical prediction model and the corresponding or A learning function is performed using another preset weight corresponding to the other five variables as an input value of the tier-rank-jungle position statistical prediction model, and according to the result of performing the learning function, the fourth player's fourth victory contribution (eg For example, 5.78 points) is calculated.

In addition, the match prediction information providing server 300 may provide fifth overall performance information, fifth statistical information, and a fifth game character type related to a fifth player corresponding to a top position among a plurality of players participating in the first game ( For example, Lee Sin), the fifth item, the fifth position (eg top), etc., and another five variables preset in the statistical prediction model of the corresponding tier-rank-top position and the corresponding or A learning function is performed using another preset weight corresponding to the other 5 variables as an input value of the tier-rank-top position statistical prediction model, and according to the result of performing the learning function, the 5th player's 5th victory contribution (eg For example, 4.67 points) is calculated.

In addition, the match prediction information providing server 300 provides sixth overall overall information, sixth statistical information, and sixth game information related to a sixth player corresponding to a supporter position among a plurality of players participating in the first game, which is an opposing team. The 6th input value including the character type (eg Soraka), the 6th item, the 6th position (eg supporter), etc. and the 5 variables (eg. For example, including DPM, Gold Earned, CS, Total Assists, Total Deaths, etc.) and the preset weights (including, for example, 3.2156, 5.6367, 1.7259, 0.6174, -4.2591, etc.) corresponding to the five variables, the tier-rank - The learning function is performed using the input value of the supporter position statistical prediction model, and the 6th victory contribution of the 6th player (for example, 2.31 points) is calculated according to the learning function performance result.

In addition, the match prediction information providing server 300 provides seventh overall overall information, seventh statistical information, and seventh game character type related to a seventh player corresponding to a one-deal position among a plurality of players participating in the first game ( For example, Ashe), the 7th input value including the 7th item, the 7th position (for example, one deal), and the other five variables (for example, DPM) preset in the corresponding tier-rank-one deal position statistical prediction model. , Gold Earned, Total Kills, Total Assists, Total Deaths, etc.) and other preset weights corresponding to the other five variables (including, for example, 3.2156, 5.6367, 1.9654, 0.6174, -4.2591, etc.) - The learning function is performed using the input value of the one-deal position statistical prediction model, and the seventh victory contribution (eg, 3.24 points) of the seventh player is calculated according to the learning function performance result.

In addition, the match prediction information providing server 300 provides eighth overall overall information, eighth statistical information, and an eighth game character type related to an eighth player corresponding to a mid position among a plurality of players participating in the first game ( eg Talon), the 8th input value including the 8th item, the 8th position (eg mid), etc. and another 5 variables preset in the corresponding tier-rank-mid position statistical prediction model and the corresponding another A learning function is performed using another preset weight corresponding to the five variables as an input value of the tier-rank-mid position statistical prediction model, and according to the result of performing the learning function, the 8th player's 8th victory contribution (for example For example, 1.21 points) is calculated.

In addition, the match prediction information providing server 300 may provide ninth overall overall information, ninth statistical information, and ninth game character type related to a ninth player corresponding to a jungle position among a plurality of players participating in the first game ( For example, eco), a ninth input value including a ninth item, a ninth position (eg jungle), etc., and another five variables preset in the corresponding tier-rank-jungle position statistical prediction model and corresponding another A learning function is performed using another preset weight corresponding to the five variables as an input value of the tier-rank-jungle position statistical prediction model, and according to the result of performing the learning function, the ninth player's ninth victory contribution (eg For example, 5.81 points) is calculated.

In addition, the match prediction information providing server 300 provides 10th overall overall information, 10th statistical information, 10th game character type ( For example, Riven), the 10th input value including the 10th item, the 10th position (eg top), etc. and another 5 variables preset in the corresponding tier-rank-top position statistical prediction model and the corresponding another A learning function is performed using another preset weight corresponding to the five variables as an input value of the tier-rank-top position statistical prediction model, and according to the result of performing the learning function, the 10th player's 10th victory contribution (for example For example, 4.57 points) is calculated (S310).

Thereafter, the match prediction information providing server 300 generates match prediction information, which is in-game information, based on the contribution to victory of a plurality of players for each team calculated for each position. Here, the match prediction information (or match prediction information for each team) is expected win probability for each team, average number of KDA (kill/death/assist) per team, average damage per team, average gold per team, average ward installation per team, game information for each player who is a team member (For example, including game nickname, recent winning streak/loss streak information, game character type, win rate, score, level, record, rank, preferred position, recent play result, etc.) and the like. At this time, the ward may be an item that illuminates the field of view of the game character in LOL.

That is, since the match prediction information providing server 300 has different levels (or tiers) or ranks between a plurality of players constituting the same team, in order to standardize them, the levels and ranks of the plurality of players belonging to the same team Calculate (or calculate) the average value of (or the average value of the level for each team and the average value of the rank for each team) respectively.

In addition, the match prediction information providing server 300 applies the calculated average value of level for each team and average value of rank for each team to a plurality of players included in each team.

In addition, the match prediction information providing server 300 calculates the average value of the level for each team and the average value of the rank for each team and statistics according to the position among the statistical values for each level-rank-position managed by the match prediction information providing server 300 Check the values (including, for example, mean, standard deviation, etc.).

In addition, the match prediction information providing server 300 standardizes the player's contribution to victory for each position in the corresponding team based on the calculated contribution to victory of the player and the statistical value according to the position of the player (or standardized the contribution to victory). calculate/calculate).

In addition, the match prediction information providing server 300 calculates an average of standardized victory contributions (or an average of standardized victory contributions for each team) for each of a plurality of players within the same team.

In addition, the game prediction information providing server 300 checks a point where the average of the calculated team-specific standardized contributions to victory is located in a preset standard normal distribution table, and confirms the confirmed point as the expected win probability of the corresponding team. (or calculate/calculate)

In addition, the match prediction information providing server 300 determines the predicted victory probability for each team, average KDA (kill/death/assist number) for each team, average damage for each team, average gold for each team, average ward installation for each team, and games for each player who is a team member. The match prediction information including information and the like is generated.

For example, the match prediction information providing server 300 determines the first player's tier (eg 3) and rank (eg 2) of the team to which the first player belongs, and the second player's tier (eg 2). eg 2) and rank (eg 4), 3rd player's tier (eg 2) and rank (eg 2), 4th player's tier (eg 3) and rank (eg 3) and the tier average of the same team for the tier (eg 3) and rank (eg 1) of the fifth player (eg (3+2+2+3+3)/5 = 2.6) Calculate the rank average of the same team (e.g. (2+4+2+3+1)/5 = 2.4) respectively.

In addition, the match prediction information providing server 300 applies a tier value 3 obtained by rounding off the calculated tier average (eg, 2.6) to one decimal place to the tiers of the first to fifth players, , The rank value 2 obtained by rounding the calculated rank average (eg, 2.4) to one decimal place is applied to the ranks of the first to fifth players.

In addition, the match prediction information providing server 300 calculates the average value of the level for each team and the average value of the rank for each team among the statistical values for each level-rank-position managed by the match prediction information providing server 300, and the position supporter Check the first statistical value (e.g., the first mean of 2.14, the first standard deviation of 1.35).

In addition, the game prediction information providing server 300 determines the calculated first player's contribution to victory (eg, 2.34), a first statistical value (eg, 1 A first standardized victory contribution (for example, 0.58) of the first player at the corresponding supporter position is calculated based on an average of 2.14 and a first standard deviation of 1.35.

According to the above schemes, the match prediction information providing server 300 determines the second player's second standardized victory contribution (eg, 0.64) and the third player's third standardized victory contribution (eg, 0.37). ), the fourth player's fourth standardized win contribution (eg 0.82), the fifth player's fifth standardized win contribution (eg 0.71), the sixth player's sixth standardized win contribution ( eg 0.57), the 7th player's 7th standardized win contribution (eg 0.63), the 8th player's 8th standardized win contribution (eg 0.38), the 9th player's 9th normalized contribution The winning contribution (for example, 0.84) and the 10th standardized winning contribution (for example, 0.59) of the tenth player are calculated.

In addition, the match prediction information providing server 300 calculates the first average (for example, 0.624) of the first standardized victory contribution to the fifth standardized victory contribution of our team and the sixth standardized victory contribution of the opposing team. A second average (for example, 0.602) of the winning contribution to the 10th normalized winning contribution is calculated, respectively.

In addition, the match prediction information providing server 300 checks the points where the averages of the standardized victory contributions for each of the calculated teams in the preset standard normal distribution table are located, and determines the expected victory probability of our team (for example, 52.70%). ) and the expected winning probability of the opposing team (eg 47.30%).

In addition, the match prediction information providing server 300 includes our team's expected winning probability (eg 52.70%), our team's first average KDA (eg 2.28:1 5.4/4.8/5.6), our team's 1st average damage (eg 16,360), our team's 1st average gold (eg 10,540), our team's 1st average ward installation (eg 10), opponent team's expected win probability (eg 10,540) 47.30%), opponent's 2nd average KDA (eg 2.53:1 5.1/4.6/6.4), opponent's 2nd average damage (eg 12,667), opponent's 2nd average gold (eg 2.53:1 5.1/4.6/6.4) 9639), the opponent team's second average ward installation (for example, 14), the first match prediction information including the game information of the first player to the tenth player, etc. is generated (S320).

Thereafter, the game prediction information providing server 300 transmits the searched (or confirmed) game nickname corresponding to the overall result information, statistical information, and the generated game prediction information to the terminal 100.

As an example, the match prediction information providing server 300 transmits first overall performance information and first statistical information corresponding to the searched first game nickname, and the generated first match prediction information to the first terminal Do (S330).

Thereafter, the terminal 100 receives comprehensive overall information, statistical information, match prediction information, etc. transmitted from the game prediction information providing server 300 in response to the previously transmitted game nickname.

In addition, the terminal 100 displays the game nickname detailed information screen including the received overall score information by default. At this time, when the match prediction information is received by the terminal 100, the terminal 100 may display a preset on state icon on one side of the in-game information tab of the game nickname detailed information screen. Here, the game nickname detailed information screen includes a game nickname item of the corresponding player, a ranking item, a record update menu item, a recent update information item, a level and record information item in solo rank mode, a level in free 5:5 rank mode and Stats information item, Stats tab to check overall stats information, Stats tab to check statistical information, In-game information to check match predictions for the same team and the opposing team while participating in the game in real time including tabs, etc.

For example, the first terminal responds to the previously transmitted first game nickname and the like, and the first comprehensive information and first statistical information corresponding to the corresponding first game nickname transmitted from the match prediction information providing server 300, the The first game prediction information and the like are received.

In addition, as shown in FIG. 12, the first terminal displays the first comprehensive overall information 1200 corresponding to the received first game nickname on the general overall information tab in the game nickname detailed information screen (S340). .

In addition, when a statistics tab (or menu/item) in the game nickname detailed information screen displayed on the terminal 100 is selected, the terminal 100 determines information related to the received corresponding game nickname corresponding to the selected statistics tab. Display statistical information.

For example, when the statistics tab 1210 provided on one side of the screen displaying the first power synthesis information 1200 is selected, as shown in FIG. 13, the first control unit displays the received first game information. First statistical information 1300 corresponding to the nickname is displayed (S350).

In addition, when an in-game information tab in the game nickname detailed information screen displayed on the terminal 100 is selected, the terminal 100 displays a player corresponding to the received game nickname corresponding to the selected in-game information tab. Displays match prediction information related to the game being played. At this time, the in-game information tab may be activated only when the match prediction information is received from the terminal 100 .

For example, when the in-game information tab 1220 provided on one side of the screen displaying the first power synthesis information 1200 is selected, as shown in FIG. The first match prediction information 1400 of the same team and the opposing team including the corresponding first game nickname in the participating first game is displayed in real time (S360).

15 is a flowchart illustrating a method of providing match prediction information according to a second embodiment of the present invention.

First, the terminal 100 accesses a website provided by the match prediction information providing server 300 in order to check match prediction information related to a game in which the terminal 100 is participating, and provides information provided by the website. Display the initial screen. At this time, the initial screen includes a record search menu for checking the player's record, a multi-search menu for providing match prediction information of a team composed of a plurality of players, a champion analysis menu for providing information by game character, and game ranking information. It includes a ranking menu to provide, a community menu to share information between players, an e-sports menu to provide match prediction events for each major game, and a setting menu to set user information. Here, the terminal 100 is registered as a member of the match prediction information providing server 300 that provides the website, and uses an ID and password according to membership registration, a barcode or QR code including the ID, and the like. When accessing the above website, a login procedure is performed to perform one or more functions of the website (e.g., a function to check overall record information for each player, a function to check statistical information for each player, a function to check in-game information for each player, and a function to check multi-search for each player). etc.) can be performed.

In addition, when a preset multi-search menu is selected on the initial screen displayed on the terminal 100, the terminal 100 displays match prediction information of a team composed of a plurality of players selected by the player of the terminal 100. Displays the multi-search screen for confirmation. Here, the multi-search screen includes a plurality of game nickname input items for inputting game nicknames (or summoner names) of a plurality of players who will form the same team. At this time, when accessing the initial screen, the terminal 100 may display a total search screen corresponding to the total search menu by default.

For example, the first terminal accesses a first website (eg https://wa.gg) that provides match prediction information, and as shown in FIG. 16, the initial screen set by default, total search. Screen 1600 is displayed. At this time, the first terminal performs a log-in procedure after accessing the first website using the account information (for example, ID 1-2 and password 1-2) of the first player of the first terminal. can be

In addition, when the multi-search menu 1610 displayed on one side of the initial screen 1600 displayed on the first terminal is selected, as shown in FIG. 17, a multi-search menu corresponding to the selected multi-search menu 1610 is selected. The search screen 1700 is displayed (S1510).

Thereafter, the terminal 100 displays a plurality of game nickname input items corresponding to a plurality of game nickname input items according to player input (or player/user selection/touch/control) of the terminal 100 on the multi-search screen displayed on the terminal 100. Receives the game nickname of

In addition, the terminal 100 transmits the received plurality of game nicknames and identification information of the terminal 100 to the match prediction information providing server 300 . Here, the identification information of the terminal 100 includes MDN, mobile IP, mobile MAC, Sim (subscriber identification module) card-specific information, serial number, and the like.

For example, the first terminal 100 displays the 21st game nickname of the 21st player according to the first player input on the multi-search screen 1700 displayed on the first terminal (for example, a girl who handles guns), The 22nd game nickname of the 22nd player (eg, El Lim), the 23rd game nickname of the 23rd player (eg, Iggy Sipdang), the 24th game nickname of the 24th player (eg, Duburangto), and 25 Receives the 25th game nickname of the player (eg, Dabong Rammus).

In addition, the first terminal receives the received 21st game nickname (eg, gun handling girl), 22nd game nickname (eg, El Lim), 23rd game nickname (eg, Iggy Shipdang), th The 24th game nickname (eg Duburangtto), the 25th game nickname (eg Dabong Rammus), and the identification information of the first terminal are transmitted to the match prediction information providing server 300 (S1520).

Thereafter, the match prediction information providing server 300 receives a plurality of game nicknames and identification information of the terminal 100 transmitted from the terminal 100 .

In addition, the match prediction information providing server 300 is the match prediction information providing server 300, among the plurality of game nickname-specific overall score information and statistical information, respectively corresponding to the plurality of game nicknames received. Comprehensive score information and search (or check) statistical information, respectively. Here, the overall overall information includes information on players (or summoners) who played games with the same team (eg, including game nickname/summoner name, number of games, win/loss information, win rate, etc.), rank information (eg For example, including overall rank, solo rank, free 5:5 rank, etc.), recent rank win rate information (for example, including win/loss information for each game character, win rate, etc.), preferred position information (for example, including game win rate for each position, etc.) ), recent game win rate information (eg, record and win/loss information, win rate, etc.), information on the results of recently played games (eg game mode, game date, win/loss information, game time, game character type, used items) Type, rating, kill involvement, earned gold, CS (creep slain: number of minions and jungle monsters hunted), role IQ (or an indicator calculated by player's understanding of the game + physical ability), information about players by team including etc.) and the like. In addition, the statistical information includes the number of games, win/loss information, rating, gold, CS, maximum number of kills, maximum number of deaths, average amount of damage inflicted, average amount of damage received, number of double kills, number of triple kills, number of quadra kills, and number of penta kills for each game character. Include etc.

For example, the match prediction information providing server 300 transmits the 21st game nickname (eg, gun handling girl), the 22nd game nickname (eg, El Lim), and the 23rd game nickname transmitted from the first terminal. A nickname (eg, Iggy Shipdang), a 24th game nickname (eg, Duburangto), a 25th game nickname (eg, Dabong Rammus), and identification information of the first terminal are received.

In addition, the match prediction information providing server 300 provides a 21st overall score corresponding to the received 21st game nickname from among the match score prediction information providing server 300 and statistical information for each game nickname. information and 21st statistical information, 22nd overall overall information and 22nd statistical information corresponding to the received 22nd game nickname, 23rd overall overall information and 23rd statistical information corresponding to the received 23rd game nickname, the The 24th overall overall information and 24th statistical information corresponding to the received 24th game nickname and the 25th overall overall information and 25th statistical information corresponding to the received 25th game nickname are searched (S1530).

Thereafter, the match prediction information providing server 300 performs a learning function (or machine learning/artificial intelligence/deep learning), and calculates (or calculates/generates) the winning contribution of each player for each position according to the result of the learning function. At this time, the match prediction information providing server 300 calculates the victory contribution for each position for the plurality of game nicknames, and calculates the highest value among the victory contributions for each position of the specific game nickname of the specific game nickname. You can also set it as a win contribution at that position.

That is, the match prediction information providing server 300 provides statistical prediction of tier-rank-top positions by providing information including overall player record information and statistical information, game character type, etc. according to a plurality of positions for each of the plurality of game nicknames. Model, Tier-Rank-Jungle Position Statistical Prediction Model, Tier-Rank-Mid Position Statistical Prediction Model, Tier-Rank-One Deal Position Statistical Prediction Model, Tier-Rank-Supporter Position Statistical Prediction Model, etc. Each is performed, and the player's contribution to victory is calculated for each position according to the learning function execution result. Here, the statistical prediction model for each tier-rank-position is a preset number of variables that greatly affect win/loss among a plurality of variables that affect win/loss of the game provided by the corresponding game server 200 for each position. Variables may be set differently.

In addition, the game prediction information providing server 300 matches and manages a specific position showing the highest contribution to victory among the contribution to victory of the player calculated for each position with the contribution to victory of the corresponding player.

At this time, the match prediction information providing server 300 includes a preset number of variables set for each position, a preset weight corresponding to the corresponding variable, and a value corresponding to the variable among the searched player-specific overall record information (or information/data) as input values of the statistical prediction model for each tier-rank-position to perform a learning function, and calculate the player's contribution to victory for each position according to the result of performing the learning function.

For example, the match prediction information providing server 300 is configured to use a 21st input value including 21st overall overall information and 21st statistical information related to the 21st player, and a corresponding tier-rank-supporter position statistical prediction model. 5 preset variables (e.g. DPM, Gold Earned, CS, Total Assists, Total Deaths, etc.) and preset weights corresponding to those 5 variables (e.g. 3.2156, 5.6367, 1.7259, 0.6174, -4.2591, etc.) included) as the input value of the tier-rank-supporter position statistical prediction model, and performs the learning function, and according to the learning function performance result, the 21st player's 21-1 victory contribution (for example, (3.2156 * 1st player of DPM) + (5.6367 * Player 1's Gold Earned) + (1.7259 * Player 1's CS) + (0.6174 * Player 1's Total Assists) + (-4.2591 * Player 1's Total Deaths) = 1.21) yields

In addition, the match prediction information providing server 300 corresponds to the 21st input value for the 21st player, other 5 variables preset in the corresponding tier-rank-wondeal position statistical prediction model, and the other 5 variables. Then, the learning function is performed by using the preset other weights as the input values of the tier-rank-wondeal position statistical prediction model, and the 21st player's contribution to the 21st-2nd win (for example, 1.16 points) according to the result of performing the learning function yields

In addition, the match prediction information providing server 300 includes the 21st input value for the 21st player, another 5 variables preset in the corresponding tier-rank-mid position statistical prediction model, and the other 5 variables A learning function is performed using another preset weight corresponding to , as an input value of the tier-rank-mid position statistical prediction model, and according to the result of performing the learning function, the 21st player's 21-3 victory contribution (for example 2.19 points) is calculated.

In addition, the match prediction information providing server 300 includes the 21st input value for the 21st player, another 5 variables preset in the corresponding tier-rank-jungle position statistical prediction model, and the other 5 variables A learning function is performed using another preset weight corresponding to , as an input value of the tier-rank-jungle position statistical prediction model, and according to the result of performing the learning function, the 21st player's 21-4 victory contribution (for example, 1.81 points) is calculated.

In addition, the match prediction information providing server 300 includes the 21st input value for the 21st player, another 5 variables preset in the corresponding tier-rank-top position statistical prediction model, and the other 5 variables Correspondingly, a learning function is performed using another preset weight as an input value of the tier-rank-top position statistical prediction model, and according to the learning function performance result, the 21st player's 21st-5th victory contribution (for example, 1.67 points) is calculated.

In addition, the match prediction information providing server 300 determines the 21-3 victory contribution having the highest value among the calculated 21-1 victory contribution to 21-5 victory contribution for the 21st player (for example, 2.19 points) is confirmed, and the 21st victory contribution (eg, 2.19 points) corresponding to the 21-3 victory contribution and a related position (eg, mid position) are matched with the corresponding 21st player.

According to the above methods, the match prediction information providing server 300 determines the 22nd to 25th victory contributions (for example, 4.81, 5.13, 2.43, 3.99) and Each related position (eg, jungle, one deal, jungle, top) is matched (S1540).

Thereafter, the game prediction information providing server 300 calculates (or calculates) an expected victory probability based on the victory contributions of the plurality of players calculated for each position.

That is, since the match prediction information providing server 300 is in a state where the levels (or tiers) or ranks of a plurality of players constituting the same team are different, in order to standardize them, the average value of the levels and ranks of the plurality of players (or The average value of the level of each team and the average value of the rank of each team) are calculated (or calculated) respectively.

In addition, the match prediction information providing server 300 applies the calculated average value of level for each team and average value of rank for each team to a plurality of players, respectively.

In addition, the match prediction information providing server 300 calculates the average value of the level for each team and the average value of the rank for each team and statistics according to the position among the statistical values for each level-rank-position managed by the match prediction information providing server 300 Check the values (including, for example, mean, standard deviation, etc.).

In addition, the match prediction information providing server 300 standardizes the player's contribution to victory for each position in the corresponding team based on the calculated contribution to victory of the player and the statistical value according to the position of the player (or standardized the contribution to victory). calculate/calculate).

In addition, the match prediction information providing server 300 calculates an average of standardized victory contributions (or an average of standardized victory contributions for each team) for each of a plurality of players within the same team.

In addition, the game prediction information providing server 300 checks a point where the average of the calculated team-specific standardized contributions to victory is located in a preset standard normal distribution table, and confirms the confirmed point as the expected win probability of the corresponding team. (or calculate/calculate).

For example, the game prediction information providing server 300 determines the tier (eg 3) and rank (eg 3) of the 21st player of the team to which the 21st player belongs, and the tier (eg 22nd player) of the 22nd player. 3) and rank (eg 3), the 23rd player's tier (eg 2) and rank (eg 2), the 24th player's tier (eg 2) and rank (eg 3) and the tier average of the same team (eg (3+2+2+3+3)/5 = 2.6) for the 25th player's tier (eg 3) and rank (eg 2). Calculate the rank average (eg (3+3+2+3+2)/5 = 2.6) of each.

In addition, the match prediction information providing server 300 applies a tier value 3 obtained by rounding off the calculated tier average (for example, 2.6) to one decimal place to the tiers of the 21st to 25th players, , The rank value 3 obtained by rounding the calculated rank average (for example, 2.6) to one decimal place is applied to the ranks of the 21st to 25th players.

In addition, the match prediction information providing server 300 calculates the average value of the level for each team and the average value of the rank for each team among the statistical values for each level-rank-position managed by the match prediction information providing server 300, and the position mid Check the 21st statistical value (for example, the 21st mean 2.41, the 21st standard deviation 1.31).

In addition, the game prediction information providing server 300 calculates the 21st victory contribution of the 21st player (for example, 2.19) and the 21st statistical value (for example, 21st A 21 st standardized victory contribution (for example, 0.31) of the 21 th player at the mid position is calculated based on an average of 2.41 and a 21 st standard deviation of 1.31).

According to the above methods, the match prediction information providing server 300 determines the 22nd standardized victory contribution of the 22nd player (eg 0.61) and the 23rd standardized victory contribution of the 23rd player (eg 0.66 ), the 24th standardized victory contribution (eg 0.41) of the 24th player, and the 25th standardized victory contribution (eg 0.54) of the 25th player.

Also, the match prediction information providing server 300 calculates a 21st average (eg, 0.506) of the 21st standardized contribution to victory to the 25th standardized contribution to victory.

In addition, the match prediction information providing server 300 checks the point where the average of the calculated standardized victory contributions in the preset standard normal distribution table is located, and the team composed of the 21st game nickname to the 25th game nickname Check the expected win probability (for example, 41.2%) (S1550).

Thereafter, the game prediction information providing server 300 transmits the calculated expected winning probability, the recommended position information for each of the identified plurality of game nicknames, and the summary information for each of the searched (or confirmed) plurality of game nicknames to the terminal Send to (100). Here, the summary information (or game information) is a game nickname, recent winning streak/loss streak information, level, pick rate and win rate and IQ for each preferred game character, pick rate and win rate and IQ for each preferred position, and recent play results (e.g., win and loss information) , KDA, and game date information.

For example, the match prediction information providing server 300 calculates the calculated expected win probability (eg 41.2%), the confirmed position distribution for each play (eg 21st player mid, 22nd player jungle, 23rd player mid) player wondeal, 24th player jungle, 25th player top), summary information of the 21st player to summary information of the 25th player, etc. are transmitted to the first terminal (S1560).

Thereafter, the terminal 100 receives the expected winning probability, recommended position information for each of the plurality of game nicknames, and the plurality of game nicknames transmitted from the match prediction information providing server 300 in response to the plurality of game nicknames transmitted previously. receive summary information, etc.

In addition, the terminal 100 displays (or outputs) the received expected win probability, recommended position information for each of the plurality of game nicknames, and summary information for each of the plurality of game nicknames.

For example, the first terminal transmits the predicted win probability (for example, 41.2%) transmitted from the match prediction information providing server 300 in response to the previously transmitted 21st to 25th game nicknames, and the play-by-play Position distribution (for example, the 21st player mid, the 22nd player jungle, the 23rd player one deal, the 24th player jungle, the 25th player top), the summary information of the 21st player to the 25th player, etc. are received. do.

In addition, as shown in FIG. 18, the first terminal receives the expected win probability (for example, 41.2%), the position distribution for each play (for example, the 21st player mid, the 22nd player jungle, 23 player one deal, 24th player jungle, 25th player top), summary information of the 21st player to summary information of the 25th player, etc. (1800) are displayed (S1570).

As described above, in an embodiment of the present invention, game information of a plurality of players in a team including a user of a terminal and game information of a plurality of other players included in an opposing team are used as input values of a statistical prediction model in an in-game state. to perform a learning function, and to provide match prediction information between teams according to the result of performing the learning function, to check whether the game is won or lost according to a combination of multiple players constituting the same team together, and to increase the interest and interest of the players. This can increase your concentration on the game.

In addition, as described above, the embodiment of the present invention provides match prediction information according to multi-search when forming a team with a plurality of players selected by the player, and a virtual team combination between the plurality of players before game play. As the optimal team members are formed according to the above, the player's interest can be increased.

The foregoing may be modified and modified by those skilled in the art without departing from the essential characteristics of the present invention. Therefore, the embodiments disclosed in the present invention are not intended to limit the technical idea of the present invention, but to explain, and the scope of the technical idea of the present invention is not limited by these embodiments. The protection scope of the present invention should be construed according to the claims below, and all technical ideas within the equivalent range should be construed as being included in the scope of the present invention.

10: match prediction information providing system 100: terminal
200: game server 300: match prediction information providing server

Claims (10)

When a player corresponding to a game nickname transmitted from a match prediction information providing server is in a game in real time, checking a game in which the player corresponding to the game nickname is participating, and the game nickname participating in the identified game together Check the game nicknames of a plurality of players who are the same team and the game nicknames of the other players who are the opposing team, respectively, and check the game character type, selected item type, and selected position for each of the plurality of players participating in the game, respectively a game server that provides game prediction information to the server; and
Among the overall performance information and statistical information for each of the plurality of game nicknames managed by the game prediction information providing server, the game nicknames of the plurality of players of the same team as the game nicknames and the game nicknames of the plurality of other players of the opposing team Comprehensive information and statistical information are retrieved, respectively, and overall comprehensive information and statistical information corresponding to the searched game nickname for each player participating in the game, game character type for each player, input value including selected item type and selected position By performing a learning function based on the result of performing the learning function, the contribution to victory of each player for each position is calculated, and based on the contribution to victory of a plurality of players for each team calculated for each position, match prediction information is generated, The match prediction information providing system including the match prediction information providing server which transmits the match prediction information providing server, which transmits the match prediction information, the generated match prediction information, and overall performance information of the player corresponding to the searched game nickname to a terminal. According to claim 1,
The match prediction information,
Provision of match prediction information comprising at least one of expected win probability per team, average KDA (kill/death/assist) per team, average damage per team, average gold per team, average ward installation per team, and game information per player who is a member of the team system. checking, by a match prediction information providing server, a game in which the player corresponding to the game nickname is participating, when the player corresponding to the game nickname transmitted from the terminal is playing a game in real time, in conjunction with the game server;
Game nicknames of a plurality of players who are on the same team as the game nickname participating in the confirmed game together with the game server by the match prediction information providing server, and game nicknames of a plurality of other players who are the opposing team checking each, and checking the game character type, the selected item type, and the selected position for each of the plurality of players participating in the game;
Game nicknames of a plurality of players who are on the same team as the game nickname and other plurality of opponent teams among the overall performance information and statistical information for each of a plurality of game nicknames managed by the game prediction information providing server by the match prediction information providing server Retrieving comprehensive overall information and statistical information corresponding to the game nickname of the player of the above;
By the match prediction information providing server, an input value including overall performance information and statistical information corresponding to the searched game nickname for each player participating in the game, game character type for each of a plurality of players, selected item type, and selected position Performing a learning function based on the basis, and calculating the victory contribution of each player for each position based on the result of performing the learning function;
generating, by the match prediction information providing server, match prediction information based on contributions to victory of a plurality of players for each team calculated for each position; and
and transmitting, by the match prediction information providing server, comprehensive game record information, statistical information, and the generated match prediction information of the player corresponding to the searched game nickname to the terminal. According to claim 3,
In the step of calculating the victory contribution of each player for each position,
According to the position of each player included in our team and the opposing team participating in the same game, the information including the player's overall record and statistical information, game character type, and selected item type for each position is displayed as Tier-Rank-Position A method for providing match prediction information, characterized in that a learning function is performed using an input value of each statistical prediction model, and a player's contribution to victory is calculated for each position according to a result of performing the learning function. According to claim 3,
In the step of calculating the victory contribution of each player for each position,
Information including a preset number of variables set for each position, a preset weight corresponding to the variables, and a value corresponding to the variable among the searched player-specific overall record information is used as a statistical prediction model for each tier-rank-position. A method for providing match prediction information, characterized in that a learning function is performed using an input value, and a player's contribution to victory is calculated for each position according to a result of performing the learning function. According to claim 3,
receiving, by the terminal, comprehensive overall information, statistical information, and match prediction information of a player corresponding to the game nickname transmitted from the match prediction information providing server;
displaying, by the terminal, a game nickname detailed information screen including the received comprehensive information by default; and
When a statistics tab in a game nickname detailed information screen displayed on the terminal is selected, displaying, by the terminal, statistical information related to the received game nickname corresponding to the selected statistics tab. How to provide match prediction information. According to claim 3,
receiving, by the terminal, comprehensive overall information, statistical information, and match prediction information of a player corresponding to the game nickname transmitted from the match prediction information providing server; and
and displaying match prediction information related to a game in which the game nickname is participating when an in-game information tab in the game nickname detailed information screen displayed on the terminal is selected. A plurality of game nicknames transmitted from the terminal and identification information of the terminal are received, and among the plurality of game nicknames and statistical information managed by the game prediction information providing server, the game totals corresponding to the received plurality of game nicknames Information and statistical information are retrieved, and a learning function is performed based on input values including overall overall information and statistical information for each of the retrieved game nicknames, and the player's contribution to victory for each position based on the result of performing the learning function , calculates an expected win probability based on the contribution to victory of a plurality of players calculated for each position, and calculates the expected win probability, recommended position information for each plurality of game nicknames, and summary for each of the plurality of game nicknames retrieved. The match prediction information providing server for transmitting information to the terminal; and
The match prediction information providing system including the terminal displaying the expected win probability transmitted from the match prediction information providing server, recommended position information for each of a plurality of game nicknames, and summary information for each of the plurality of game nicknames. According to claim 8,
The summary information is
A system for providing match prediction information, characterized in that it includes at least one of a game nickname, recent winning/losing streak information, level, pick rate, win rate and IQ for each preferred game character, pick rate and win rate and IQ for each preferred position, and recent play results. Receiving, by a match prediction information providing server, a plurality of game nicknames transmitted from a terminal and identification information of the terminal;
By the match prediction information providing server, the match prediction information providing server retrieves comprehensive overall information and statistical information corresponding to the received plurality of game nicknames from among comprehensive overall information and statistical information for each of the plurality of game nicknames managed by the match prediction information providing server. doing;
The match prediction information providing server performs a learning function based on input values including overall overall information and statistical information for each of the retrieved game nicknames, and the player's contribution to victory for each position based on the result of performing the learning function Calculating each;
calculating, by the game prediction information providing server, an expected victory probability based on the contributions to victory of a plurality of players calculated for each position;
transmitting, by the match prediction information providing server, the calculated expected winning probability, recommended position information for each of a plurality of game nicknames, and summary information for each of the plurality of game nicknames retrieved to the terminal; and
and displaying, by the terminal, an expected win probability transmitted from the match prediction information providing server, recommended position information for each of a plurality of game nicknames, and summary information for each of the plurality of game nicknames.

Images