Information display method, device and system for Go game content

Abstract

An information display method for Go game content includes: displaying a video of a Go game between a first player and a second player on a screen of a user terminal based on streaming transmission to the user terminal in response to a request of the user terminal; displaying a first selection means capable of selecting a first information display mode on one side of the screen, wherein the first information display mode is used for displaying first information related to the Go game; if the first information display mode is selected, calculating a final win rate predicted value, namely a win rate value, of at least one player on the basis of an artificial intelligence engine whenever chessmen fall in the go game proceeding process; information relating to the calculated winning rate value is displayed as first information on a screen on which the Go game is displayed in real time. Therefore, various information related to the Go game based on artificial intelligence engine analysis can be provided.

Description

Technical Field

[0001] This invention relates to a method, apparatus, and system for displaying information about Go game content, and more specifically, to a method, apparatus, and system for displaying information about Go game content that can provide users with useful information related to Go game content based on artificial intelligence. Background Technology

[0002] Artificial intelligence (AI), generally speaking, is a branch of computer science that aims to artificially replicate human learning, reasoning, and perception abilities. In the field of information engineering, it can be considered a fundamental technology. In other words, AI is a computer system with intelligent capabilities; it is the technology that artificially implements human intelligence on machines and other devices.

[0003] Although the concept of artificial intelligence was proposed and developed in the early years, it remained at the stage of conceptual implementation or simple application for a long time. However, with the continuous development of deep learning technology and big data processing technology, its maturity has been significantly improved in recent years.

[0004] On the other hand, Go, a two-player game where players take turns placing black and white stones on a square board to compete for territory, is both a zero-sum game and a form of intellectual sport, enjoying immense popularity worldwide. Go players can place their stones on 361 intersections on a board formed by 19 vertical and 19 horizontal lines. The goal of Go is to create more space than one's opponent with one's own stones. While the rules of Go are simple, they require extremely profound strategic thinking.

[0005] Compared to board games like chess, Go presents a far greater number of possible game positions, making traditional AI techniques like brute-force search difficult to apply effectively. Therefore, computers were long considered unlikely to defeat human players. However, after Google acquired the British startup DeepMind in 2014, the development of AlphaGo, an AI-based game, entered a substantial phase. The release of AlphaGo Zero in 2017, which achieved overwhelming victories against humans, led to AlphaGo being hailed as ushering in a new era of AI. Simultaneously, the Go community widely believed that AlphaGo's innovative strategies and game philosophy, which challenged traditional understanding, were poised to revolutionize the millennia-old Go paradigm.

[0006] On the other hand, while Go games can provide players with information that helps them progress or achieve victory, traditional methods only offer fragmented information, making it difficult to provide information that can actually improve win rates or enhance skill. Furthermore, offline games or broadcast matches offer virtually no relevant information. Artificial intelligence-based Go systems can calculate various types of information; if they can provide appropriate information during a Go game, they can offer a practical and highly satisfying service to both players and viewers. Summary of the Invention

[0007] The problem the invention aims to solve

[0008] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method, device and system for displaying Go game content information that can provide the information required for Go game content based on artificial intelligence.

[0009] Problem Solving Methods

[0010] To achieve the above objectives, one aspect of the present invention provides a method for displaying information about Go game content. The method, executed by a server, includes: responding to a request from a user terminal, displaying a video of a Go game between a first player and a second player on the screen of the user terminal based on streaming transmission to the user terminal; displaying a first selection means on one side of the screen to select a first information display mode, the first information display mode being used to display first information related to the Go game; if the first information display mode is selected, then during the Go game, whenever a move is made, calculating the final win rate prediction value, i.e., the win rate value, of at least one player based on an artificial intelligence engine; and displaying information related to the calculated win rate value as the first information in real time on the screen displaying the Go game.

[0011] The real-time display step includes: after a move is made, superimposing the predicted final winning probability value (i.e., the win rate value) of the player who made the move, calculated for the move, onto the point on the chessboard where the move occurred; and if the next move occurs, removing the win rate value of the previous move that has already been displayed.

[0012] The real-time display step includes: after a move is made, comparing the win rate value of the player who made the move, calculated for the move, with the win rate value before the move; and based on the comparison of the win rate values, displaying on the screen whether the win rate after the move has increased or decreased compared to the win rate before the move, so that the user can identify it.

[0013] The real-time display step includes: after a move is made, comparing the win rate of the player who made the move with the win rate of the opponent; and based on the comparison of the win rate, displaying on the screen whether the player who made the move has a higher probability of winning or a higher probability of losing, so that the user can identify the outcome.

[0014] The method for displaying information about the Go game content further includes: displaying a second selection means on the other side of the screen, wherein the second information display mode is used to display second information related to the Go game; if the second information display mode is selected, then setting an analysis target player between the first player and the second player; before the set analysis target player makes a move, calculating the expected winning probability, i.e., the win rate value, corresponding to each possible move point based on an artificial intelligence engine; among the calculated win rate values, detecting a predetermined number of possible move points with the highest win rate value when the analysis target player makes a move; and displaying the detected possible move points as recommended move points on the Go game board so that the user can identify them.

[0015] The method for displaying information about the Go game content involves displaying the expected winning probability (i.e., the win rate value) in relation to the recommended move point when a move is made, so as to facilitate identification.

[0016] The recommended move points are multiple. The method for displaying information about the Go game content distinguishes and displays the relative levels of the win rates between the recommended move points based on the win rate value corresponding to each recommended move point, so as to facilitate identification.

[0017] The method for displaying information about the Go game content further includes: setting a start time and an end time for reviewing the Go game based on the user interface; and when a review request signal is received, displaying a review screen from the start time to the end time, and displaying the changing win rate value for each move.

[0018] On the other hand, to achieve the above objectives, one aspect of the present invention provides an information display device for Go game content. The Go game content information display device includes: a game display unit, which, in response to a request from a user terminal, displays a video of a Go game between a first player and a second player on the screen of the user terminal based on streaming transmission to the user terminal; and a processor unit, which displays a first selection means for selecting a first information display mode on one side of the screen, the first information display mode being used to display first information related to the Go game; if the first information display mode is selected, during the Go game, whenever a move is made, a final win rate prediction value, i.e., a win rate value, is calculated based on an artificial intelligence engine for at least one player; and information related to the calculated win rate value is displayed in real time on the screen displaying the Go game as the first information.

[0019] After a move is made, the processor calculates the final winning probability prediction value (i.e., the win rate value) for the player who made the move and displays it superimposed on the point where the move occurred on the chessboard; if the next move occurs, the win rate value of the previous move that has been displayed is removed.

[0020] After a move is made, the processor compares the win rate calculated for the player who made the move with the win rate before the move. Based on the comparison, it displays on the screen whether the win rate after the move has increased or decreased compared to the win rate before the move, so that the user can identify the difference.

[0021] After a move is made, the processor compares the win rate of the player who made the move with the win rate of the opponent. Based on the comparison of the win rates, it displays on the screen whether the player who made the move has a higher probability of winning or a higher probability of losing, so that the user can make an identification.

[0022] The processor displays a second selection method on the other side of the screen, which allows selection of a second information display mode. The second information display mode is used to display second information related to the Go game. If the second information display mode is selected, an analysis target player is set between the first player and the second player. Before the set analysis target player makes a move, the expected winning probability, i.e., the win rate value, is calculated based on an artificial intelligence engine when a move is made at each possible move point. Among the calculated win rate values, a predetermined number of possible move points with the highest win rate value are detected when the analysis target player makes a move. The detected possible move points are displayed as recommended move points on the Go game board so that the user can identify them.

[0023] The information display device for the Go game content displays the expected winning probability, i.e., the win rate value, in relation to the recommended move point when the move is made, so as to facilitate identification.

[0024] The recommended move points are multiple. The Go game content information display device distinguishes and displays the relative high and low win rates between the recommended move points according to the win rate value corresponding to each recommended move point for identification.

[0025] The processor unit, based on the user interface, sets the start and end times for reviewing the Go game; when a review request signal is received, it displays the review screen from the start to the end time, and shows the changing win rate value for each move.

[0026] On the other hand, to achieve the above objectives, one aspect of the present invention provides an information display system for Go game content. The information display system for Go game content includes: an artificial intelligence engine; and a device unit that, in response to a request from a user terminal, displays a video of a Go game between a first player and a second player on the screen of the user terminal based on streaming transmission to the user terminal; displays a first selection means for selecting a first information display mode on one side of the screen, the first information display mode being used to display first information related to the Go game; if the first information display mode is selected, during the Go game, whenever a move is made, the final win rate prediction value, i.e., the win rate value, of at least one player is calculated based on the artificial intelligence engine; and information related to the calculated win rate value is displayed in real time on the screen displaying the Go game as the first information.

[0027] After a move is made, the device calculates the final winning probability prediction value (i.e., the win rate value) for the player who made the move and displays it superimposed on the point where the move occurred on the chessboard; if the next move occurs, the win rate value of the previous move that has been displayed is removed.

[0028] After a move is made, the device compares the win rate calculated for the player who made the move with the win rate before the move. Based on the comparison of the win rate, the device displays on the screen whether the win rate after the move has increased or decreased compared to the win rate before the move, so that the user can identify the win rate.

[0029] After a move is made, the device compares the win rate of the player who made the move with the win rate of the opponent. Based on the comparison of the win rates, it displays on the screen whether the player who made the move has a higher probability of winning or a higher probability of losing, so that the user can make an identification.

[0030] The device displays a second selection method on the other side of the screen, allowing selection of a second information display mode. This second information display mode displays second information related to the Go game. If the second information display mode is selected, an analysis target player is set between the first player and the second player. Before the set analysis target player makes a move, an artificial intelligence engine calculates the expected winning probability (win rate) for each possible move. Among the calculated win rate values, a predetermined number of possible moves with the highest win rate are detected. The detected possible moves are displayed as recommended moves on the Go game board for user identification.

[0031] The Go game information display system displays the expected winning probability (i.e., the win rate value) in relation to the recommended move point when the move is made, so as to facilitate identification.

[0032] The recommended move points are multiple. The Go game content information display system distinguishes and displays the relative high and low win rates between the recommended move points based on the win rate value corresponding to each recommended move point for identification.

[0033] The device, based on the user interface, sets the start and end times for reviewing the Go game; when a review request signal is received, it displays the review screen from the start to the end time, and shows the changing win rate value for each move.

[0034] On the other hand, to achieve the above objectives, one aspect of the present invention provides a calculator program stored in a medium to a computer, which, in response to a request from a user terminal, displays a video of a Go game between a first player and a second player on the screen of the user terminal based on streaming transmission to the user terminal; displays a first selection means for selecting a first information display mode on one side of the screen, the first information display mode being used to display first information related to the Go game; if the first information display mode is selected, during the Go game, whenever a move is made, calculates the final win rate prediction value, i.e., the win rate value, of at least one player based on an artificial intelligence engine; and displays information related to the calculated win rate value as the first information on the screen displaying the Go game in real time.

[0035] Invention Effects

[0036] As described above, according to the present invention, when users watch Go game content provided by television, online live streaming, OTT, etc., they can conveniently obtain the information required for Go games generated by an artificial intelligence engine. Attached Figure Description

[0037] Figure 1 A system structure block diagram for implementing the method according to a preferred first embodiment of the present invention.

[0038] Figure 2 for Figure 1 The diagram shows a detailed structural block diagram of the first server.

[0039] Figure 3 For illustrative purposes Figure 2 The flowchart shown illustrates the operation of the first server, representing the process of an online Go game method based on artificial intelligence according to a preferred embodiment of the present invention.

[0040] Figure 4 An example diagram illustrating the win probability value corresponding to a possible move point calculated by the first information display processing unit of the first server based on the first artificial intelligence engine.

[0041] Figure 5 This is an example diagram illustrating an instance where the first information display processing unit of the first server detects a predetermined plurality of possible placement points with the highest win rate.

[0042] Figure 6 This example shows the first information display processing unit representing multiple recommended moves on the chessboard.

[0043] Figure 7 This example illustrates how the first information display and processing unit uses multiple colors to represent recommended placement points on a chessboard.

[0044] Figure 8 This is a flowchart illustrating the process by which the first information display processing unit of the first server sets the number of recommended placement points.

[0045] Figure 9 An example diagram is shown to illustrate the win probability value of the first player in the current situation, displayed on one side of the screen after a move is made in a Go game.

[0046] Figure 10 A system structure block diagram for implementing the method according to a preferred second embodiment of the present invention.

[0047] Figure 11 for Figure 10 The diagram shows the detailed structure of the second server.

[0048] Figure 12 and Figure 13 For illustrative purposes Figure 2 The flowchart shown is a process flow diagram of the second server operation, in which, Figure 12 Explain the operational process of the Chess Information Processing Department. Figure 13Explain the operation flow of the second game execution unit and the second information display processing unit.

[0049] Figure 14 This is an example diagram illustrating the state of displaying possible move points and the first player's playing style on the Go game screen.

[0050] Figure 15 A system structure block diagram for implementing the method according to a preferred third embodiment of the present invention.

[0051] Figure 16 for Figure 15 The detailed structural block diagram of the third device section is shown.

[0052] Figure 17 For illustrative purposes Figures 15 to 16 The flowchart shows the operation process of the first device unit.

[0053] Figure 18 This is an example diagram illustrating how the first device acquires and displays images of a Go game via a camera.

[0054] Figure 19 This is an example diagram illustrating the display of graphical objects on a screen showing a Go game.

[0055] Figure 20 This is an example diagram illustrating the win probability value of the player who made the previous move on the board at the current position.

[0056] Figure 21 This is an example diagram illustrating how a player's win rate value changes after a move is made.

[0057] Figure 22 This example illustrates how, after a player makes a move, their win rate is compared to their opponent's win rate, and how the player is more likely to win or lose.

[0058] Figure 23 This example illustrates how, after a player makes a move, their win rate is compared to their opponent's win rate, and an emoji is used to indicate whether the player has a higher probability of winning or losing.

[0059] Figure 24 This is a flowchart illustrating the process of setting up an analysis target player and displaying recommended move points for that player.

[0060] Figure 25 This is an example diagram illustrating the means by which players can select analytical targets from those playing a game of Go.

[0061] Figure 26 This is an example diagram illustrating the state of recommended move points for the corresponding player being analyzed.

[0062] Figure 27 This illustrates how the first device uses multiple colors to represent multiple recommended placement points on a chessboard.

[0063] Figure 28 This is a flowchart illustrating the process of the first device displaying a recap of the Go game after it has ended.

[0064] Figure 29 A system structure block diagram for implementing the method according to a preferred fourth embodiment of the present invention.

[0065] Figure 30 for Figure 29 The detailed structural block diagram of the fourth device section is shown.

[0066] Figure 31 For illustrative purposes Figures 29 to 30 The flowchart of the operation of the second device shown illustrates the display process of Go game-related information according to the fourth embodiment of the present invention.

[0067] Figure 32 The process of acquiring a code image is illustrated by the second device unit capturing a code located near the Go board where the Go game is played or a code displayed on the screen of the Go game.

[0068] Figure 33 This is a flowchart illustrating the process of setting up an analysis target player and displaying recommended move points for that player.

[0069] Figure 34 A system structure block diagram for implementing the method according to a preferred fifth embodiment of the present invention.

[0070] Figure 35 for Figure 34 The diagram shows the detailed structure of the fifth server.

[0071] Figure 36 For illustrative purposes Figures 34 to 35 The flowchart shown below illustrates the operation process of the fifth server, illustrating the display process of Go game-related information according to the fifth embodiment of the present invention.

[0072] Figure 37 This is a flowchart illustrating the process of setting up an analysis target player and displaying recommended move points for that player.

[0073] Figure 38 This is a structural block diagram illustrating an apparatus for performing the method according to the first to fifth embodiments of the present invention. Detailed Implementation

[0074] The preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings. In the process of describing the present invention, for ease of understanding, the same reference numerals are used for the same components in the drawings, and repeated descriptions of the same components are omitted.

[0075] In preferred embodiments of the present invention, in scenarios such as online Go games, offline Go games, and Go game content provided through broadcast or streaming media services, various information is calculated based on an artificial intelligence engine. This allows for easy display of various information on the screen during the Go game, including the player's final winning probability (win rate value) for each move, recommended moves calculated based on the win rate values ​​of each possible move, player playing style information, and Go game replays.

[0076] First, in the first embodiment, a subsystem will be introduced that provides an online Go game and generates and displays various information related to the Go game based on artificial intelligence.

[0077] Figure 1 A system structure block diagram for implementing the method according to a preferred first embodiment of the present invention.

[0078] like Figure 1 As shown, the first server 100 can be linked with the database 10 and multiple user terminals UR via a communication network. The user terminals UR can be network-connected computing devices such as smartphones, personal computer terminals, laptops, and tablets.

[0079] The first server 100 can connect to user terminals UR and provide them with a Go game that allows players to compete against each other or between players and computer players over the network. During the Go game, the first server 100 can display on the screen a win rate calculation based on an artificial intelligence engine and recommended move points based on that calculation.

[0080] Figure 2 for Figure 1 The diagram shows a detailed structural block diagram of the first server 100.

[0081] Figure 2 As shown, the first server 100 may include a first Go game execution unit 110, a first information display and processing unit 120, and a first artificial intelligence engine 130, etc. These units can interact with each other and can also interact with the database 10. The detailed functions of each unit and their interaction processes will be described in detail below.

[0082] Figure 3 For illustrative purposes Figure 2 The flowchart shown illustrates the operation of the first server 100, representing the flow of an online Go game method based on artificial intelligence according to a preferred embodiment of the present invention.

[0083] See Figures 1 to 3 First, after the first user terminal UR1 connects to the first server 100, it can request to execute a Go game. Then, the first Go game execution unit 110 of the first server 100 can execute a Go game between the first player and the second player through the screen of the first user terminal UR1 (step: S1).

[0084] The Go game allows the first and second players to choose black or white stones respectively, and take turns placing their stones in the desired positions on a graphically displayed board to gradually build their respective territories and determine the winner.

[0085] The first player can be the game player corresponding to the account of the first user terminal UR1. That is, after the first user of the first user terminal UR1 connects to the first server 100, he logs into the first server 100 through his own account and becomes the first player to play games with other players or artificial intelligence players.

[0086] The second player can be the game player corresponding to the account of the second user terminal UR2 connected to the first server 100. That is, after the second user of the second user terminal UR2 connects to the first server 100, they log in to the first server 100 through their own account and become the second player to play the game with the first player. The matchmaking used to facilitate the battle between the first player and the second player can be based on automatic matchmaking based on a matchmaking algorithm, or it can be specified according to the first player's choice.

[0087] On the other hand, the second player can also be a player from an AI player account that can play automated Go. The first player can choose whether to play against other players or against the AI ​​player by sending a request to the first server 100 through the user interface on the first user terminal UR1.

[0088] When the first information display processing unit 120 of the first server 100 is executing an online Go game, if the first player makes a move on a possible point before the first player's move order, the first artificial intelligence engine 130 can calculate the corresponding win rate value for each player (step: S2).

[0089] The first AI engine 130 can perform predictive learning, enabling it to quickly calculate the final winning probability (win rate) of both players when placing a stone on a possible move, before the Go player makes their move, by learning from experience to calculate possible moves based on the situation in a Go game. For example, the first AI engine 130 can be trained on at least one of the following: supervised learning of policy networks, which learns from large amounts of game records using a convolutional neural network (CNN) to grasp the positional information and patterns of moves and predict the next move; reinforcement learning of policy networks, which improves the performance of policy networks through repeated self-play to overcome the limitations of optimizing only game records; and reinforcement learning of value networks, which enhances the analytical capabilities of value networks by assigning weights to the win rates obtained from self-play of policy networks that reinforce Go outcome prediction.

[0090] Figure 4 An example diagram is provided to illustrate a winning probability value corresponding to a possible move point calculated by the first information display processing unit 120 of the first server 100 based on the first artificial intelligence engine 130.

[0091] like Figure 4 As shown, the first artificial intelligence engine 130 can calculate the winning probability, or win rate, of the first player when each player places a piece on a possible landing point in the current state. The win rate can be expressed as a percentage. For example, the first artificial intelligence engine 130 can calculate a win rate of "61%" when placing a piece on point A5, and a win rate of "54%" when placing a piece on point B4, etc.

[0092] Next, the first information display processing unit 120 of the first server 100 can detect from the calculated win rate values ​​the multiple predetermined possible placement points with the highest win rate when the first player places a piece (step: S3).

[0093] Figure 5 This is an example diagram illustrating an instance where the first information display processing unit 120 of the first server 100 detects a predetermined plurality of possible placement points with the highest win rate.

[0094] like Figure 5As shown, the first information display processing unit 120 can detect the six possible placement points with the highest win rate. For example, the first information display processing unit 120 can detect the six points with the highest expected win rate after the first player places a piece among all possible placement points, namely A5, B43, B6, C7, E2, and E7.

[0095] The so-called six highest-probability placements refer to the six highest-probability placements (from the first to the sixth highest, a predetermined number). That is, they refer to the N placements sorted by win rate (N is a predetermined integer of 2 or higher). Figure 6 In the example shown, N is 6, which is a possible placement point.

[0096] Next, the first information display processing unit 120 of the first server 100 can display the detected multiple possible move points as recommended move points on the Go game board so that the first player can identify them (step: S4).

[0097] Figure 6 This represents an example of the first information display processing unit 120 displaying multiple recommended moves on the chessboard.

[0098] like Figure 6 As shown, on the Go board in the first player's game screen, the six possible moves with the highest detected win rates are represented as six recommended moves for the first player to visually identify. For example, these six recommended moves could be displayed on the board with stones of a third color, distinct from black and white stones. For example, in... Figure 6 In the example shown, the recommended move point can be represented by the red piece CM. Recommended move points can be represented in various visually recognizable forms, such as dashed lines in the shape of a piece on the board, or pieces with special patterns.

[0099] On the other hand, the calculated win rate value is not displayed when showing recommended move points. However, according to one embodiment of the present invention, multiple recommended move points can be displayed differently based on the calculated win rate value. For example, they can be displayed in different colors according to the calculated win rate value.

[0100] Figure 7 This illustrates an example where the first information display processing unit 120 uses multiple colors to represent multiple recommended move points on a chessboard.

[0101] like Figure 7 As shown, on the Go board in the first player's game screen, the six possible moves with the highest detected win rates are displayed as six recommended moves. Different colors are used to represent the win rates corresponding to these recommended moves.

[0102] For example, among the six recommended moves with the highest win rates, the two moves with the highest win rates can be represented by a first color, such as red. Furthermore, besides the red moves, the two moves with the highest win rates can be represented by a second color, such as blue. Additionally, besides the red and blue moves, the two moves with the highest win rates can be represented by a third color, such as green. This color information based on the win rate is displayed on the screen of the first user terminal UR1, allowing the first player to intuitively identify the color of the recommended moves based on the win rate.

[0103] On the other hand, the number of recommended moves can be determined based on the first player's Go skill level. The first artificial intelligence engine 130 can analyze the first player's past games, set the first player's Go skill level, and store it in the database 10. The Go skill level refers to a rating system based on a prescribed grading standard for Go strength.

[0104] Figure 8 This is a flowchart illustrating the process by which the first information display processing unit 120 of the first server 100 sets the number of recommended move points.

[0105] like Figure 8 As shown, the first server 100 extracts the Go skill level of the first player from the database 10 (step: S11), and sets the number of recommended moves based on the extracted Go skill level of the first player (step: S12).

[0106] For example, the first server 100 can be configured such that the higher the first player's Go skill level, the more recommended moves are displayed. That is, there is a linear relationship between a player's Go skill level and the number of recommended moves. For instance, the higher a player's Go skill level, the more recommended moves are displayed, thus increasing the difficulty of choosing a move and broadening the range of possible moves. Conversely, the lower a player's Go skill level, the fewer recommended moves are displayed, thus reducing the difficulty of choosing a move.

[0107] On the other hand, the recommended move points can also be displayed when the first player uses their game items. In this case, the higher the level of the game items possessed by the first player, the fewer recommended move points will be displayed; conversely, the lower the level of the game items possessed by the first player, the more recommended move points will be displayed. The game items can refer to Go game items purchased by the player, obtained through events, or purchased or obtained based on Go game milestones, etc.

[0108] On the other hand, the recommended move points can be displayed when the first player has a specified paid membership level or higher. Furthermore, the higher the first player's paid membership level, the fewer recommended move points will be displayed.

[0109] See again Figures 1 to 3 If the first player places a stone at any of the multiple recommended placement points displayed, the first server 100 can calculate the first player's win rate in the current situation after the placement based on the first artificial intelligence engine 130 (step: S5). Next, the first server 100 displays the calculated win rate of the first player on one side of the Go game screen so that the first player can identify it (step: S6).

[0110] Figure 9 An example diagram is shown to illustrate the win probability value of the first player in the current situation, displayed on one side of the screen after a move is made in a Go game.

[0111] like Figure 9 As shown, when the first player places a piece at one of the recommended points, the win rate calculated by the first artificial intelligence engine 130 can be displayed through a win rate display window. This win rate display window can be displayed on the game board screen near the first player's identification information. The displayed win rate value for the first player is only visible to the first player. Conversely, the second player's win rate value will be displayed on the screen of the corresponding second player's second user terminal UR2.

[0112] On the other hand, after the first player makes their move, while the win rate is displayed in the win rate display window, the second player makes their move. After the second player makes their move, the first artificial intelligence engine 130 calculates the first player's win rate and updates the win rate displayed in the win rate display window with the calculated win rate.

[0113] As described above, according to a preferred embodiment of the present invention, information required for Go games can be provided based on artificial intelligence. For example, predictive learning can be performed so that an AI engine, which learns from experience to calculate possible moves based on the situation in a Go game, can quickly calculate the win rate of placing a stone on a possible move before the player makes their move, and detect multiple possible moves with high win rates as recommended moves, thereby providing the user with appropriate move information needed to win. Furthermore, after each player makes a move, the win rate under the current situation is confirmed in real time, thereby providing optimal information for the player's appropriate situation judgment, situational transitions, and strategic settings.

[0114] Below, in the second embodiment, a system will be introduced that provides an online Go game, reflects the player's playing style information, and generates and displays various information related to the Go game based on artificial intelligence.

[0115] Figure 10A system structure block diagram for implementing the method according to a preferred second embodiment of the present invention.

[0116] like Figure 10 As shown, the second server 200 can be linked with the database 20 and multiple user terminals UR via a communication network. The user terminals UR can be network-connected computing devices such as smartphones, personal computer terminals, laptops, and tablets.

[0117] The second server 200 can connect to user terminals UR and provide them with a Go game that allows players to compete against each other or against a computer player over the network. The second server 200 can analyze game records related to players, generate player style information based on this, and store and manage it in database 20. During a Go game, it can calculate the win rate based on the player's style information using an artificial intelligence engine, and display recommended moves on the screen based on the calculated win rate.

[0118] Figure 11 for Figure 10 The detailed structural block diagram of the second server 200 is shown.

[0119] Figure 11 As shown, the second server 200 may include a game style information processing unit 240, a second Go game execution unit 210, a second information display processing unit 220, a second artificial intelligence engine 230, etc. These units can interact with each other and can also interact with the database 20. The detailed functions of each unit and their interaction processes will be described in detail below.

[0120] Figure 12 and Figure 13 For illustrative purposes Figure 2 The flowchart shown is the operation process of the second server 200, in which... Figure 12 Explain the operation process of Chess Wind Information Processing Unit 240. Figure 13 The operation flow of the second game execution unit 210 and the second information display processing unit 220 is explained.

[0121] First, refer to Figure 12 The style information processing unit 240 of the second server 200 analyzes the game information of the first player based on the second artificial intelligence engine 230, thereby generating style information that reflects the first player's Go style (step: S21). Next, the style information processing unit 240 can store the generated style information in the database 20 (step: S22).

[0122] Go styles can be broadly divided into two categories: one is the combative Go style, which tends to gain advantages through meticulous reading of the game and engaging in close combat with individual stones, rather than simply defining boundaries; the other is the territorial Go style, which avoids life-or-death battles and gradually defines boundaries with exquisite situational judgment, ultimately winning by a slight difference in territory. In contrast to the combative style, it tends to win without fighting.

[0123] The second artificial intelligence engine 230 can calculate the percentage of a player's combat-oriented style and territory-oriented style by analyzing the player's past game information.

[0124] The chess style information can be obtained by analyzing players' game records and grouping or classifying them according to the proportion of combative and territorial tendencies in their chess styles.

[0125] For example, assuming the sum of combat and territory tendencies is 100%, if a player's game record analysis shows an 80% combat tendency and a territory tendency of less than 20%, they are classified as the first type, "Extreme Combat Type"; if the combat tendency is greater than 60% and less than 80%, and the territory tendency is less than 40% and greater than 20%, they are classified as the second type, "Ordinary Combat Type"; if the combat tendency is greater than 40% and less than 60%, and the territory tendency is less than 60% and greater than 40%, they are classified as the third type, "Mixed Type"; if the combat tendency is greater than 20% and less than 40%, and the territory tendency is less than 80% and greater than 60%, they are classified as the fourth type, "Ordinary Territory Type"; and if the combat tendency is less than 20% and the territory tendency is greater than 80%, they are classified as the fifth type, "Extreme Territory Type".

[0126] like Figure 13 As shown, after the first user terminal UR1 connects to the second server 200, it can request to execute a Go game. Then, the second Go game execution unit 210 of the second server 200 can execute a Go game between the first player and the second player through the screen of the first user terminal UR1 (step: S31).

[0127] The Go game allows the first and second players to choose black or white stones respectively, and take turns placing their stones in the desired positions on a graphically displayed board to gradually build their respective territories and determine the winner.

[0128] The first player can be the game player corresponding to the account of the first user terminal UR1. That is, after the first user of the first user terminal UR1 connects to the second server 200, he logs into the second server 200 through his own account and becomes the first player to play games with other players or artificial intelligence players.

[0129] The second player can be the game player whose account corresponds to the second user terminal UR2 connected to the second server 200. That is, after the second user of the second user terminal UR2 connects to the second server 200, they log in to the second server 200 through their own account and become the second player to play the game against the first player. The matchmaking used to facilitate the battle between the first player and the second player can be based on an automatic matchmaking algorithm or can be specified according to the first player's choice.

[0130] On the other hand, the second player can also be a player from an AI player account that can play automated Go. Whether the first player plays against other players or against the AI ​​player can be selected by the first user terminal UR1 through a request sent to the second server 200 via the user interface.

[0131] The second information display processing unit 220 of the second server 200 can extract the playing style information of the first player from the database 20 (step: S32). When the online Go game is in progress, before the first player makes a move, the second information display processing unit 220 of the second server 200 can calculate the win rate value corresponding to the first player's move at each possible point based on the extracted playing style information and the first artificial intelligence engine 130 (step: S33).

[0132] The second artificial intelligence engine 230 can perform predictive learning, enabling it to learn from experience and calculate possible moves based not only on the aforementioned wind analysis but also on the board position information. This allows the AI ​​engine to quickly calculate the win rate when a player makes a move on a possible point before the player makes their move.

[0133] For example, the second AI engine 230 can be trained based on at least one of the following: supervised learning of policy networks, which learns from large amounts of game records through convolutional neural networks (CNNs) to master the positional information and patterns of move points in order to predict the next move; reinforcement learning of policy networks, which improves the performance of policy networks by repeatedly playing against themselves to overcome the limitation of only optimizing game records; and reinforcement learning of value networks, which improves the analysis capabilities of value networks by mastering the win rates of game records obtained from self-play of policy networks that reinforce Go result prediction, assigning weights to them, and using them to make the next move.

[0134] Next, the second information display processing unit 220 of the second server 200 can detect from the calculated win rate values ​​the multiple predetermined possible placement points with the highest win rate when the first player places a piece (step: S34).

[0135] The second information display processing unit 220 of the second server 200 can display multiple detected possible move points as recommended move points on the Go game board so that the first player can identify them (step: S35).

[0136] For example, the second server 200 can display the multiple possible moves with the highest detected win rates as recommended moves on the Go board in the first player's game screen, so that the first player can identify them intuitively. For example, the recommended moves can be displayed on the board with stones of a third color that is different from black and white stones.

[0137] Figure 14 This is an example diagram illustrating the state of displaying possible move points and the first player's playing style on the Go game screen.

[0138] like Figure 14 As shown, recommended placement points on a Go board can be represented by red stones CM. According to another embodiment, recommended placement points can be represented in various visually recognizable forms, such as by dotted lines in the shape of Go stones on the board, or by stones with special patterns.

[0139] On one side of the Go game, such as the area near the first player, the first player's playing style information can be displayed. This allows the player to confirm their own playing style tendencies. The opponent's playing style information may not be disclosed.

[0140] On the other hand, the calculated win rate value is not displayed when showing recommended move points. However, according to one embodiment of the present invention, multiple recommended move points can be displayed differently based on the calculated win rate value. For example, as described above. Figure 7 The description can be displayed in different colors based on the calculated win rate value.

[0141] On the other hand, the number of recommended moves can be determined based on the first player's Go skill level. The second AI engine 230 can analyze the first player's past games, set the first player's Go skill level, and store it in the database 20. The Go skill level refers to a rating system based on a prescribed grading standard for Go strength.

[0142] The second server 200 extracts the Go skill level of the first player from the database 20, and sets the number of recommended moves based on the extracted Go skill level of the first player.

[0143] For example, the second server 200 could be configured such that the higher the Go skill level of the first player, the more recommended moves are displayed. That is, there is a linear relationship between a player's Go skill level and the number of recommended moves. For instance, the higher a player's Go skill level, the more recommended moves are displayed, thus increasing the difficulty of choosing a move and broadening the range of possible moves. Conversely, the lower a player's Go skill level, the fewer recommended moves are displayed, thus reducing the difficulty of choosing a move.

[0144] On the other hand, the recommended move points can also be displayed when the first player uses their game items. In this case, the higher the level of the game items possessed by the first player, the fewer recommended move points will be displayed; conversely, the lower the level of the game items possessed by the first player, the more recommended move points will be displayed. The game items can refer to Go game items purchased by the player, obtained through events, or purchased or obtained based on Go game milestones, etc.

[0145] On the other hand, the recommended move points can be displayed when the first player has a specified paid membership level or higher. Furthermore, the higher the first player's paid membership level, the fewer recommended move points will be displayed.

[0146] See again Figure 13 If the first player places a stone at any of the multiple recommended placement points displayed, the second server 200 can calculate the first player's win rate in the current situation after the placement based on the second artificial intelligence engine 230 (step: S36). Next, the second server 200 displays the calculated win rate of the first player on one side of the Go game screen so that the first player can identify it (step: S37).

[0147] When the first player places a piece at one of the recommended points, the win rate calculated by the second artificial intelligence engine 230 can be displayed through a win rate display window. This win rate display window is located on the game board screen near the first player's identification information. The displayed win rate value for the first player is only visible to the first player. Conversely, the second player's win rate value will be displayed on the screen of the corresponding second player's second user terminal UR2.

[0148] On the other hand, after the first player makes their move, while the win rate is displayed in the win rate display window, the second player makes their move. After the second player makes their move, the second AI engine 230 calculates the first player's win rate and updates the win rate displayed in the win rate display window with the calculated win rate.

[0149] As described above, according to a preferred embodiment of the present invention, information required for the Go game can be provided based on artificial intelligence, according to the player's playing style preferences. For example, predictive learning can be performed based on playing style information, enabling an AI engine that has learned through experience to calculate possible moves based on the Go game situation to quickly calculate the win rate of placing a stone on a possible move before the player makes their move, and detect multiple possible moves with high win rates as recommended moves, thereby providing the user with appropriate move information needed to win. Furthermore, after each player makes a move, the win rate under the current situation is confirmed in real time, thereby providing optimal information for the player's appropriate situation judgment, situational transitions, and strategic settings.

[0150] The third embodiment describes a system for acquiring images of Go games played offline or online, and for displaying various information related to the Go games generated based on artificial intelligence.

[0151] Figure 15 A system structure block diagram for implementing the method according to a preferred third embodiment of the present invention.

[0152] like Figure 15 As shown, the first device unit DV1 can be linked with the database 30 and the third server via a communication network. The third server 300 may have a third artificial intelligence engine 330. The first device unit DV1 may be implemented based on a user terminal. For example, the first device unit DV1 may be a network-connected computing device such as a smartphone, personal computer terminal, laptop, or tablet computer.

[0153] The first device unit DV1 can download application installation files from a third server 300 or a third-party online application store, and has modules for executing processes by executing the downloaded installation files.

[0154] The first device unit DV1, as a user account-authenticated device, can acquire images of the Go game between the first player and the second player via camera 371, and display the Go game between the first player and the second player on the screen based on the acquired images. When the Go game is played, whenever a move is made, the first device unit DV1 calculates the predicted final winning probability and winning percentage of at least one player based on the third artificial intelligence engine 330, and displays the information related to the winning percentage in real time on one side of the screen after the move occurs, thereby displaying various information related to the Go game on the Go game screen.

[0155] Figure 16 for Figure 15 The detailed structural block diagram of the third device section is shown.

[0156] like Figure 16As shown, the first device unit DV1 may include a camera 371 and a first processor unit 375, etc. These units can interact with each other and can also interact with the database 30 and the third server 300. The detailed functions of each unit and their interaction process will be described in detail below.

[0157] Figure 17 For illustrative purposes Figures 15 to 16 The flowchart of the operation of the first device unit DV1 shown illustrates the process of displaying Go game-related information according to the third embodiment of the present invention.

[0158] First, the first player and the second player can play a game of Go, and the user can watch the Go game between the first player and the second player. This can be done by the user actually watching the Go game between the first player and the second player, or through a television screen or monitor. The Go game allows the first player and the second player to choose black or white stones respectively, taking turns placing their stones in desired positions on the board, gradually building their respective territories to determine the winner. In this embodiment, the user can obtain various information related to the Go game they are watching through the first device.

[0159] See Figures 15 to 17 First, the first processor unit 375 of the first device unit DV1 can acquire images of the Go game between the first player and the second player via the camera 371 (step: S41). For example, the user can activate the processor unit of the first device unit DV1, and the first device unit can acquire images of the Go game by activating the camera 371 built into or connected to the first device.

[0160] Next, the first processor unit 375 may display the Go game between the first player and the second player on the screen based on the acquired image (step: S42). For example, the first processor unit 375 may generate a Go game graphic image between the first player and the second player in real time based on the acquired image, and display the generated Go game graphic image on the screen.

[0161] The first processor unit 375 can display various graphic objects related to the Go game on the real-time Go game image acquired by the camera 371, i.e., the real-scene image. The Go game graphic image may include a first image layer corresponding to the acquired image, and a second image layer overlaid on the first image layer and including graphic objects representing information related to the Go game.

[0162] Figure 18 This is an example diagram illustrating how the first device unit DV1 acquires images of a Go game via camera 371 and displays them on the screen.

[0163] like Figure 18As shown, the first device unit DV1 captures an actual Go game or a Go game displayed on a monitor or television using a camera 371, thereby obtaining an image of the Go game and displaying it on the screen of the first device unit DV1. The first device can overlay graphic objects providing Go-related information onto the real-time Go game image displayed on the screen.

[0164] To this end, the first processor unit 375 can detect Go game information such as the area of ​​the chessboard contained in the first image layer, the points on the chessboard, the placement position and color of the pieces, and display the graphic object at the required position on the chessboard based on the detected Go game information.

[0165] On the other hand, according to another embodiment, the first device unit DV1 can analyze the real-time images acquired by the camera 371 and generate a graphic representing the situation of the Go game in real time. Furthermore, graphic objects representing information related to the Go game can be arranged on the generated graphic.

[0166] That is, the first device unit DV1 can display graphic objects for providing information on the real-world image of the Go game acquired by the camera 371, or generate graphics such as animations representing the Go game by analyzing the real-world image and then display graphic objects for providing information on it.

[0167] See again Figure 17 When playing the Go game, the first device DV1 can calculate the predicted final winning probability value, i.e., the winning rate value, of at least one player based on the third artificial intelligence engine 330 each time a move is made (step: S43).

[0168] For example, the first device unit DV1 can calculate the win rate of the first player in the current situation after a move has been made, based on the third artificial intelligence engine 330. The win rate represents the probability that the first player will ultimately win in the current situation, and as mentioned above, it can be expressed as a percentage. The first device unit DV1 can also calculate the win rate of the second player based on the first player's win rate. For example, since the sum of the win rates of the two players should be 100%, the second player's win rate can be 100 minus the difference between the first player's win rate and the second player's win rate.

[0169] The third AI engine 330 is an AI engine that learns from experience to calculate possible moves based on the situation in a Go game. During a Go game, it can quickly calculate the final winning probability (win rate) for both players in the current situation. Therefore, the third AI engine 330 can calculate the winning probability of both players both after the most recent move and also before a specific player makes a move, considering the player's possible moves (i.e., assuming a move has been made).

[0170] For example, the third AI engine 330 can be trained based on at least one of the following: supervised learning of policy networks, which learns from large amounts of game records through convolutional neural networks (CNNs) to grasp the positional information and patterns of move points in order to predict the next move; reinforcement learning of policy networks, which improves the performance of policy networks by repeatedly playing against themselves to overcome the limitation of only optimizing game records; and reinforcement learning of value networks, which improves the analysis capabilities of value networks by mastering the win rates of game records obtained from self-play of policy networks that reinforce Go result prediction, assigning weights to them, and using them to make the next move.

[0171] Next, the first device unit DV1 can display the winning probability value and related information of the player playing Go in real time on one side of the screen after the move is made (step: S44).

[0172] Figure 19 The example diagram, which exemplifies the display of graphical objects on a screen showing a Go game, represents an example of the predicted probability, or win rate, of the first and second players in the current position.

[0173] like Figure 19 As shown, the screen of the first device unit DV1 displays a Go board used in a game of Go, and indicates the colors of the pieces of the first player and the second player, the current state of the game, etc. The first device unit DV1 can display the win rate of the first player and the win rate of the second player in the current situation on the screen respectively.

[0174] For example, a win rate value such as "current win rate 71%" can be displayed near the graphic object representing the first player using black, and a win rate value such as "current win rate 29%" can be displayed near the graphic object representing the second player using white. Therefore, users can monitor the win rates of both players in real time while watching a Go game, greatly enhancing the viewing experience by observing the dynamically changing win rate values ​​after each move. Although this embodiment exemplifies the representation of win rate values ​​in numerical form, according to an embodiment of the present invention, graphic objects whose shape, color, or size changes with the numerical value, such as emoticons or graphics, can also be used for representation.

[0175] The first device, DV1, can display the most recently placed stone on the Go game screen. For example, as... Figure 19As shown, the first device unit DV1 can indicate the black piece of the first player who recently made a move in the current situation by using a red border, which can be identified by the user; it can also indicate the place of the most recently made move by making the most recently made piece flash or by adjusting its color to a special hue that distinguishes it from other pieces.

[0176] On the other hand, according to one embodiment of the present invention, the first device unit DV1 may also, after a move is made, overlay and display the predicted final winning probability value (i.e., the win rate value) of the player who made the move, calculated for that move, onto the point on the chessboard where the move occurred. When the next move is made, the first device unit DV1 may clear the win rate value of the previous move and overlay and display the win rate value corresponding to the situation of the next move onto the point of the move.

[0177] Figure 20 This is an example diagram illustrating the win probability value of the player who made the previous move on the board at the current position.

[0178] like Figure 20 As shown, assuming the current situation is as follows after the first player places a piece at a specific point, the first device DV1 can calculate the win rate value in the current situation and display the value superimposed above the piece at the most recent place. Therefore, the user can intuitively identify the most recent place and the change in win rate caused by the placing action each time a piece is placed.

[0179] On the other hand, according to an embodiment of the present invention, after a move is made, the first device unit DV1 compares the win rate value of the player who made the move, calculated for the move, with the win rate value before the move; based on the comparison of the win rate values, it displays on the screen whether the win rate after the move has increased or decreased compared to the win rate before the move, so that the user can identify it.

[0180] Figure 21 This is an example diagram illustrating how a player's win rate value changes after a move is made.

[0181] like Figure 21 As shown, after a move is made, the area near the player's graphic object can display a graphic object indicating whether the player's win rate has increased or decreased after the move. For example, assuming the first player's win rate has increased after the move of the black piece, the first device unit DV1 can display a graphic object IM1 indicating the increase in win rate compared to before the move in the area near the graphic object representing the first player.

[0182] The graphic object IM1 can display an upward arrow in a first color when the win rate increases after a move, and can also display text describing the increase in the win rate (e.g., "UP"). Conversely, it can display a downward arrow in a second color when the win rate decreases after a move, and can also display text describing the decrease in the win rate (e.g., "DOWN").

[0183] On the other hand, the first device unit DV1 can also compare the win rate value of the player who made the move with the win rate value of the opponent after the move is made; based on the comparison of the win rate values, it can display on the screen whether the player who made the move has a higher probability of winning or a higher probability of losing, so that the user can make an identification.

[0184] Figure 22 This example illustrates how, after a player makes a move, their win rate is compared to their opponent's win rate, and how the player is more likely to win or lose.

[0185] like Figure 22 As shown, the first device unit DV1 can adjust the size of the graphic object used to identify players based on the win rate value. For example, when a specific player's win rate value is higher than that of the opponent player in the current situation, the first device unit DV1 can display the graphic object used to identify that player at a relatively larger size than the graphic object used to identify the opponent player.

[0186] For example, if after the first player makes a move, the first player's winning probability is higher than the second player's in the current situation, then when displaying the Go game, the graphic object IM2 used to identify the first player is displayed as a larger size than the graphic object used to identify the second player.

[0187] Furthermore, the first device unit DV1 can also display text on the graphic object used to identify players, indicating whether the current win rate is higher or lower than that of the opponent. For example, within or near the graphic object IM2 used to identify the first player, whose win rate is relatively high in the current situation, text such as "High probability of winning" can be displayed to indicate that the second player has a higher win rate than their opponent. Conversely, within or near the graphic object used to identify the second player, text such as "Low probability of winning" can be displayed to indicate that the second player has a lower win rate than the first player.

[0188] Figure 23 This example illustrates how, after a player makes a move, their win rate is compared to their opponent's win rate, and an emoji is used to indicate whether the player has a higher probability of winning or losing.

[0189] like Figure 23As shown, the first device unit DV1 can adjust the shape of the graphic object used to identify players based on their win rate. For example, when a specific player's win rate is higher than that of their opponent in the current situation, the first device unit DV1 can insert emoticons that typically have a positive meaning (e.g., a happy expression or behavior, sunny weather, etc.) into the graphic object used to identify that player. Conversely, emoticons that typically have a negative meaning (e.g., a sad or angry expression or behavior, or rainy weather, etc.) can be inserted into the graphic object used to identify players whose win rate is lower than that of their opponent.

[0190] For example, assuming that after the first player makes a move, their winning probability is higher than the second player's in the current situation, when displaying the Go game, an emoticon, such as a "smiling emoticon," can be inserted on the graphic object IM3 used to identify the first player to visually indicate that their winning probability is higher than the second player's. Conversely, an emoticon, such as a "sad emoticon," can be inserted on the graphic object identifying the second player, whose winning probability is relatively lower.

[0191] On the other hand, the processor of the first device unit DV1 can provide the function of setting either of the two players playing Go as the analysis target player and displaying the recommended move points as described in the previous first to second embodiments to the analysis target player.

[0192] Figure 24 This is a flowchart illustrating the process of setting up an analysis target player and displaying recommended move points for that player.

[0193] like Figure 24 As shown, the first device unit DV1 can display the selection means for selecting the analysis target player among the first player or the second player (step: S51).

[0194] Figure 25 This is an example diagram illustrating the means by which players can select analytical targets from those playing a game of Go.

[0195] like Figure 25 As shown, the first device unit DV1 can display a selection method UI1 for selecting either the first or second player in a Go game. The user can select the icon corresponding to the player to be analyzed using the selection method via touch or other means. The first device unit DV1 can then set the player selected through the selection method as the target player for analysis.

[0196] Next, before the player selected by the selection means makes a move, the first device unit DV1 calculates the expected winning probability, i.e. the win rate value, when making a move at each possible point based on the third artificial intelligence engine 330 (step: S52).

[0197] The first device unit DV1 can detect from the calculated win rate values ​​a predetermined number of possible moves with the highest win rate when the analyzed player makes a move (step: S53). The first device unit DV1 can display the detected possible moves as recommended moves on the Go board displaying the Go game screen for the user to identify (step: S54).

[0198] For example, assuming the user selects the first player as the analysis target player through a selection method, the first device unit DV1 can set the first player as the analysis target player. Then, when it is the first player's turn to make a move, the first device unit DV1 can calculate the win rate value corresponding to the first player's move at each possible move point in the current state based on the third artificial intelligence engine 330. From the calculated win rate values, the first device unit DV1 detects the possible move points with the highest win rate value when the first player makes a move, a predetermined number (e.g., N, where N is a predetermined integer of 2 or higher), and displays the detected possible move points as recommended move points on the Go game board for the user to identify.

[0199] Figure 26 This is an example diagram illustrating the state of recommended move points for the corresponding player being analyzed.

[0200] like Figure 26 As shown, in the first player's turn (i.e., before their next move), the predetermined number of points with the highest win rates detected on the Go game display screen, such as six possible moves, are represented as six recommended moves for intuitive user identification. For example, these six recommended moves could be displayed on the board using stones of a different color than black and white stones. Figure 26 In the example shown, the recommended move point can be represented by the red piece CM. Recommended move points can be represented in various visually recognizable forms, such as dashed lines in the shape of a piece on the board, or pieces with special patterns.

[0201] On the other hand, according to one embodiment of the present invention, there are multiple recommended move points. Based on the win rate value corresponding to each recommended move point, the relative levels of win rates among the recommended move points are distinguished for identification. That is, multiple recommended move points can be displayed differently based on the calculated win rate value. For example, they can be displayed with different colors, shapes, sizes, etc., based on the calculated win rate value.

[0202] Figure 27 This illustrates an example where the first device unit DV1 uses multiple colors to represent multiple recommended placement points on a chessboard.

[0203] like Figure 27 As shown, before the first player, designated as the analysis target, makes their move, the six recommended moves on the Go board (representing the game's visuals) represent the six points with the highest win rate for the first player. These recommended moves are represented using various colors based on their corresponding win rates.

[0204] For example, among the six recommended moves with the highest win rates, the two moves with the highest win rates can be represented by the first color, such as red. Furthermore, besides the red moves, the two moves with the highest win rates can be represented by the second color, such as blue. Additionally, besides the red and blue moves, the two moves with the highest win rates can be represented by a third color, such as green. This color information based on the win rate is displayed on the screen using icons or prompts, allowing users to intuitively identify the color of the recommended moves based on the win rate.

[0205] Furthermore, when the first device unit DV1 places a piece at the recommended placement point, it displays the expected winning probability, i.e., the win rate value, in relation to the displayed recommended placement point for identification. For example, such as Figure 27 As shown, the first device unit DV1 can simultaneously display the corresponding win rate value when displaying the recommended move point. For example, the win rate value can be displayed inside the piece representing the recommended move point or in its vicinity.

[0206] As described above, according to a preferred embodiment of the present invention, when it is the turn of the player designated as the analysis target to make a move, a predetermined number of recommended moves can be displayed on the Go board in various ways. Therefore, when watching a Go game, the user can select a player of interest or one they wish to analyze. Each time it is the turn of the player to be analyzed, the system can confirm recommended moves with high win rates in real time and also confirm the actual point where the player to be analyzed places their move, thereby enhancing the viewing experience of the Go game. Alternatively, when the user needs to provide Go instruction or guidance to a player, and thus requires advice or guidance during the game, suggestions or guidance can be provided based on the real-time displayed recommended moves, making it an excellent educational aid.

[0207] Figure 28 This is a flowchart illustrating the process of the first device displaying a recap of the Go game after it has ended.

[0208] like Figure 28As shown, the first processor unit 375 of the first device can set the start time and end time of the review that the user wants to review based on the user interface (step: S61).

[0209] When a replay request signal is received, the first processor unit 375 can display the replay screen from the start time to the end time, and display the changing win rate value for each move (step: S62). Therefore, even after the Go game ends, the user can still view the win rate value that changes with each move, thereby enhancing the enjoyment of the replay.

[0210] As described above, according to a preferred embodiment of the present invention, when a user is playing Go in real-world conditions or watching Go games on screen, the device can obtain information needed for the Go game based on artificial intelligence, such as the predicted win rate values ​​for both players updated in real time for each move, changes in win rates, and recommended move points for the corresponding analyzed player. Therefore, it not only enhances the enjoyment of watching Go games but also serves as a highly valuable educational tool for users learning Go.

[0211] The fourth embodiment describes a system that uses a device to acquire images of Go games played offline or online, identifies the players playing the Go game, and uses the players' playing style information, skill information, and other information generated based on artificial intelligence to display various information related to the Go game.

[0212] Figure 29 A system structure block diagram for implementing the method according to a preferred fourth embodiment of the present invention.

[0213] like Figure 29 As shown, the second device unit DV2 can be linked with the database 40 and the fourth server via a communication network. The fourth server 400 may have a fourth artificial intelligence engine 430. The second device unit DV2 may be implemented based on a user terminal. For example, the second device unit DV2 may be a network-connected computing device such as a smartphone, personal computer terminal, laptop, or tablet computer.

[0214] The second device unit DV2 can download application installation files from the fourth server 400 or a third-party online application store, and has modules for executing processes by executing the downloaded installation files.

[0215] The second device unit DV2, as a user account-authenticated device, extracts first player information (including identification information, playing style information, and playing strength information of the first player) and second player information (including identification information, playing style information, and playing strength information of the second player) from the database 40. It can acquire images of the Go game between the first player and the second player via camera 471 and display the Go game between the first player and the second player on the screen based on the acquired images. During the Go game, the fourth artificial intelligence engine 430 uses the first player information and the second player information to calculate in real time the predicted final winning probability value (i.e., the win rate value) of at least one player in the current game state. Information related to the calculated win rate value is displayed on one side of the screen. Here, based on the real-time calculated win rate value, the information related to the win rate value displayed at each move is updated.

[0216] Figure 30 for Figure 29 The detailed structural block diagram of the fourth device section is shown.

[0217] like Figure 30 As shown, the second device unit DV2 may include a camera 471 and a second processor unit 475, etc. These units can be interconnected and can also be linked with the database 40 and the fourth server 400. The detailed functions of each unit and their interconnection process will be described in detail below.

[0218] Figure 31 For illustrative purposes Figures 29 to 30 The flowchart of the operation of the second device unit DV2 shown illustrates the process of displaying Go game-related information according to the fourth embodiment of the present invention.

[0219] First, a first player and a second player can play a game of Go, and a user can watch the Go game between the first player and the second player. This can be done by the user actually watching the Go game between the first player and the second player, or through a television screen or monitor. The Go game allows the first player and the second player to choose black or white stones respectively, taking turns placing their stones in desired positions on the board, gradually building their respective territories to determine the winner. In one embodiment, the user can obtain various information related to the Go game they are watching through a first device.

[0220] See Figures 29 to 31 First, the second processor unit 475 of the second device unit DV2 can extract first player information containing the identification information, playing style information, and playing strength information of the first player and second player information containing the identification information, playing style information, and playing strength information of the second player from the database 40 respectively (step: S71).

[0221] To this end, the second device unit DV2 uses the camera 471 to capture images of codes near the Go board where the Go game is played or codes displayed on the screen of the Go game, and obtains code images. By decoding the obtained code images, it can access at least one of the first player information and the second player information stored in the database 40.

[0222] The code is encrypted information containing an address that allows access to the first player's information and the second player's information in database 40. It can be a two-dimensional barcode, a QR (Quick Response) code, etc. The second device unit DV2 captures the code image using camera 471, and after decoding, it can access database 40 to obtain the first player's information and the second player's information.

[0223] Figure 32 The process of acquiring a code image by capturing a code near the Go board or displayed on the screen of a Go game, as an example, is illustrated by the second device unit DV2.

[0224] like Figure 32 As shown, codes are placed near the Go board or on the screen displaying the Go game. For example, according to one embodiment, in order to enable viewers to obtain various information related to the Go game during a real-world Go game, the codes can be placed around the Go board or in specific locations related to the venue, such as in printed materials or on a display screen, and provided to the user.

[0225] On the other hand, according to one embodiment, when a Go game is played on a television or monitor via broadcast or streaming services, the code can also be displayed on one side of the corresponding screen.

[0226] The code can be generated, distributed, and provided by the organizer of the Go game or the Go game information service provider in this embodiment. While the code can be made public and photographed by all Go spectators, according to one embodiment, the photographing permission can be limited to users who meet specific conditions, such as specific paid members or Go academy members who can receive the Go game information service. In this case, the Go information display service can be provided to users as a limited service.

[0227] For example, in offline Go venues, the code can be posted in areas accessible only to specific individuals (such as paid members, members of a certain level or higher, or members of a specific group), thus ensuring that only authorized personnel can photograph the code. Similarly, when a Go match is broadcast live via broadcast or streaming services, the code is displayed only on screens of authorized user accounts (such as paid members, members of a certain level or higher, or members of a specific group), ensuring that only authorized personnel can photograph the code.

[0228] The second device unit DV2 captures the code image using camera 471, and by decoding the acquired code image, it can access player information stored in database 40, such as first player information or second player information.

[0229] On the other hand, according to another embodiment, the second device unit DV2 can use camera 471 to capture the face of the first player or the second player, analyze the information through the facial recognition module, and then access database 40 to obtain information about the first player or the second player. For example, the second device unit DV2 can use the camera 471 to capture the face of at least one of the first player and the second player to obtain a facial image, and then query the database 40 based on the acquired facial image to access at least one of the first player information and the second player information.

[0230] In the database 40, each player's playing style information and external strength information can be stored in relation to the corresponding player identification information. The fourth artificial intelligence engine 430 can be trained based on big data of game information containing players' playing style and strength information to predict win rates under different game conditions.

[0231] For example, the fourth AI engine 430 is an AI engine that learns from experience based on big data of Go games, including information on each player's Go playing style and strength, to calculate possible moves based on the current situation in a Go game. During a Go game, it can quickly calculate the final winning probability (win rate) of both players based on player information. Therefore, the fourth AI engine 430 can calculate the winning probability of both players both after the most recent move and before a specific player makes a move, considering the possibility of the player making a move at a possible point (i.e., assuming a move has been made).

[0232] For example, the fourth AI engine 430 can be trained based on at least one of the following: supervised learning of policy networks, which learns from large amounts of game records using convolutional neural networks (CNNs) to grasp the positional information and patterns of moves and predict the next move; reinforcement learning of policy networks, which improves the performance of policy networks through repeated self-play to overcome the limitation of only optimizing game records; and reinforcement learning of value networks, which improves the analytical capabilities of value networks by mastering the win rates of game records obtained through self-play of policy networks that reinforce Go outcome prediction, assigning weights to these win rates to predict the next move. See again. Figure 31 Next, the second device unit DV2 can acquire images of the Go game between the first player and the second player via camera 471 (step: S72). The second device unit DV2 can display the Go game between the first player and the second player on the screen based on the acquired images (step: S73).

[0233] For example, the second device unit DV2 can generate a graphic image of a Go game between the first player and the second player in real time based on the acquired image, and display the generated Go game graphic image on the screen. The second processor unit 475 can display various graphic objects representing information related to the Go game on the real-time Go game image, i.e., the live-action image, acquired by the camera 471. The Go game graphic image may include a first image layer corresponding to the acquired image, and a second image layer overlaid on the first image layer and including graphic objects representing information related to the Go game.

[0234] For example, as previously mentioned Figure 18 The second device unit DV2 captures images of an actual Go game or a Go game displayed on a monitor or television using a camera 471, and displays these images on the screen of the second device unit DV2. The second device can overlay graphic objects providing Go-related information onto the real-time Go game image displayed on the screen.

[0235] Therefore, the second processor unit 475 can detect Go game information such as the area of ​​the chessboard included in the first image layer, the points on the chessboard, the placement position and color of the pieces, and display the graphic object at the required position on the chessboard based on the detected Go game information.

[0236] On the other hand, according to another embodiment, the second device unit DV2 can analyze the real-time images acquired by the camera 471 and generate a graphic representing the situation of the Go game in real time. Furthermore, graphic objects representing information related to the Go game can be arranged on the generated graphic.

[0237] That is, the second device unit DV2 can display graphic objects for providing information on the real-world image of the Go game acquired by the camera 471, or generate graphics such as animations representing the Go game by analyzing the real-world image and then display graphic objects for providing information on it.

[0238] See again Figure 31 When playing the Go game, the second device DV2 uses the fourth artificial intelligence engine 430 to calculate in real time the predicted final winning probability value, i.e., the winning rate value, of at least one player in the current game state based on the first player information and the second player information (step: S74).

[0239] For example, when playing the Go game, the second device unit DV2 can calculate the win rate of the at least one player based on the fourth artificial intelligence engine 430, which reflects the playing style and strength of the first player and the second player respectively, when each move is made.

[0240] For example, the second device unit DV2 can calculate the win rate of the first player in the current situation after a move has been made, based on the fourth artificial intelligence engine 430. The win rate represents the probability that the first player will ultimately win in the current situation, and as mentioned above, it can be expressed as a percentage. The second device unit DV2 can also calculate the win rate of the second player based on the first player's win rate. For example, since the sum of the win rates of the two players should be 100%, the second player's win rate can be 100 minus the first player's win rate.

[0241] Next, the second device unit DV2 can display information related to the calculated win rate value on one side of the screen (step: S75). Based on the win rate value calculated in real time, the information related to the win rate value displayed at each move can be updated (step: S76).

[0242] For example, the screen of the second device unit DV2 displays a Go board game and shows the colors of the pieces of the first and second players, the current state of the game, etc. The second device unit can display the win rate of the first player and the win rate of the second player in the current situation on the screen separately.

[0243] For example, a win rate value such as "current win rate 71%" can be displayed near the graphic object representing the first player using black, and a win rate value such as "current win rate 29%" can be displayed near the graphic object representing the second player using white. Therefore, users can monitor the win rates of both players in real time while watching a Go game, greatly enhancing the viewing experience by observing the dynamically changing win rate values ​​after each move. Although this embodiment exemplifies the representation of win rate values ​​in numerical form, according to an embodiment of the present invention, graphic objects whose shape, color, or size changes with the numerical value, such as emoticons or graphics, can also be used for representation.

[0244] The second device, DV2, can display the most recent move on the Go game screen. For example, as previously mentioned... Figure 19 As exemplarily described, the second device unit DV2 can indicate the black piece of the first player who recently made a move in the current situation by using a red border, which can be identified by the user; or it can indicate the place of the most recently made move by making the most recently made piece flash, or by adjusting its color to a special hue that distinguishes it from other pieces.

[0245] On the other hand, according to one embodiment of the present invention, the second device unit DV2 may also, after a move is made, overlay and display the predicted final winning probability value (i.e., the win rate value) of the player who made the move, calculated for that move, onto the point on the chessboard where the move occurred. When the next move occurs, the second device unit DV2 may clear the win rate value of the previous move and overlay and display the win rate value corresponding to the situation of the next move onto the point of the move.

[0246] For example, as previously mentioned Figure 20 Using the same concept, assuming the current situation is as follows after the first player places a piece at a specific point, the second device (DV2) can calculate the win rate value in the current situation and display the value superimposed above the piece at the most recent placement point. Therefore, the user can intuitively identify the most recent placement point and the change in win rate caused by the placement action each time a piece is placed.

[0247] On the other hand, according to an embodiment of the present invention, after a move is made, the second device unit DV2 compares the win rate value of the player who made the move, calculated for the move, with the win rate value before the move; based on the comparison of the win rate values, it displays on the screen whether the win rate after the move has increased or decreased compared to the win rate before the move, so that the user can identify it.

[0248] For example, as previously mentioned Figure 21The same concept applies here: after a move is made, the area near the player's graphic object can display a graphic object indicating whether the player's win rate has increased or decreased after the move. For example, assuming the first player's win rate has increased after making a move with black, the second device unit DV2 can display a graphic object indicating the increase in win rate compared to before the move in the area near the first player's graphic object.

[0249] The graphic object can display an upward arrow in a first color when the win rate increases after a move, and can represent text describing the increase in the win rate (e.g., "UP"). Conversely, it can display a downward arrow in a second color when the win rate decreases after a move, and can represent text describing the decrease in the win rate (e.g., "DOWN").

[0250] On the other hand, the second device unit DV2 can also compare the win rate value of the player who made the move with the win rate value of the opponent after the move is made; based on the comparison of the win rate values, it can display on the screen whether the player who made the move has a higher probability of winning or a higher probability of losing, so that the user can make an identification.

[0251] For example, as previously mentioned Figure 22 Similarly, the second device unit DV2 can adjust the size of the graphic object used to identify players based on the win rate value. For example, when a particular player's win rate value is higher than that of the opponent player in the current situation, the second device unit DV2 can display the graphic object used to identify that player at a relatively larger size than the graphic object used to identify the opponent player.

[0252] For example, if after the first player makes a move, the first player's winning probability is higher than the second player's in the current situation, then when displaying the Go game, the graphic object used to identify the first player is displayed at a larger size than the graphic object used to identify the second player.

[0253] Furthermore, the second device unit DV2 can also display text on the graphic object used to identify players, indicating whether the current win rate is higher or lower than that of the opponent. For example, within or near the graphic object identifying the first player (whose win rate is relatively high in the current situation), text such as "High Win Rate" can be displayed to indicate that the second player has a higher win rate than their opponent. Conversely, within or near the graphic object identifying the second player, text such as "Low Win Rate" can be displayed to indicate that the second player has a lower win rate than the first player.

[0254] In addition, as previously referred to Figure 23The second device unit DV2 can adjust the shape of the graphic object used to identify players based on their win rate. For example, when a specific player's win rate is higher than that of their opponent in the current situation, the second device unit DV2 can insert emoticons that typically have a positive meaning (e.g., a happy expression or behavior, sunny weather, etc.) into the graphic object used to identify that player. Conversely, emoticons that typically have a negative meaning (e.g., a sad or angry expression or behavior, or rainy weather, etc.) can be inserted into the graphic object used to identify players whose win rate is lower than their opponent's.

[0255] For example, assuming that after the first player makes a move, their winning probability is higher than the second player's, an emoticon, such as a "smiling emoticon," can be inserted onto the graphic object identifying the first player when displaying the Go game. Conversely, an emoticon, such as a "sad emoticon," can be inserted onto the graphic object identifying the second player, whose winning probability is relatively lower.

[0256] On the other hand, the processor of the second device unit DV2 can provide the function of setting either of the two players playing Go as the analysis target player and displaying the recommended move points as described in the previous first to third embodiments to the analysis target player.

[0257] Figure 33 This is a flowchart illustrating the process of setting up an analysis target player and displaying recommended move points for that player.

[0258] like Figure 33 As shown, the second device unit DV2 can display the selection means for selecting the analysis target player among the first player or the second player (step: S81).

[0259] For example, the second device unit DV2 can display a selection method for choosing either the first or second player in a Go game. The user can select the icon corresponding to the player to be analyzed using the selection method, such as by touch. The second device unit DV2 can then set the player selected by the selection method as the target player for analysis. Related examples are as previously mentioned. Figure 25 The same concept as described above.

[0260] Next, before the player selected by the selection means makes a move, the second device unit DV2 calculates the expected winning probability, i.e. the win rate value, when making a move at each possible point, based on the fourth artificial intelligence engine 430, reflecting the player's playing style (step: S82).

[0261] The second device unit DV2 can detect from the calculated win rate values ​​a predetermined number of possible moves with the highest win rate when the analyzed player makes a move (step: S83). The second device unit DV2 can display the detected possible moves as recommended moves on the Go board displaying the Go game screen for the user to identify (step: S84).

[0262] For example, assuming the user selects the first player as the analysis target player through a selection method, the second device unit DV2 can set the first player as the analysis target player. Then, when it is the first player's turn to make a move, the second device unit DV2 can calculate the win rate value corresponding to the first player's move at each possible move point in the current state, based on the fourth artificial intelligence engine 430. From the calculated win rate values, the second device unit DV2 detects the possible move points with the highest win rate value when the first player makes a move, a predetermined number (e.g., N, where N is a predetermined integer of 2 or higher), and displays the detected possible move points as recommended move points on the Go game board for the user to identify.

[0263] For example, in the first player's turn (before their next move), the predetermined number of points with the highest win rates detected on the Go game display screen, such as six possible moves, are represented as six recommended moves for intuitive user identification. For example, these six recommended moves could be displayed on the board using stones of a different color than black and white stones. Figure 26 In the example shown, recommended placement points can be represented by red pieces. Recommended placement points can be represented in various visually recognizable forms, such as dashed lines in the shape of pieces on the board, or pieces with special patterns.

[0264] On the other hand, according to one embodiment of the present invention, there are multiple recommended move points. Based on the win rate value corresponding to each recommended move point, the relative levels of win rates among the recommended move points are distinguished for identification. That is, multiple recommended move points can be displayed differently based on the calculated win rate value. For example, they can be displayed with different colors, shapes, sizes, etc., based on the calculated win rate value.

[0265] For example, before the first player, designated as the target of analysis, makes their move, on the Go board representing the game, six recommended moves with the highest win rate for the first player are identified. These recommended moves are represented using various colors based on their corresponding win rates.

[0266] For example, as previously mentioned Figure 27The six recommended moves (six places with the highest win rate) are as follows: the two recommended moves with the highest win rate can be represented by a first color, such as red. Furthermore, the two recommended moves with the highest win rate, excluding those represented by red, can be represented by a second color, such as blue. Additionally, the two recommended moves with the highest win rate, excluding those represented by red and blue, can be represented by a third color, such as green. This color information based on the win rate is displayed on the screen using icons or prompts, allowing users to intuitively identify the color of the recommended moves based on the win rate.

[0267] Furthermore, when the second device unit DV2 places a piece at the recommended placement point, it displays the expected winning probability (i.e., the win rate value) in relation to the displayed recommended placement point for identification. For example, such as... Figure 27 As shown, the second device unit DV2 can simultaneously display the corresponding win rate value when displaying the recommended move point. For example, the win rate value can be displayed inside the piece representing the recommended move point or in its vicinity.

[0268] As described above, according to a preferred embodiment of the present invention, when it is the turn of the player designated as the analysis target to make a move, a predetermined number of recommended moves can be displayed on the Go board in various ways. Therefore, when watching a Go game, the user can select a player of interest or one they wish to analyze. Each time it is the turn of the player to be analyzed, the system can confirm recommended moves with high win rates in real time and also confirm the actual point where the player to be analyzed places their move, thereby enhancing the viewing experience of the Go game. Alternatively, when the user needs to provide Go instruction or guidance to a player, and thus requires advice or guidance during the game, suggestions or guidance can be provided based on the real-time displayed recommended moves, making it an excellent educational aid.

[0269] On the other hand, the second processor unit 475 of the second device unit DV2 can set the start and end times of the replay based on the user interface. When a replay request signal is received, the second processor unit 475 can display the replay screen from the start time to the end time, and show the changing win rate value for each move. Therefore, even after the Go game ends, the user can still view the changing win rate value for each move, thereby enhancing the enjoyment of the replay.

[0270] As described above, according to a preferred embodiment of the present invention, when a user plays Go in real-world conditions or watches Go games on screen, the device can acquire information about the players in the Go game. Based on artificial intelligence that learns about the players' playing styles and skill levels, the device generates necessary information for the Go game, such as real-time updates of predicted win rates for both players, win rate changes, and recommended moves for the corresponding analyzed player. Therefore, this not only enhances the enjoyment of watching Go games but also serves as a highly valuable educational tool for users learning Go.

[0271] Figure 34 A system structure block diagram for implementing the method according to a preferred fifth embodiment of the present invention.

[0272] like Figure 34 As shown, the fifth server 500 can be linked with the database 50 and the user terminal UR3 via a communication network. The user terminal UR3 can be a network-connected computing device such as a user's smartphone, personal computer terminal, laptop, or tablet computer.

[0273] The fifth server 500 can access user terminal UR3 and provide viewing services for Go games between players to the accessed user terminal UR3 through broadcasting or streaming, and can display various information related to Go games on the screen of user terminal UR3.

[0274] For example, the fifth server 500 can respond to a request from the user terminal UR3 and, based on streaming transmission to the user terminal UR3, display a video of the Go game between the first player and the second player on the screen of the user terminal UR3; on one side of the screen, a first selection means is displayed to select a first information display mode, which is used to display first information related to the Go game; if the first information display mode is selected, then during the Go game, whenever a move is made, the fifth artificial intelligence engine 530 calculates the final win rate prediction value, i.e., the win rate value, of at least one player; and the information related to the calculated win rate value is used as the first information and displayed in real time on the screen displaying the Go game.

[0275] Figure 35 for Figure 34 The diagram shows the detailed structure of the fifth server 500.

[0276] Figure 35 As shown, the fifth server 500 may include a game display unit 572, a third processor unit 574, and a fifth artificial intelligence engine 530, etc. These units can interact with each other and can also interact with the database 50. The detailed functions of each unit and their interaction processes will be described in detail below.

[0277] Figure 36 For illustrative purposes Figures 34 to 35 The flowchart of the operation process of the fifth server 500 shown illustrates the process of displaying Go game-related information according to the fifth embodiment of the present invention.

[0278] See Figures 34 to 36 The game display unit 572 of the fifth server 500 can respond to the request of the user terminal UR3 and display the image of the Go game between the first player and the second player on the screen of the user terminal UR3 based on the streaming transmission of the image to the user terminal UR3 (step: S91).

[0279] The third processor unit 574 of the fifth server 500 displays a first selection means for selecting a first information display mode on one side of the screen. The first information display mode is used to display first information related to the Go game (step: S92).

[0280] If the user selects the first information display mode through the first selection means, the fifth server 500 can overlay various information related to the Go game generated based on the fifth artificial intelligence engine 530 onto the Go game image between the first player and the second player in the form of graphic objects.

[0281] For example, when the first information display mode is selected, the fifth server 500 can calculate the predicted final winning probability value, i.e., the win rate value, of at least one player based on the fifth artificial intelligence engine 530 whenever a move is made in the Go game (step: S93). The fifth artificial intelligence engine 530 can learn and generate information as described in the third artificial intelligence engine 330 or the fourth artificial intelligence engine 430 in the foregoing embodiments.

[0282] For example, the fifth server 500 can calculate the win rate of the first player in the current situation after a move has been made, based on the fifth artificial intelligence engine 530. The win rate represents the probability that the first player will ultimately win in the current situation, and as mentioned earlier, it can be expressed as a percentage. The fifth server 500 can also calculate the win rate of the second player based on the first player's win rate. For example, since the sum of the win rates of the two players should be 100%, the second player's win rate can be 100 minus the first player's win rate.

[0283] Next, the fifth server 500 can use the real-time calculated information related to the win rate value as the first information and display it in real time on the screen of the user terminal UR3 that displays the Go game screen (step: S94).

[0284] For example, the user terminal UR3 displays a Go board game screen, showing the colors of the first and second players' pieces, the current state of the game, etc. The fifth server 500 can display the win rates of the first and second players respectively on the screen.

[0285] For example, a win rate value such as "current win rate 71%" can be displayed near the graphic object representing the first player using black, and a win rate value such as "current win rate 29%" can be displayed near the graphic object representing the second player using white. Therefore, users can monitor the win rates of both players in real time while watching a Go game, greatly enhancing the viewing experience by observing the dynamically changing win rate values ​​after each move. Although this embodiment exemplifies the representation of win rate values ​​in numerical form, according to an embodiment of the present invention, graphic objects whose shape, color, or size changes with the numerical value, such as emoticons or graphics, can also be used for representation.

[0286] Server 500 can display the most recent move on the Go game screen. For example, as previously mentioned... Figure 19 As exemplarily described, the fifth server 500 can use a red border to indicate the black piece of the first player who recently made a move in the current situation, which can be identified by the user; it can also indicate the place of the most recently made move by making the most recently made piece flash, or by adjusting its color to a special hue that distinguishes it from other pieces.

[0287] On the other hand, according to one embodiment of the present invention, the fifth server 500 may also, after a move is made, overlay and display the predicted final winning probability value (i.e., the win rate value) of the player who made the move, calculated for that move, onto the point on the chessboard where the move occurred. When the next move occurs, the fifth server 500 may clear the win rate value represented by the previous move and overlay and display the win rate value corresponding to the situation of the next move onto the point of the move.

[0288] For example, as previously mentioned Figure 20 The same concept applies here. Assuming the current situation is as follows: after the first player places a piece at a specific point, the fifth server 500 can calculate the win rate value in the current situation and display this value overlaid on the piece at the most recent placement point. Therefore, users can intuitively identify the most recent placement point and the change in win rate caused by their placement behavior each time they make a move.

[0289] On the other hand, according to an embodiment of the present invention, after a move is made, the fifth server 500 compares the win rate value of the player who made the move, calculated for the move, with the win rate value before the move; based on the comparison of the win rate values, the server displays on the screen whether the win rate after the move has increased or decreased compared to the win rate before the move, so that the user can identify it.

[0290] For example, as previously mentioned Figure 21 The same concept applies here: after a move is made, the area near the player's graphic object can display graphic objects indicating whether the player's win rate has increased or decreased. For example, assuming the first player's win rate has increased after making a move with black, the fifth server 500 can display a graphic object indicating the increase in win rate compared to before the move in the area near the first player's graphic object.

[0291] The graphic object can display an upward arrow in a first color when the win rate increases after a move, and can represent text describing the increase in the win rate (e.g., "UP"). Conversely, it can display a downward arrow in a second color when the win rate decreases after a move, and can represent text describing the decrease in the win rate (e.g., "DOWN").

[0292] On the other hand, the fifth server 500 can also compare the win rate of the player who made the move with the win rate of the opponent after a move is made; based on the comparison of the win rate, the server can display on the screen whether the player who made the move has a higher probability of winning or a higher probability of losing, so that the user can identify the probability.

[0293] For example, as previously mentioned Figure 22 The same concept applies to the fifth server 500, which can adjust the size of the graphical object used to identify players based on win rate values. For example, when a particular player's win rate is higher than that of the opponent in the current situation, the fifth server 500 can display the graphical object used to identify that player at a relatively larger size than the graphical object used to identify the opponent.

[0294] For example, if after the first player makes a move, the first player's winning probability is higher than the second player's in the current situation, then when displaying the Go game, the graphic object used to identify the first player is displayed at a larger size than the graphic object used to identify the second player.

[0295] Furthermore, Server 500 can display text on the graphical objects used to identify players, indicating whether their current win rate is higher or lower than their opponent's. For example, within or near the graphical object identifying the first player (whose win rate is relatively high in the current situation), the text "High Win Rate" can be displayed to indicate that the second player has a higher win rate than their opponent. Conversely, within or near the graphical object identifying the second player, the text "Low Win Rate" can be displayed to indicate that the second player has a lower win rate than the first player.

[0296] In addition, as previously referred to Figure 23The fifth server 500 can adjust the shape of the graphical object used to identify players based on their win rate. For example, when a specific player's win rate is higher than that of their opponent in the current situation, the fifth server 500 can insert emoticons that typically have a positive meaning (e.g., happy expressions or behaviors, sunny weather, etc.) into the graphical object used to identify that player. Conversely, emoticons that typically have a negative meaning (e.g., sad or angry expressions or behaviors, or rainy weather, etc.) can be inserted into the graphical object used to identify players whose win rate is lower than their opponent's.

[0297] For example, assuming that after the first player makes a move, their winning probability is higher than the second player's, an emoticon, such as a "smiling emoticon," can be inserted onto the graphic object identifying the first player when displaying the Go game. Conversely, an emoticon, such as a "sad emoticon," can be inserted onto the graphic object identifying the second player, whose winning probability is relatively lower.

[0298] On the other hand, the third processor unit 574 of the fifth server 500 can provide the function of setting either of the two players playing Go as the analysis target player and displaying the recommended move points as described in the previous first to fourth embodiments for the analysis target player.

[0299] Figure 37 This is a flowchart illustrating the process of setting up an analysis target player and displaying recommended move points for that player.

[0300] like Figure 37 As shown, the fifth server 500 displays a second selection means on the other side of the screen, which allows selection of a second information display mode. The second information display mode is used to display second information related to the Go game (step: S101).

[0301] When the second information display mode is selected, the fifth server 500 can set the analysis target player between the first player and the second player (step: S102). For example, the fifth server 500 can display a selection method that allows selection of either the first player or the second player to play the Go game. The user can select the icon corresponding to the player to be analyzed using the selection method by touch or other means. Then the fifth server 500 can set the player selected by the selection method as the analysis target player. Related examples are as previously mentioned. Figure 25 The same concept as described above.

[0302] Next, before the player being analyzed makes a move, the fifth server 500, based on the fifth artificial intelligence engine 530, calculates the expected winning probability, i.e., the win rate value, corresponding to each possible move point (step: S103); among the calculated win rate values, it detects the predetermined number of possible moves with the highest win rate value when the player being analyzed makes a move (step: S104); and displays the detected possible moves as recommended moves on the Go game board so that the user can identify them (step: S105).

[0303] For example, assuming the user selects the first player as the analysis target through a selection process, the fifth server 500 can set the first player as the analysis target. Then, when it's the first player's turn to make a move, the fifth server 500 can calculate the win rate of the first player at each possible move point in the current state, based on the fourth artificial intelligence engine 530. From the calculated win rate values, the fifth server 500 detects a predetermined number (e.g., N, where N is a predetermined integer greater than 2) of possible moves with the highest win rate when the first player makes a move, and displays these detected possible moves as recommended moves on the Go game board for the user to identify.

[0304] For example, in the first player's turn (before their next move), the predetermined number of points with the highest win rates detected on the Go game display screen, such as six possible moves, are represented as six recommended moves for intuitive user identification. For example, these six recommended moves could be displayed on the board using stones of a different color than black and white stones. Figure 26 In the example shown, recommended placement points can be represented by red pieces. Recommended placement points can be represented in various visually recognizable forms, such as dashed lines in the shape of pieces on the board, or pieces with special patterns.

[0305] On the other hand, according to one embodiment of the present invention, there are multiple recommended move points. Based on the win rate value corresponding to each recommended move point, the relative levels of win rates among the recommended move points are distinguished for identification. That is, multiple recommended move points can be displayed differently based on the calculated win rate value. For example, they can be displayed with different colors, shapes, sizes, etc., based on the calculated win rate value.

[0306] For example, before the first player, designated as the target of analysis, makes their move, on the Go board representing the game, six recommended moves with the highest win rate for the first player are identified. These recommended moves are represented using various colors based on their corresponding win rates.

[0307] For example, as previously mentioned Figure 27The six recommended moves (six places with the highest win rate) are as follows: the two recommended moves with the highest win rate can be represented by a first color, such as red. Furthermore, the two recommended moves with the highest win rate, excluding those represented by red, can be represented by a second color, such as blue. Additionally, the two recommended moves with the highest win rate, excluding those represented by red and blue, can be represented by a third color, such as green. This color information based on the win rate is displayed on the screen using icons or prompts, allowing users to intuitively identify the color of the recommended moves based on the win rate.

[0308] Furthermore, when the fifth server 500 places a piece at the recommended placement point, it displays the expected winning probability (i.e., the win rate value) in relation to the displayed recommended placement point for identification. For example, as... Figure 27 As shown, the fifth server 500 can simultaneously display the corresponding win rate value when displaying recommended move points. For example, the win rate value can be displayed inside the piece representing the recommended move point or in its vicinity.

[0309] As described above, according to a preferred embodiment of the present invention, in the second information display mode, when it is the turn of the player designated as the analysis target to make a move, a predetermined number of recommended move points can be displayed on the Go game display screen in various ways. Therefore, when watching a Go game, the user can select a player of interest or one they wish to analyze. Whenever it is the turn of the player to be analyzed, the system can confirm the recommended move points with high winning probabilities in real time, and also confirm in real time where the player actually places their move, thereby enhancing the viewing experience of the Go game. Alternatively, when the user needs to provide Go instruction or guidance to the player, and thus needs to offer suggestions or guidance during the Go game, suggestions or guidance can be provided based on the real-time displayed recommended move points, making it an excellent educational aid.

[0310] On the other hand, the third processor unit 574 of the fifth server 500 can set the start and end times of the replay based on the user interface. When a replay request signal is received, the third processor unit 574 can display the replay screen from the start time to the end time, and show the changing win rate value for each move. Therefore, even after the Go game ends, the user can still view the changing win rate value for each move, thereby enhancing the enjoyment of the replay.

[0311] Figure 38 This is a hardware structure block diagram illustrating an apparatus for performing the method according to the first to fifth embodiments of the present invention.

[0312] The device can be made of, for example Figure 38 The computing system shown is configured as follows. Figure 38As shown, the computing device 1000 may include a flash memory 1010, a processor 1020, RAM 1030, an input device 1040, and a power supply 1050. Additionally, the flash memory 1010 may include a memory device 1011 and a memory controller 1012. Furthermore, although not shown in... Figure 38 The text states that the computing device 1000 may also include a port that can communicate with video cards, sound cards, memory cards, USB devices, etc., or with other electronic instruments.

[0313] The computing system 1000 can be implemented via a personal computer, or via portable electronic devices such as laptops, mobile phones, PDAs (personal digital assistants), and cameras.

[0314] Processor 1020 can perform specific calculations or tasks. Depending on the embodiment, processor 1020 may be a microprocessor or a central processing unit (CPU). Processor 1020 can communicate with RAM 1030, input / output devices 1040, and flash memory 1010 via buses 1060, including address bus, control bus, and data bus.

[0315] According to one embodiment, the processor 1020 may also be connected to an expansion bus such as a Peripheral Component Interconnect (PCI) bus.

[0316] RAM 1030 can store data required for the operation of computing system 1000. For example, it can be used as random access memory (RAM) 1030 of any type, including DRAM, mobile DRAM, SRAM, PRAM, FRAM, MRAM, RRAM, etc.

[0317] Input / output device 1040 may include input devices such as keyboard, buttons, and mouse, and output devices such as printer and monitor. Power supply device 1050 provides the operating voltage required for the operation of computing system 1000.

[0318] The processor 1020 can be configured to execute the method according to a preferred embodiment of the present invention. More specifically, the operation of the processor may follow a method for signing online Go game pieces or an information display method for online Go games according to an embodiment of the present invention.

[0319] According to the method of the invention, it can be implemented in computer-readable code on a computer-readable recording medium. The computer-readable recording medium includes all types of recording media that store data that can be interpreted by a computer system. For example, it may include ROM (Read Only Memory), RAM (Random Access Memory), magnetic tape, magnetic disk, flash memory, optical data storage device, etc. Furthermore, the computer-readable recording medium can be distributed across computer systems connected via a computer communication network and can be distributed to store and execute computer-readable code.

[0320] The accompanying drawings and embodiments are for illustrative purposes only and are not intended to limit the invention. Those skilled in the art should understand that modifications, variations, or equivalent substitutions can be made to the invention without departing from its spirit and scope, and all such modifications and substitutions should be covered within the scope of the claims of the invention.

[0321] Specifically, the features described herein may be executed within digital electronic circuits or computer hardware, firmware, or a combination thereof. For example, the features may be executed in a computer program product implemented in a machine-readable storage device, to be executed by a programmable processor. Alternatively, the features may be executed by a programmable processor of a program that executes instructions of a program by running on input data and generating output to perform the functions of the embodiments described herein. The features may be executed within more than one computer program, which may be executed on a programmable system comprising at least one programmable processor, at least one input device, and at least one output device combined for receiving and transmitting data and instructions from and to a data storage system. A computer program includes a set of instructions used directly or indirectly within a computer to perform a specific operation on a specified result. A computer program is written in one of the programming languages, including compiled or parsed languages, and may be used as a module, element, subroutine, or other unit suitable for use in a computer environment, or as a program that can operate independently.

[0322] A processor for executing instructions, such as a multiprocessor including general-purpose or special-purpose microprocessors, individual processors, or other types of computers. Additionally, a storage device for implementing the computer program instructions and data for implementing the aforementioned features includes semiconductor memory devices such as EPROM, EEPROM, and flash memory devices, and magnetic devices such as internal hard disks and erasable disks, as well as non-volatile memories such as optical discs, CD-ROMs, and DVD-ROMs. The processor and memory can be integrated within or added via an ASIC (application-specific integrated circuit).

[0323] The embodiments described herein are merely illustrative of a series of functional modules and are not intended to limit the invention. Those skilled in the art should understand that modifications, variations, or equivalent substitutions can be made to the invention without departing from its spirit and scope, and all such modifications and substitutions should be covered within the scope of the claims of the invention.

[0324] The combination of the foregoing embodiments is not limited to the foregoing embodiments. Various combinations of the foregoing embodiments may be provided in addition to the foregoing embodiments, depending on the implementation and / or needs.

[0325] In the foregoing embodiments, the method is described based on a flowchart of a series of steps or modules. However, the present invention is not limited to the order of the steps, and some steps may occur differently from or in a different order than those described above, or simultaneously. Furthermore, those skilled in the art should understand that the method is not limited to the steps in the flowchart and may include other steps, or one or more steps in the flowchart may be deleted without affecting the scope of the present invention.

[0326] The foregoing embodiments include examples of various forms. While it is impossible to describe all possible combinations representing various aspects, those skilled in the art will understand that other combinations may exist. Therefore, the invention includes all other substitutions, modifications, and alterations that fall within the scope of the following claims.

[0327] The embodiments described are for illustrative purposes only and are not intended to limit the invention. Those skilled in the art should understand that modifications, variations, or equivalent substitutions can be made to the invention without departing from its spirit and scope, and all such modifications and substitutions should be covered within the scope of the claims. Therefore, any future changes to the embodiments of the invention should not exceed the technical scope of the invention.

Claims

1. A method for providing information about Go game content, executed by a server, characterized in that, include: In response to a request from a user terminal, and based on streaming to the user terminal, a video of the Go game between the first player and the second player is displayed on the screen of the user terminal. A first selection method is displayed on one side of the screen, which allows selection of a first information display mode. The first information display mode is used to display first information related to the Go game. If the first information display mode is selected, then during the Go game, whenever a move is made, the final win rate prediction value, i.e. the win rate value, of at least one player is calculated based on the artificial intelligence engine. as well as Information related to the calculated win rate value is used as the first information and displayed in real time on the screen displaying the Go game.

2. The method for providing information on Go game content according to claim 1, characterized in that: The real-time display step includes: After a move is made, the predicted final winning probability (i.e., the win rate) of the player who made that move is superimposed and displayed on the point where the move occurred on the chessboard; and If the next move occurs, the win rate value of the previous move that was already displayed is removed.

3. The method for providing information on Go game content according to claim 1, characterized in that: The real-time display step includes: After a move is made, the win rate of the player who made that move is compared with their win rate before the move; and Based on the comparison of the win rate value, the screen displays whether the win rate after the move has increased or decreased compared to the win rate before the move, so that the user can make an identification.

4. The method for providing information on Go game content according to claim 1, characterized in that: The real-time display step includes: After a move is made, the win rate of the player who made that move is compared with the win rate of the opponent; and Based on the comparison of the win rate values, the screen displays whether the player who made the move has a higher probability of winning or a higher probability of losing, so that the user can make an identification.

5. The method for providing information on Go game content according to claim 1, characterized in that, Also includes: On the other side of the screen, a second selection means is displayed to select a second information display mode, which is used to display second information related to the Go game; If the second information display mode is selected, then the analysis target player is set between the first player and the second player; Before the player being analyzed makes their move, the AI ​​engine calculates the expected winning probability (win rate) for each possible move at each available point. Among the calculated win rate values, the predetermined number of possible placement points with the highest win rate value are detected when the analyzed player places a piece. as well as The detected possible moves are displayed as recommended moves on the Go board so that users can make their own choices.

6. The method for providing information on Go game content according to claim 5, characterized in that, Also includes: When placing a piece at the recommended placement point, the expected winning probability, i.e., the win rate value, is displayed in relation to the recommended placement point for identification.

7. The method for providing information on Go game content according to claim 5, characterized in that: There are multiple recommended moves. Based on the win rate value corresponding to each recommended move, the relative win rates of the recommended moves are displayed to distinguish and identify them.

8. The method for providing information on Go game content according to claim 1, characterized in that, Also includes: Based on the user interface, set the start and end times for reviewing the Go game. When a replay request signal is received, the replay screen from the start time to the end time is displayed, along with the change in win rate value for each move.

9. A device for providing information on Go game content, characterized in that, include: The game display unit, in response to a request from the user terminal, displays a video of the Go game between the first player and the second player on the screen of the user terminal based on streaming transmission to the user terminal; as well as The processor unit displays a first selection means on one side of the screen to select a first information display mode, the first information display mode being used to display first information related to the Go game; If the first information display mode is selected, then during the Go game, whenever a move is made, the final win rate prediction value, i.e. the win rate value, of at least one player is calculated based on the artificial intelligence engine. Information related to the calculated win rate value is used as the first information and displayed in real time on the screen displaying the Go game.

10. The information providing device for Go game content according to claim 9, characterized in that: After a move is made, the processor calculates the final winning probability prediction value (i.e., the win rate value) for the player who made the move and displays it superimposed on the point where the move occurred on the chessboard; if the next move occurs, the win rate value of the previous move that has been displayed is removed.

11. The information providing device for Go game content according to claim 9, characterized in that: After a move is made, the processor compares the win rate calculated for the player who made the move with the win rate before the move. Based on the comparison, it displays on the screen whether the win rate after the move has increased or decreased compared to the win rate before the move, so that the user can identify the difference.

12. The information providing device for Go game content according to claim 9, characterized in that: After a move is made, the processor compares the win rate of the player who made the move with the win rate of the opponent. Based on the comparison of the win rates, it displays on the screen whether the player who made the move has a higher probability of winning or a higher probability of losing, so that the user can make an identification.

13. The information providing device for Go game content according to claim 9, characterized in that: The processor displays a second selection method on the other side of the screen, which allows selection of a second information display mode. The second information display mode is used to display second information related to the Go game. If the second information display mode is selected, an analysis target player is set between the first player and the second player. Before the set analysis target player makes a move, the expected winning probability, i.e., the win rate value, is calculated based on an artificial intelligence engine when a move is made at each possible move point. Among the calculated win rate values, a predetermined number of possible move points with the highest win rate value are detected when the analysis target player makes a move. The detected possible move points are displayed as recommended move points on the Go game board so that the user can identify them.

14. The information providing device for Go game content according to claim 13, characterized in that: When placing a piece at the recommended placement point, the expected winning probability, i.e., the win rate value, is displayed in relation to the recommended placement point for identification.

15. The information providing device for Go game content according to claim 13, characterized in that: There are multiple recommended moves. Based on the win rate value corresponding to each recommended move, the relative win rates of the recommended moves are displayed to distinguish and identify them.

16. The information providing device for Go game content according to claim 9, characterized in that: The processor unit, based on the user interface, sets the start and end times for reviewing the Go game; when a review request signal is received, it displays the review screen from the start to the end time, and displays the changing win rate value for each move.

17. A method for displaying Go game information, executed by a Go game information display device, characterized in that, include: Extract first player information containing the first player's identification information, playing style information, and playing strength information, and second player information containing the second player's identification information, playing style information, and playing strength information from the database respectively; Use a camera to capture images of the Go game between the first player and the second player; Based on the acquired images, a Go game between the first player and the second player is displayed on the screen; When playing the Go game, an artificial intelligence engine is used to calculate in real time the predicted final winning probability value, i.e., the win rate value, of at least one player in the current game state based on the information of the first player and the information of the second player. as well as Information related to the calculated win rate value is displayed on one side of the screen; Here, based on the win rate value calculated in real time, the information related to the win rate value displayed at each move is updated.

18. The method for displaying Go game information according to claim 17, characterized in that, Also includes: The camera is used to capture images of codes located near the Go board where the Go game is played or displayed on the screen of the Go game, in order to obtain code images. as well as By decoding the obtained code image, at least one of the first player information and the second player information stored in the database can be accessed.

19. The Go game information display method according to claim 17, characterized in that, Also includes: The camera is used to capture the faces of at least one of the first player and the second player to obtain facial images; as well as The database is queried based on the acquired facial image to access at least one of the first player information and the second player information stored in the database.

20. The Go game information display method according to claim 17, characterized in that: In the database, each player's playing style information and external strength information are stored in relation to the corresponding player identification information; The artificial intelligence engine is trained based on big data of game information, including information on players' playing styles and skill levels, to predict win rates under different game conditions.

Images