Computer system

Abstract

To provide a new technology capable of obtaining various evaluations for a game play.SOLUTION: A game system can analyze and evaluate given game situation information irrespective of the amount of information, and can provide a player with evaluation information based on a result of the analysis and evaluation of the information on the game situation in which the player has played a game. Prior to the provision of the evaluation information, the game system selects selection game situation information out of the player's game situation information. The analysis and evaluation is performed on the basis of the selection game situation information, and the evaluation information is generated on the basis of the analysis and evaluation result based on the player's game situation information, and is provided to the player.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a computer system and the like. [Background technology]

[0002] For video game players, receiving evaluations for their gameplay is an important element that adds excitement to the game. Evaluations are also meaningful as a factor in improving skill. Evaluations can be meaningful whether they are mechanically obtained by a computer system or by other players / users.

[0003] For example, evaluation comments posted by viewers on a live broadcast of a gameplay can boost the players' enthusiasm and create a lively atmosphere for the broadcast. Furthermore, evaluation comments can also help players improve their playing skills on the spot by referring to them.

[0004] For example, Patent Document 1 discloses a technology that uses an AI model (artificial intelligence) to analyze the results of a user's progress in a competition and present the user with instructional content to help them progress in the competition to their advantage. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Publication No. 2022-26503 Summary of the Invention [Problem to be solved by the invention]

[0006] However, in conventional games such as those described in Patent Document 1, the prevailing view was that one AI model was sufficient. However, when one fixed AI model was used, the output of the AI ​​model became uniform, which raised concerns about the game becoming monotonous. Similar issues arose when conventional technology was applied to evaluate gameplay.

[0007] The problem to be solved by the present invention is to provide a new technique that can obtain diverse evaluations of game play. [Means for solving the problem]

[0008] The first invention for solving the above problems comprises an analysis and evaluation means for analyzing and evaluating given game situation information (for example, the primary analysis and evaluation 10, the secondary analysis and evaluation 20 in FIG. 2, the analysis and evaluation unit 230 in FIG. 10, step S92 in FIG. 13, and step S118 in FIG. 14); evaluation information provision control means (e.g., provision of primary evaluation information 16 and provision of secondary evaluation information 26 in FIG. 2, evaluation information provision control unit 240 in FIG. 10, step S96 in FIG. 13, step S122 in FIG. 14) that causes the analysis and evaluation means to analyze and evaluate game situation information in which a player has played a game, and controls the provision of evaluation information to the player based on the analysis and evaluation results; information selection means for selecting selected game situation information from the game situation information of the player (for example, the primary selected game situation information 751, the secondary selected game situation information 752 in FIG. 2, the information selection unit 224 in FIG. 10, step S90 in FIG. 13, and step S116 in FIG. 14); wherein the evaluation information provision control means provides the player with the evaluation information based on the analysis and evaluation result obtained by having the analysis and evaluation means analyze and evaluate the selected game situation information.

[0009] If we want to evaluate a variety of gameplay aspects, we will inevitably need multiple analytical and evaluation methods with different analytical and evaluation standards. The computer system of the first invention can analyze and evaluate a game based on selected game status information selected from among game status information, and provide players with evaluation information based on the analysis and evaluation results. Therefore, even if a single game is equipped with an AI model that evaluates games based on game status information, players can efficiently obtain diverse evaluations of their gameplay. Diverse evaluations prevent games from becoming monotonous, allowing players to enjoy the same game for a longer period of time from various perspectives.

[0010] A second invention comprises an analysis and evaluation means for analyzing and evaluating given game situation information, and an evaluation information provision control means for controlling the analysis and evaluation means to analyze and evaluate game situation information in which a player has played a game, and to provide evaluation information based on the analysis and evaluation results to the player, The analysis and evaluation means has a plurality of individual analysis and evaluation means (for example, the primary individual analysis and evaluation AI model 6a to the primary individual analysis and evaluation AI model 6c in FIG. 4), The evaluation information provision control means is a computer system that has an individual selection means (e.g., second individual selection unit 241 in Figure 10, step S34 in Figure 13, YES in step S50 → step S52 in Figure 15) that selects a selected individual analysis evaluation means from the plurality of individual analysis evaluation means, and provides the player with the evaluation information based on the analysis evaluation results by the selected individual analysis evaluation means.

[0011] Furthermore, a third invention is a computer system comprising an analysis and evaluation means (for example, primary individual analysis and evaluation AI model 6a to primary individual analysis and evaluation AI model 6c in Figure 4) that analyzes and evaluates given game situation information, and an evaluation information provision control means that causes the analysis and evaluation means to analyze and evaluate the game situation information in which a player has played the game, and controls the provision of evaluation information based on the analysis and evaluation results to the player, wherein the analysis and evaluation means has a plurality of individual analysis and evaluation means, and individual selection means (for example, individual selection unit 236 in Figure 10, step S60 and thereafter in Figure 15) that selects a selected individual analysis and evaluation means from the plurality of individual analysis and evaluation means based on the game situation information, and the selected individual analysis and evaluation means analyzes and evaluates the game situation information.

[0012] If we want to evaluate a variety of gameplay aspects, we will inevitably need multiple analytical and evaluation methods with different analytical and evaluation standards. The computer system of the second or third invention can select and operate an individual analysis and evaluation means for performing analysis and evaluation from among multiple individual analysis and evaluation means. Therefore, even if a single game is equipped with an AI model that performs evaluation based on game situation information, players can efficiently obtain diverse evaluations of their gameplay. Furthermore, since the evaluation information provided to the player can change depending on which individual analysis and evaluation means is selected, it prevents players from becoming bored and allows them to enjoy the same game for a longer period of time from various perspectives.

[0013] A fourth aspect of the present invention is a game system comprising: analysis and evaluation means for analyzing and evaluating given game situation information; and evaluation information provision control means for controlling the analysis and evaluation means to analyze and evaluate game situation information in which a player has played a game, and to provide evaluation information based on the analysis and evaluation results to the player, The analysis and evaluation means is a computer system that has individual analysis and evaluation means for each layer from the input layer to the output layer (for example, the individual analysis and evaluation unit 233 on the input layer side and the individual analysis and evaluation unit 234 on the output layer side in Figure 10), and performs hierarchical analysis and evaluation in which the individual analysis and evaluation results of the individual analysis and evaluation means on the input layer side are individually analyzed and evaluated by the individual analysis and evaluation means on the output layer side.

[0014] According to the fourth aspect of the present invention, the computer system can obtain further analytical evaluation results for the output layer by using the analytical evaluation results for the input layer as input for the output layer. Therefore, even when a single game title is equipped with an AI model that evaluates based on game situation information, players can efficiently obtain diverse evaluations of their gameplay. Furthermore, the computer system can provide players with hierarchical evaluation information. Hierarchical evaluations give players a variety of perspectives, preventing them from becoming bored and allowing them to enjoy the same game for a long time.

[0015] A fifth invention is the computer system described above, wherein the analysis and evaluation means is capable of analyzing and evaluating the game situation information based on a given selected evaluation item from a plurality of evaluation items; Further provided is an evaluation item selection means (for example, the evaluation item selection unit 220 in FIG. 10, steps S52, S62, and S76 in FIG. 15) for selecting the selected evaluation item from the plurality of evaluation items, The evaluation information provision control means is a computer system that controls the analysis and evaluation means to analyze and evaluate the game situation information based on the selected evaluation items.

[0016] According to the fifth aspect of the present invention, the computer system can select a selection evaluation item from a plurality of evaluation items and perform analysis and evaluation based on the selection evaluation item, thereby improving the diversity of evaluation information provided to players.

[0017] A sixth invention is a computer system in which the evaluation item selection means selects the selected evaluation item based on the operation of the player (for example, step S14 in Figure 13, steps S50 to S52 in Figure 15).

[0018] According to the sixth aspect of the present invention, the computer system can select the selection evaluation item based on the player's operation.

[0019] A seventh aspect of the present invention is the computer system described above, wherein the analysis and evaluation means is capable of analyzing and evaluating the game situation information based on a given selected evaluation item from a plurality of evaluation items; an evaluation item selection means for selecting the selected evaluation item from the plurality of evaluation items; The individual selection means further includes the selected individual analysis and evaluation means based on the selected evaluation items, The evaluation information provision control means is a computer system that controls the selected individual analysis and evaluation means to perform analysis and evaluation based on the selected evaluation items.

[0020] According to the seventh aspect of the present invention, the computer system can select a selection evaluation item from a plurality of evaluation items and perform analysis and evaluation based on the selected evaluation item, thereby improving the diversity of evaluation information provided to players.

[0021] An eighth invention is a computer system in which, in the above-mentioned computer system, the evaluation item selection means selects the selected evaluation item based on the operation of the player (for example, step S14 in Figure 13, steps S50 to S52 in Figure 15).

[0022] According to the eighth aspect of the present invention, the computer system can select the selection evaluation item based on the player's operation.

[0023] A ninth invention is the computer system described above, wherein the individual selection means selects the selected individual analysis and evaluation means based on the operation of the player (for example, step S14 in FIG. 13, steps S50 to S52 in FIG. 15).

[0024] According to the ninth aspect of the present invention, the computer system can select the selected individual analytical evaluation means based on the operation of the player.

[0025] A tenth invention is a computer system, in which the above-mentioned computer system further comprises an evaluation item selection means for selecting a selected evaluation item from a plurality of evaluation items, and the analysis and evaluation means analyzes and evaluates the game situation information based on the selected evaluation item.

[0026] According to the tenth aspect of the present invention, the computer system can select a selection evaluation item from a plurality of evaluation items and perform analysis and evaluation based on the selected evaluation item, thereby improving the diversity of evaluation information provided to players.

[0027] An eleventh aspect of the present invention is the computer system, wherein the evaluation item selection means selects the selected evaluation item based on the game situation information.

[0028] According to the eleventh aspect, the computer system can select the selection evaluation item based on the game situation information.

[0029] A twelfth aspect of the present invention is the computer system described above, wherein the analysis and evaluation means performs the analysis and evaluation using an AI model for analyzing and evaluating the game situation information.

[0030] According to the twelfth aspect of the present invention, the computer system can perform analysis and evaluation using an AI model.

[0031] A thirteenth invention is a computer system, further comprising: a feedback information receiving means (e.g., the feedback information receiving control unit 242 in FIG. 10, steps S98 and S124 in FIG. 14) for receiving feedback information based on the player's operation in response to the evaluation information; and an AI model updating means (e.g., the AI ​​model updating unit 244 in FIG. 10, steps S100 and S126 in FIG. 14) for updating the AI ​​model based on the feedback information.

[0032] According to the thirteenth aspect of the present invention, the computer system can update the AI ​​model based on feedback information based on the player's operation in response to the evaluation information.

[0033] A fourteenth invention is a computer system in which, in the above-mentioned computer system, there are a plurality of individual analysis and evaluation means in the output layer, and the evaluation information provision control means generates the evaluation information based on the analysis and evaluation results of the plurality of individual analysis and evaluation means in the output layer.

[0034] According to the fourteenth aspect of the present invention, the computer system can generate evaluation information based on a plurality of analytical evaluation results in the output layer.

[0035] A fourteenth aspect of the present invention is the computer system, wherein the individual analysis and evaluation means performs the individual analysis and evaluation using an AI model for individually analyzing and evaluating the game situation information; The computer system further comprises a feedback information receiving means for receiving feedback information based on the player's operation in response to the evaluation information, and an AI model update means for updating the AI ​​model of the individual analysis evaluation means based on the feedback information.

[0036] According to the fifteenth aspect of the present invention, the computer system can update the AI ​​model used for the individual analytical evaluation based on feedback information based on the player's operation in response to the evaluation information. [Brief explanation of the drawings]

[0037] [Figure 1] FIG. 1 is a system configuration diagram showing an example of the configuration of a game system. [Figure 2] FIG. 10 is a diagram for explaining an overview of analysis and evaluation of game play. [Figure 3] FIG. 10 is a diagram for explaining game situation information and data selected for analysis and evaluation. [Figure 4] FIG. 1 is a diagram for explaining a primary analytical evaluation. [Figure 5] FIG. 10 is a diagram for explaining secondary analytical evaluation. [Figure 6] A diagram to explain an example of the application of primary and secondary analytical evaluation to a game. [Figure 7] A diagram to explain an example of the application of primary and secondary analytical evaluation to a game. [Figure 8] A diagram to explain an example of the application of primary and secondary analytical evaluation to a game. [Figure 9] FIG. 2 is a diagram showing examples of programs and data stored in the server system. [Figure 10] FIG. 2 is a diagram showing an example of a functional unit realized by a control board of the server system. [Figure 11] FIG. 10 is a diagram showing a display example of a first request reception screen. [Figure 12] FIG. 4 is a diagram showing an example of the data configuration of game situation information. [Figure 13] 10 is a flowchart illustrating the flow of processing executed by the server system in relation to one game play. [Figure 14] Flowchart continued from Figure 13. [Figure 15] 10 is a flowchart illustrating the flow of a process for selecting an AI model to be used. [Figure 16] FIG. 10 is a diagram for explaining a modified example. [Figure 17] FIG. 10 is a diagram for explaining a modified example. [Figure 18] FIG. 10 is a diagram for explaining a modified example. DETAILED DESCRIPTION OF THE INVENTION

[0038] Hereinafter, examples of embodiments of the present invention will be described, but it goes without saying that the forms to which the present invention can be applied are not limited to the following embodiments.

[0039] FIG. 1 is a system configuration diagram showing an example of the configuration of a game system according to this embodiment. The game system 1000 is a computer system that controls the execution of a video game (a type of entertainment content that takes place in a virtual world), analyzes and evaluates game play, and provides evaluation information based on the results of the analysis and evaluation to the player.

[0040] The game system 1000 is also a computer system that distributes gameplay videos, accepts viewer comments, and provides a comment sharing service that shares these with players and viewers.

[0041] The game system 1000 is a computer system including a server system 1100 and user terminals 1500 (1500a, 1500b, . . . ) for each user, which are connected via a network 9 so as to enable data communication.

[0042] The network 9 refers to a communication path that allows data communication. That is, the network 9 includes a dedicated line (dedicated cable) for direct connection, a LAN (Local Area Network) such as Ethernet (registered trademark), a telephone communication network, a cable network, the Internet, and the like.

[0043] The server system 1100 is a computer system that performs various processes such as managing and controlling registered user information, controlling the progress of the game, analyzing and evaluating, and presenting evaluation information. The server system 1100 also performs various processes such as controlling the management functions of the video distribution website, controlling the distribution of gameplay videos, controlling the acceptance of viewer comments, and controlling the shared display of viewer comments within gameplay videos.

[0044] The server system 1100 includes a main unit 1101 equipped with a control board 1150 . The control board 1150 is equipped with various microprocessors such as a CPU (Central Processing Unit) 1151, a GPU (Graphics Processing Unit), and a DSP (Digital Signal Processor), various IC memories 1152 such as a VRAM, RAM, and ROM, and a communication device 1153. Note that some or all of the functions equipped on the control board 1150 may be realized by an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or an SoC (System on a Chip).

[0045] Although the server system 1100 is depicted as if it were a single server device, it may be configured to be realized by a plurality of devices. For example, the server system 1100 may be configured by providing a plurality of servers that share functions and connecting them to each other via an internal bus or network 9 so that they can communicate data. The server system 1100 may also include a database or online storage.

[0046] The user terminal 1500 serves as a man-machine interface (MMIF) for user 2 to play a game as a player. The user terminal 1500 also serves as a man-machine interface for user 2 to watch gameplay videos as a viewer. In other words, the user terminal 1500 for user 2 who is a player functions as a game play terminal for playing a game. The user terminal 1500 for user 2 who is a viewer functions as a viewing terminal for watching gameplay videos.

[0047] The user terminals 1500 (1500a, 1500b, . . . ) are computer systems connectable to the network 9, such as personal computers, smartphones, wearable computers, portable game devices, home game devices, and tablet computers.

[0048] The user terminal 1500 is a computer equipped with an operation input device, an image display device, and a control board 1550 that performs arithmetic processing. Examples of the operation input device include a touch panel 1506, a keyboard, a game controller, and a mouse. Examples of the image display device include the touch panel 1506, a head-mounted display, and a glasses-type display.

[0049] The control board 1550 is equipped with a CPU 1551, various microprocessors such as a GPU and a DSP, various IC memories 1552 such as a VRAM, a RAM and a ROM, a communication module 1553 connected to the network 9, and the like. These elements equipped on the control board 1550 are electrically connected via a bus circuit or the like, and are connected so as to be able to read and write data and send and receive signals. Part or all of the control board 1550 may be an ASIC, an FPGA, or an SoC.

[0050] The control board 1550 stores in the IC memory 1552 programs and various data for realizing the functions of the user terminal 1500. The user terminal 1500 executes predetermined application programs to realize the functions of a play terminal and a viewing terminal.

[0051] Specifically, the user terminal 1500 functions as a play terminal by executing an application program that serves as a predetermined game client program. The user terminal 1500 executes an application program that serves as a predetermined web browser program, thereby accessing a video distribution website managed by the server system 1100, enabling the user to watch gameplay videos and post comments.

[0052] FIG. 2 is a diagram for explaining an overview of the analysis and evaluation of game play. The server system 1100 stores and manages game situation information 700 for each game play. The game situation information 700 includes a wide variety of data related to game play. A portion of the data in the game situation information 700 is selected and used as a source for analysis and evaluation.

[0053] FIG. 3 is a diagram for explaining the game situation information 700 and the data selected for analysis and evaluation. The game status information 700 includes basic information 710, game parameter history data 712, viewer comment information 716, other information 718, past primary analysis and evaluation results 731, and past secondary analysis and evaluation results 732. Of course, data other than these may also be included as appropriate.

[0054] Basic information 710 may include, for example, identification information for the type of game being played, the time of play (elapsed time and remaining time since the start of play), data on the player level for each player, data on the type and attributes of the player character for each player, etc.

[0055] Game parameter history data 712 is data that periodically collects and stores various parameters related to the progress of the game. Game parameter history data 712 may include, for example, the play time when the data was recorded and the game progress status. The game progress status includes the game stage ID and quest ID currently being played, scenario progress information, game results (for example, the degree to which the play objective was achieved, score, number of kills, etc.), etc.

[0056] The game parameter history data 712 also includes status data of the player character, status data of NPCs (Non-Player Characters), status data of the game field, and the like.

[0057] The player character's status data and NPC's status data may include, for example, position coordinates, the character's physical condition, the character's experience points, ability parameter values, skill types, skill levels, etc. They may also include emotional parameter values, money in possession, items in possession, hunger levels, chat content between characters, etc.

[0058] The game parameter history data 712 may include secondary information determined from the player character status data and NPC status data, such as the relative positional relationship between player characters, the relative positional relationship between player characters and NPCs, and character display control information (e.g., color information and shape information).

[0059] The game field status data may include, for example, identification information for the game field, data relating to the placement position and display status (e.g., color, shape, whether or not damaged) of objects other than characters, such as background objects, and the like.

[0060] The viewer comment information 716 may include the content of comments posted by viewers, statistical data on the number of comments, statistical data on the content of comments divided into predetermined categories through natural language analysis, and the like.

[0061] Other information 718 may include, for example, chat content exchanged between players during the game, chat content exchanged between players and AI-controlled NPCs during the game, and the like.

[0062] The past primary analysis and evaluation result 731 and the past secondary analysis and evaluation result 732 will be described later.

[0063] Returning to Figure 2, at a given primary timing, the server system 1100 selects primary selected game status information 751 from the game status information 700, performs primary analysis evaluation 10 based on the primary selected game status information 751, and determines primary analysis evaluation result 12.

[0064] The "primary timing" is set to a timing that occurs during a relatively short period of gameplay, such as at a given time interval during gameplay, at the end of a game stage, or at the end of a battle with an enemy NPC. Of course, the primary timing may also be determined randomly. The primary analysis and evaluation 10 may be performed at a predetermined interval, and if the primary analysis and evaluation result 12 satisfies a predetermined change condition, this may be considered to be the primary timing.

[0065] The primarily selected game status information 751 is not the entire game status information 700 but a part of it. Specifically, the primarily selected game status information 751 is information selected from the information and data included in the basic information 710, game parameter history data 712, and viewer comment information 716 (see FIG. 3).

[0066] The primary analysis and evaluation 10 is an evaluation performed on predetermined evaluation items by inputting information and data contained in the game situation information 700, and a primary analysis and evaluation result 12 is output as the result of the evaluation.

[0067] The "evaluation items" can be set appropriately depending on the game content and game rules. For example, in an action RPG in which multiple players form parties and engage in encounter battles with enemy NPCs, the attacking skills of each player can be an evaluation item. Evaluation items may also be the defensive skills of each player, how each player uses items, how well the group / party coordinates their gameplay, etc.

[0068] 4 is a diagram for explaining the primary analytical evaluation 10. The primary analytical evaluation 10 is realized by selecting and using one or more primary individual analytical evaluation AI models 6 (6a, 6b, . . . ).

[0069] The primary individual analysis and evaluation AI model 6 receives information included in the game situation information 700 as input, and outputs an evaluation (primary analysis and evaluation result 12) for predetermined evaluation items that are preset in the AI ​​model.

[0070] The primary individual analysis and evaluation AI model 6 is a trained model (AI model) that has been trained with training data that uses information included in the primary selection game status information 751 as input and outputs a primary analysis and evaluation result 12 for a predetermined evaluation item. Alternatively, the primary individual analysis and evaluation AI model 6 may be realized as a decision tree model that arrives at one of the primary analysis and evaluation results 12 in light of a number of judgment criteria for the primary selection game status information 751. Hereinafter, the primary individual analysis and evaluation AI model 6 will be described as an AI model.

[0071] The primary individual analysis and evaluation AI model 6 has different setting elements for application requirements, evaluation criteria, evaluation type, evaluation items, and tone of voice for each model.

[0072] The application requirements for the primary individual analysis and evaluation AI model 6 are described as conditions that must be met for the information and data included in the game situation information 700.

[0073] For example, the game status information 700 may include, as one piece of basic information, the play time, which is measured as the play progresses. Furthermore, a threshold value or range of the play time may be set as an application requirement for the primary individual analysis and evaluation AI model 6. In this case, when the application requirement is met according to the progress of play over time after the start of play, the primary individual analysis and evaluation AI model 6 may begin to be used. As a result, the items to be evaluated may change.

[0074] For example, suppose the game status information 700 includes a game parameter history. The game parameters may include, for example, data indicating the game progress, status data of the player character, status data of the NPC, status data of the game field, game results, etc. Furthermore, suppose that a condition based on the game progress information is set as an application requirement for the primary individual analysis and evaluation AI model 6. In this case, when a specific game progress state (e.g., a situation where a dungeon puzzle cannot be solved) is reached and the application requirement is satisfied, the primary individual analysis and evaluation AI model 6 can be started to be used. If the application requirement is set as a condition using the character's status data, the primary individual analysis and evaluation AI model 6 can be started to be used when the player character or the NPC reaches a specific state and the application requirement is satisfied.

[0075] For example, suppose that the game situation information 700 includes information on viewer comments. Also, suppose that conditions based on the content of the viewer comments are set as application requirements for the primary individual analysis and evaluation AI model 6. In this case, when a certain type of comment is made by a viewer and the application requirements are satisfied, use of the primary individual analysis and evaluation AI model 6 can be initiated. As a result, it appears that evaluation by the primary individual analysis and evaluation AI model 6 begins in response to the viewer comments.

[0076] The application requirements set for each of the multiple primary individual analysis and evaluation AI models 6 may be unique to each model, or may be partially common to all models. In Figure 4, these are represented as "application requirements A" and "application requirements B."

[0077] The evaluation criteria of the primary individual analysis and evaluation AI model 6 are so-called yardsticks for determining grades / ranks / levels / stages such as high / low / superior / good / bad. For example, they may be "lenient," "normal," or "strict."

[0078] The evaluation type of the primary individual analysis evaluation AI model 6 may be either "positive (praise, pointing out good points)", "negative (scold, pointing out bad points and areas for improvement)", or "a mixture of positive and negative".

[0079] One or more evaluation items are set for the primary individual analysis and evaluation AI model 6. For example, the evaluation items may be subdivided such that a comprehensive A evaluation item is "the growth level of the individual player," a detailed A1 evaluation item is "the growth level in attack," and an A2 evaluation item is "the growth level in defense."

[0080] The tone of speech of the primary individual analysis and evaluation AI model 6 indicates the type of language used when expressing the primary analysis and evaluation result 12 and the primary evaluation information 14 in natural language. The type of tone of speech can be set as appropriate. For example, it may be "gentle," "friendly tone," "instructor tone," "explanatory tone," "polite," etc. The tone of speech of the primary individual analysis and evaluation AI model 6 is one of these.

[0081] In the machine learning process, the primary individual analytical evaluation AI model 6 uses a limited number of predetermined types of information and data, which are designated as the primary selected game status information 751, from among the types of information and data stored as the game status information 700. Therefore, the primary selected game status information 751 is largely determined by which primary individual analytical evaluation AI model 6 is used in the primary analytical evaluation 10.

[0082] In a configuration in which the primary individual analysis and evaluation AI model 6 is realized using a decision tree model, the type of data selected as the primary selection game situation information 751 is the same as the type of data adopted as the branching decision criteria of the decision tree model.

[0083] The server system 1100 determines the primary individual analytical evaluation AI model 6 to be used for the primary analytical evaluation 10 by one of the following methods. (1) Select a primary individual analysis and evaluation AI model 6 that satisfies the specified application requirements (A application requirements, B application requirements, ... in Figure 4) set for each primary individual analysis and evaluation AI model 6. (2) The evaluation items for the primary analysis evaluation 10 are selected from candidates according to the game situation, and a primary individual analysis evaluation AI model 6 capable of evaluating the selected evaluation items is selected. (3) Determine the evaluation items required for the primary analysis and evaluation 10, and select a primary individual analysis and evaluation AI model 6 that can evaluate the determined evaluation items.

[0084] The server system 1100 basically adopts (1) to select the primary individual analysis and evaluation AI model 6, but also incorporates (2) as appropriate.

[0085] Furthermore, if there is a bias exceeding the standard in the evaluation items that have been subjected to the past primary analysis evaluation 10, the server system 1100 adopts (3) to first select evaluation items that reduce the bias, and then selects the primary individual analysis evaluation AI model 6 that can output this.

[0086] Furthermore, if the player requests in advance which primary individual analysis and evaluation AI model 6 to use for the primary analysis and evaluation 10, or which evaluation items, evaluation criteria, and evaluation types (e.g., positive type / negative type) to include in the primary analysis and evaluation 10, then (3) is adopted. Specifically, evaluation items, evaluation criteria, and evaluation types are set in advance for each primary individual analysis and evaluation AI model 6, and the server system 1100 selects an AI model with the same requested evaluation items, evaluation criteria, and evaluation type, or an AI model with settings closest to the request.

[0087] Returning to FIG. 2, the primary analysis and evaluation results 12 include the score for each evaluation item, whether it occurred, and the degree of achievement, and are stored in the game situation information 700 as past primary analysis and evaluation results 731 (see FIG. 3).

[0088] The server system 1100 determines or generates primary evaluation information 14 based on the primary analysis and evaluation results 12 . The primary evaluation information 14 may be, for example, a report-style text in natural language containing one or more primary analysis evaluation results 12 within a predetermined number of characters. The primary evaluation information 14 in natural language may include words of praise, criticism, scolding, and advice depending on the evaluation type. The primary evaluation information 14 in natural language may be generated using LLM (Large Language Models). In this case, the primary evaluation information 14 in natural language uses the "tone of voice" of the primary individual analysis evaluation AI model 6 used in the primary analysis evaluation 10 as one of its inputs, and is output in the natural language of that tone of voice.

[0089] The primary evaluation information 14 may simply be a score, a number of stars, a rank, or a graded evaluation such as "excellent," "good," "fair," or "poor." The primary evaluation information 14 may also be information indicating a reward (e.g., points, items, titles, etc.) to be given to a player depending on the level of the analysis evaluation, or information indicating a still image or video announcing the evaluation. The primary evaluation information 14 may also be information indicating a sound (e.g., applause, sighs, beeps, etc.) corresponding to the level of the analysis. In these cases, the primary evaluation information 14 may be prepared in advance by comparing one or more primary analysis evaluation results 12 with predetermined criteria, and may be determined based on the primary analysis evaluation results 12 at each time.

[0090] The server system 1100 provides the primary evaluation information 14 to the player. The provision 16 of primary evaluation information is executed appropriately according to the data format of the primary evaluation information 14. For example, if the primary evaluation information 14 is in natural language, it may be displayed as text on the game screen. If the primary evaluation information 14 is a still image or a video, it is displayed on the game screen. If the primary evaluation information 14 is audio, it is played and emitted as part of the game sound.

[0091] After providing 16 the primary evaluation information, the server system 1100 executes primary feedback information reception 18. Primary feedback information reception 18 is reception of player evaluations for the primary evaluation information 14. The reception may be performed each time the information is provided, or at key points in the game progress (for example, at the end of a battle, the end of a game stage, etc.). Primary feedback information reception 18 may be a rating. Primary feedback information reception 18 may be a graded evaluation expressed in natural language, such as "very favorable," "favorable," "average," "rather unfavorable," or "dislike."

[0092] The server system 1100 updates the learning of the primary individual analytical evaluation AI model 6 used in the primary analytical evaluation 10 based on the received player's evaluation.

[0093] The server system 1100 selects secondary selected game status information 752 from the game status information 700 at a given secondary timing, and executes secondary analysis evaluation 20 using the secondary selected game status information 752 as input to determine secondary analysis evaluation result 22.

[0094] The "secondary timing" may be a turning point in the game progress. For example, in an action RPG, the secondary timing may be the timing when a game scenario consisting of multiple game stages or quests is completed after a predetermined number of game stages or quests have been played, etc. The secondary timing may also be the timing when the rank of the player or party changes after a battle with a specific enemy character (such as a stage boss character), etc.

[0095] For example, in the case of a mahjong game, the secondary timing may be the timing when the "stage" (East stage, South stage, etc.), which is the unit of the game, changes. For example, in the case of a Sugoroku game, the secondary timing may be every predetermined number of laps or when a participant drops out. The secondary timing is basically set to a timing that occurs after playing the game for a longer period of time than the primary timing. For example, it may be set so that one secondary timing occurs after multiple primary timings have passed.

[0096] The secondary selection game status information 752 is not the entire game status information 700 but a part of it. Specifically, the secondary selection game status information 752 is a part of information and data selected from the information and data included in the basic information 710, game parameter history data 712, viewer comment information 716, past primary analysis and evaluation results 731, and past secondary analysis and evaluation results 732 (see FIG. 3).

[0097] The secondary analysis and evaluation 20 inputs information and data contained in the game situation information 700, and outputs an evaluation (secondary analysis and evaluation result 22) for predetermined evaluation items.

[0098] 5 is a diagram for explaining details of the secondary analytical evaluation 20. The secondary analytical evaluation 20 is realized by selecting and using one or more secondary individual analytical evaluation AI models 7 (7a, 7b, ...).

[0099] The secondary individual analysis and evaluation AI model 7 receives information included in the game situation information 700 as input, performs evaluation on predetermined evaluation items, and outputs the secondary analysis and evaluation result 22 as the evaluation result.

[0100] The secondary individual analysis and evaluation AI model 7 is a trained model (AI model) that has been trained with training data that uses information included in the secondary selection game status information 752 as input and outputs a secondary analysis and evaluation result 22 for a predetermined evaluation item. Alternatively, the secondary individual analysis and evaluation AI model 7 may be realized as a decision tree model that arrives at one of the secondary analysis and evaluation results 22 in light of a number of judgment criteria for the secondary selection game status information 752. Hereinafter, the secondary individual analysis and evaluation AI model 7 will basically be described as an AI model.

[0101] The secondary individual analysis and evaluation AI model 7 has different setting elements for application requirements, evaluation criteria, evaluation type, evaluation items, and tone of voice depending on the model.

[0102] The application requirements for the secondary individual analysis and evaluation AI model 7 are described as conditions that must be met for the information and data included in the basic information 710, game parameter history data 712, viewer comment information 716, other information 718, past primary analysis and evaluation results 731, and past secondary analysis and evaluation results 732 (see FIG. 3).

[0103] The evaluation criteria of the secondary individual analysis and evaluation AI model 7 are so-called yardsticks for determining grades / ranks / levels / stages such as high / low / superior / good / bad. For example, they may be "lenient," "normal," or "strict."

[0104] The evaluation type of the secondary individual analysis evaluation AI model 7 may be either "positive," "negative," or "mixed positive and negative."

[0105] The secondary individual analysis and evaluation AI model 7 has one or more evaluation items set.

[0106] The tone of speech of the secondary individual analysis and evaluation AI model 7 indicates the type of language used when expressing the secondary analysis and evaluation results 22 and the secondary evaluation information 24 in natural language. The type of tone of speech can be set as appropriate. For example, it may be "gentle," "friendly tone," "instructor tone," "explanatory tone," "polite," etc. The tone of speech of the secondary individual analysis and evaluation AI model 7 is one of these.

[0107] In the machine learning process, the secondary individual analytical evaluation AI model 7 uses a limited number of predetermined types of information and data, designated as secondary selection game status information 752, from among the types of information and data stored as game status information 700. Therefore, the secondary selection game status information 752 is largely determined by which secondary individual analytical evaluation AI model 7 is used in the secondary analytical evaluation 20.

[0108] In a configuration in which the secondary individual analysis and evaluation AI model 7 is realized using a decision tree model, the type of data selected as the secondary selection game situation information 752 is the same as the type of data adopted as the branching judgment criteria of the decision tree model.

[0109] The server system 1100 determines the secondary individual analytical evaluation AI model 7 to be used for the secondary analytical evaluation 20 by one of the following methods. (i) A secondary individual analysis and evaluation AI model 7 that satisfies the specified application requirements (α application requirements, β application requirements, ... in Figure 5) set for each secondary individual analysis and evaluation AI model 7 is selected. (b) The evaluation items for the secondary analysis evaluation 20 are selected from candidates according to the game situation, and a secondary individual analysis evaluation AI model 7 capable of evaluating the selected evaluation items is selected. (c) Determine the evaluation items required for the secondary analysis and evaluation 20, and select the secondary individual analysis and evaluation AI model 7 that can evaluate the determined evaluation items.

[0110] The server system 1100 basically adopts (A) to select the secondary individual analysis and evaluation AI model 7, but also incorporates (B) as appropriate.

[0111] Furthermore, if there is a bias exceeding the standard in the evaluation items that have been subjected to the past secondary analysis evaluation 20, the server system 1100 adopts (C). Then, the server system 1100 first selects evaluation items that reduce the bias, and selects the secondary individual analysis evaluation AI model 7 that can output them.

[0112] Furthermore, if the player requests in advance which secondary individual analytical evaluation AI model 7 to use in the secondary analytical evaluation 20 or which evaluation items to include in the secondary analytical evaluation 20, the server system 1100 adopts (c) and selects a secondary individual analytical evaluation AI model 7 that can output the evaluation items requested by the player.

[0113] Returning to FIG. 2, the secondary analysis and evaluation results 22 include the score for each evaluation item, whether it occurred, and the degree of achievement, and are stored in the game situation information 700 as past primary analysis and evaluation results 731 (see FIG. 3).

[0114] The server system 1100 determines or generates secondary evaluation information 24 based on the secondary analysis and evaluation results 22 .

[0115] The secondary evaluation information 24 may be, for example, a report-style text in natural language containing one or more secondary analysis evaluation results 22 within a predetermined number of characters. The natural language secondary evaluation information 24 may include praise, criticism, reprimand, or advice depending on the evaluation type. The natural language secondary evaluation information 24 may be generated using LLM (Large Language Models). In this case, the natural language secondary evaluation information 24 uses the "tone of voice" of the secondary individual analysis evaluation AI model 7 used in the secondary analysis evaluation 20 as one of its inputs and is output in the natural language of that tone of voice.

[0116] The secondary evaluation information 24 may simply be a score, the number of stars, a rank, or a graded evaluation such as "excellent," "good," "fair," or "poor." The secondary evaluation information 24 may also be information indicating a reward (e.g., points, items, titles, etc.) to be given to a player depending on the level of the analysis evaluation, or information indicating a still image or video announcing the evaluation. The secondary evaluation information 24 may also be information indicating a sound (e.g., applause, sighs, beeps, etc.) depending on the level of the analysis. In these cases, the secondary evaluation information 24 may be prepared in advance by comparing one or more secondary analysis evaluation results 22 with predetermined criteria, and may be determined based on the secondary analysis evaluation results 22 at each time.

[0117] The server system 1100 provides the secondary evaluation information 24 to the player. The provision 26 of the secondary evaluation information is executed appropriately according to the data format of the secondary evaluation information 24, similar to the provision 16 of the primary evaluation information.

[0118] Furthermore, the server system 1100 executes secondary feedback information reception 28 for the provision of secondary evaluation information 26, in the same manner as for the primary analytical evaluation. Then, the server system 1100 updates the learning of the secondary individual analytical evaluation AI model 7 used in the secondary analytical evaluation 20, based on the received player evaluation.

[0119] In this way, a player playing a game in the game system 1000 can receive evaluations in various hierarchical forms, including primary and secondary analysis evaluations. This enhances the player's gaming experience and serves as an important reference for the player to improve their skills. User viewers can also watch the gameplay while receiving evaluations in various hierarchical forms.

[0120] Depending on the game content and game scenario, the secondary timing may be simultaneous with the primary timing. In this case, from the player's perspective, the provision of primary evaluation information 16 based on the primary analysis evaluation and the provision of secondary evaluation information 26 appear to be performed simultaneously, approximately simultaneously, or consecutively.

[0121] 6 to 8 are diagrams for explaining examples of application of primary analysis and secondary analysis to games.

[0122] In the application example of Fig. 6, the game genre can be set appropriately, such as a board game or an action RPG. The players are user players. Comments from user viewers are accepted and shared and displayed in a comment sharing service (hereinafter referred to as "comment shared display" as appropriate).

[0123] In the application example of Figure 6, the primary analysis evaluation 10 is an evaluation by an AI viewer who evaluates and comments on the content of gameplay. Primary individual analysis evaluation AI models 6 (see Figure 4) are prepared for each type of AI viewer. In the application example of Figure 6, the primary evaluation information 14 is a comment in natural language generated based on the primary analysis evaluation result 12, and the comment is shared and displayed together with the comments of the user viewer.

[0124] In the application example of Figure 6, the secondary analysis evaluation 20 is an evaluation by a commentator who comprehensively evaluates and comments on the comments of user viewers, comments of AI viewers, and the content of gameplay. The secondary individual analysis evaluation AI model 7 (see Figure 5) is prepared for each type of commentator.

[0125] The secondary evaluation information 24 in the application example of FIG. 6 is generated as a general comment in natural language, and is displayed as text in a dedicated speech bubble by displaying a commentator NPC on the game screen.

[0126] By appropriately setting the application requirements and evaluation items for the primary individual analysis and evaluation AI model6, it is possible to have the AI ​​viewer appear depending on the number of user viewers and comment on evaluation items that were not mentioned in the user viewers' comments. As a result, the user viewers at that time can learn about evaluation items and evaluations that the user viewers have not commented on, and enjoy gaining a new perspective. Of course, for players who are playing, the AI ​​viewer's comments can also provide immediate hints for improving their skills.

[0127] The primary individual analysis and evaluation AI model 6 offers a variety of models with different evaluation criteria, evaluation types, and tone of voice, so that, combined with the combination of applicable requirements, a wide variety of evaluations appear as comments, resulting in a lively system that is far from becoming monotonous.

[0128] Similarly, by appropriately configuring the secondary individual analysis and evaluation AI model 7, it is possible to have it make comments on evaluation points that were not mentioned in the user viewer's comments or the AI ​​viewer's comments in the primary individual analysis and evaluation. Because these comments are made at a secondary timing, they may provide hints for a player who has just finished one section of play to help them play the next section.

[0129] The primary individual analysis and evaluation AI model 6 and the secondary individual analysis and evaluation AI model 7 are updated based on feedback of player evaluations in response to the provision of primary evaluation information 16 and the provision of secondary evaluation information 26.

[0130] As a result, in a configuration in which the primary individual analysis and evaluation AI model 6 and the secondary individual analysis and evaluation AI model 7 are managed for each player, the AI ​​model will, through feedback, begin to make analysis and evaluations that leave a more favorable impression on the player.In a configuration in which the primary individual analysis and evaluation AI model 6 and the secondary individual analysis and evaluation AI model 7 are shared by the game, this corrects the discrepancy between the evaluations originally envisioned by the game creator when setting the training data and the evaluations perceived by actual users.

[0131] 7, the game genre is a board game, an action RPG, etc. Players are composed of user players who are users and AI players controlled by the server system 1100. Comments from user viewers are accepted and displayed in a comment sharing display.

[0132] The primary analysis evaluation 10 in the application example of Figure 7 is an evaluation of the game situation that will be the basis for the AI ​​player's decision to act as a single player. The primary individual analysis evaluation AI model 6 (see Figure 4) is prepared for each type of AI player.

[0133] In the application example of Figure 7, the primary evaluation information 14 is generated in natural language as the AI ​​player's utterances (monologue or conversations with other players) and is displayed as text on the game screen so that the user viewer can understand it along with the utterances of other players.

[0134] In the application example of Figure 7, the secondary analysis evaluation 20 is an evaluation by a commentator who watches and evaluates the gameplay together with the user viewers. Secondary individual analysis evaluation AI models 7 (see Figure 5) are prepared for each type of commentator. Secondary evaluation information is generated as a general comment in natural language and displayed as text in a speech bubble of the commentator NPC displayed on the game screen.

[0135] By appropriately setting the application requirements and evaluation items for the primary individual analytical evaluation AI model6, it becomes possible to introduce various types of AI players depending on the user player's skill, experience, game progress, etc. As a result, the user player can virtually play with various types of AI players and be exposed to various evaluations. The primary evaluation information can also be useful for the user player in further improving their skill.

[0136] The primary individual analysis and evaluation AI model 6 offers a variety of models with different evaluation criteria, evaluation types, and tone of voice, so that, combined with the combination of application requirements, AI players with diverse play styles and personalities will appear, resulting in a lively game that is far from becoming monotonous.

[0137] Similarly, by appropriately configuring the secondary individual analysis evaluation AI model 7, it is possible to have it comment on user viewers' comments and the gameplay of the AI ​​player. As mentioned above, these comments can provide hints for the player on how to play the next section of the game.

[0138] In the application example of Fig. 8, the game genre is a game in which an instructor NPC or a trainer NPC appears. The player is a user player. Comments from user viewers are accepted and displayed in a comment sharing display.

[0139] The primary analysis evaluation 10 in the application example of Figure 8 is an evaluation by the AI ​​instructor NPC. It is an evaluation that the AI ​​instructor NPC uses as a basis for making decisions when taking training-related actions, and is an evaluation of the game situation. Primary individual analysis evaluation AI models 6 (see Figure 4) are prepared for each type of AI instructor NPC. The primary evaluation information 14 in the application example of Figure 8 is generated in natural language that is suitable for the AI ​​instructor NPC's utterances, and is displayed as text on the game screen in natural language that makes the user viewer feel as if the instructor is speaking.

[0140] In the application example shown in Figure 8, the secondary analysis and evaluation 20 is an evaluation by a commander NPC who is a superior to the AI ​​instructor NPC. It is an evaluation that the commander NPC uses as a basis for making decisions when taking training-related actions, and is an evaluation of the game situation. Secondary individual analysis and evaluation AI models 7 (see Figure 5) are prepared for each type of commander NPC. The commander NPC gives an overall evaluation of the training and selects and issues operational orders (so-called quests) as a result. The secondary evaluation information is generated in natural language that is suitable for the commander NPC's speech, and is displayed as text on the game screen in natural language that makes it seem to the user viewer as if it were being spoken by a commander.

[0141] By appropriately setting the application requirements and evaluation items for the Primary Individual Analysis and Evaluation AI Model6, it becomes possible to have various types of AI instructor NPCs appear depending on the user player's skill and experience, game progress, etc. As a result, the AI ​​instructor NPC will provide advice that is appropriate for the user player at that time.

[0142] Furthermore, a variety of models with different evaluation criteria, evaluation types, and tone of voice are available for the primary individual analysis and evaluation AI model 6. Therefore, when the user player selects the primary individual analysis and evaluation AI model 6 before starting the game, the user player can play the game while being taught by the type of instructor he or she desires (for example, a strict instructor, a kind instructor, etc.).

[0143] Similarly, for the secondary individual analysis and evaluation AI model 7, various models with different application requirements, evaluation items, evaluation criteria, evaluation types, and tone of voice are available, so commander NPCs with various command roles will appear in the game, increasing the variety and interest of the game's progress.

[0144] The positioning of the first and second analytical evaluations in a game is not limited to the application examples shown in Figures 6 to 8. The second analytical evaluation may be omitted, or further analytical evaluations such as the third analytical evaluation or later may be added. User viewer comments may also be omitted as appropriate.

[0145] FIG. 9 is a diagram showing an example of programs and data stored in the server system 1100. As shown in FIG. The server system 1100 stores in the IC memory 1152 a server program 501, evaluation item selection pattern data 510, a plurality of types of primary individual analytical evaluation AI model data 512, and a plurality of types of secondary individual analytical evaluation AI model data 514.

[0146] The server system 1100 also stores primary selected game situation information selection criteria data 516 , secondary selected game situation information selection criteria data 518 , primary evaluation information definition data 520 , and secondary evaluation information definition data 522 .

[0147] The server system 1100 also stores user registration data 600, game status information 700, and current date and time 900. Of course, other programs and data may also be stored as appropriate.

[0148] The server system 1100 executes and processes the server program 501 in the CPU 1151, thereby realizing the function of a server processing unit 200s in the control board 1150, as shown in FIG. 10, for example.

[0149] The server processing unit 200s has a user registration management unit 202, a game progress control unit 204, a distribution control unit 206, a viewer comment sharing control unit 208, and a player request reception control unit 210. The server processing unit 200s also has an evaluation item selection unit 220, an information selection unit 224, an analysis and evaluation unit 230, an evaluation information provision control unit 240, a feedback information reception control unit 242, an AI model update unit 244, and a timing unit 280.

[0150] The user registration management unit 202 executes registration procedure processing related to user registration and storage management processing of registration information.

[0151] The game progress control unit 204 executes control related to the game progress, sequentially updates the game status information 700 (see FIG. 3), and determines the game results. The game progress control unit 204 also controls AI-controlled characters such as AI viewers, commentator NPCs, AI players, AI instructor NPCs, and AI commander NPCs (see FIGS. 6, 7, and 8).

[0152] The distribution control unit 206 controls the distribution of game play images as game play videos.

[0153] The viewer comment sharing control unit 208 receives comments from viewer users and performs control to display the received comments in a comment sharing display within the gameplay video being distributed.

[0154] The player request reception control unit 210 controls the reception of player requests regarding the analysis and evaluation (primary analysis and evaluation 10, secondary analysis and evaluation 20) related to the game.

[0155] The player request reception control unit 210 displays a request reception screen W2 for receiving a request regarding the primary analysis evaluation 10 on the user terminal 1500 of the user player, as shown in FIG. 11, for example.

[0156] The request reception screen W2 includes an evaluator type selection section 41, a customization section 42, and a confirmation operation icon 43.

[0157] The evaluator type selection unit 41 regards requests for the primary analysis evaluation 10 as different settings of the primary individual analysis evaluation AI model 6 (see FIG. 3), i.e., different types of evaluators, and allows the player to select one or more. The evaluator type selection unit 41 displays a character icon prepared in advance for each AI model, which also serves as a selection operation icon. The type details display unit 44 displays in detail the settings (evaluation criteria, evaluation type, evaluation items, tone) for each primary individual analysis evaluation AI model 6 corresponding to the character icon selected by the player.

[0158] The customization section 42 is provided with selection sections for each setting of the primary individual analysis evaluation AI model 6, and the player can select each setting he or she desires. If there is a primary individual analysis and evaluation AI model 6 that matches the setting results of the customization unit 42, that AI model is used for the primary analysis and evaluation 10. If there is no primary individual analysis and evaluation AI model 6 that matches the setting results of the customization unit 42, the primary individual analysis and evaluation AI model 6 that has the highest degree of compatibility with the setting contents is used for the primary analysis and evaluation 10.

[0159] The player request reception control unit 210 displays a second request reception screen on the user terminal 1500 of the user player, which receives requests regarding the secondary analysis evaluation 20. The second request reception screen can be realized in the same way as the request reception screen W2, with the only difference being whether the setting target is the primary analysis evaluation 10 or the secondary analysis evaluation 20, so a duplicated explanation will be omitted.

[0160] The evaluation item selection unit 220 selects selected evaluation items from a plurality of evaluation items. In the first analysis evaluation 10 and the second analysis evaluation 20, an AI model that analyzes and evaluates the selected evaluation items is selected and used (see FIGS. 4 and 5).

[0161] Specifically, when a setting operation is accepted in the evaluator type selection section 41 of the request acceptance screen W2 (see FIG. 11), the evaluation item selection section 220 sets the evaluation item of the primary individual analysis evaluation AI model 6 corresponding to the character icon for which the setting operation is accepted as the selected evaluation item. When a setting operation is accepted in the customization section 42, the evaluation item selected and set in the evaluation item setting section is set as the selected evaluation item.

[0162] Furthermore, when a setting operation is not accepted on the request acceptance screen W2, the evaluation item selection unit 220 determines the selected evaluation items. At that time, the evaluation item selection unit 220 may refer to the evaluation item selection pattern data 510 (see FIG. 9).

[0163] The evaluation item selection pattern data 510 includes application requirements and a list of candidate evaluation items to be selected, and is prepared in the same number as the number of different game situations described in the application requirements.

[0164] The application requirements of the evaluation item selection pattern data 510 indicate the conditions for the game situation that must be met in order for the pattern data to be adopted and applied. Specifically, the application requirements are written by combining one or more sub-conditions with AND or OR. The sub-conditions are written as conditions that must be met for the parameter values ​​and information included in the basic information 710, game parameter history data 712, game results 714, viewer comment information 716, and other information 718 of the game situation information 700 (see FIG. 3).

[0165] 10, the information selection unit 224 selects selected game situation information from the player's game situation information 700, which will be the source of analysis and evaluation by the analysis and evaluation unit 230. This corresponds to the primary selected game situation information 751 and the secondary selected game situation information 752 in FIG.

[0166] Regardless of the amount of information, the analysis and evaluation unit 230 analyzes and evaluates the given game situation information 700. Specifically, the analysis and evaluation unit 230 receives the selected game situation information selected by the information selection unit 224 as input and outputs the analysis and evaluation results.

[0167] The analysis and evaluation unit 230 also includes a plurality of primary individual analysis and evaluation units 231 , a plurality of secondary individual analysis and evaluation units 232 , and an individual selection unit 236 .

[0168] The primary individual analytical evaluation unit 231 is realized by the primary individual analytical evaluation AI model 6 used in the primary analytical evaluation 10. The analytical evaluation results of the primary analytical evaluation 10 are used as input for the secondary analytical evaluation 20, so the primary individual analytical evaluation unit 231 can be said to be an input layer side individual analytical evaluation unit 233.

[0169] The secondary individual analytical evaluation unit 232 is realized by the secondary individual analytical evaluation AI model 7 used in the secondary analytical evaluation 20. Since the secondary analytical evaluation 20 uses the analytical evaluation results of the primary analytical evaluation 10 as input, the secondary individual analytical evaluation unit 232 can be said to be an output layer side individual analytical evaluation unit 234.

[0170] Therefore, it can be said that the analysis and evaluation unit 230 performs hierarchical analysis and evaluation in which the individual analysis and evaluation unit on the output layer side uses the analysis and evaluation results of the individual analysis and evaluation unit on the input layer side to perform analysis and evaluation.

[0171] The individual selection unit 236 selects a selected individual analysis evaluation unit to be used for the primary analysis evaluation 10 and the secondary analysis evaluation 20 from among a plurality of individual analysis evaluation units (primary individual analysis evaluation unit 231, secondary individual analysis evaluation unit 232). This corresponds to the selection of the primary individual analysis evaluation AI model 6 to be used for the primary analysis evaluation 10 and the selection of the secondary individual analysis evaluation AI model 7 to be used for the secondary analysis evaluation 20 (see FIGS. 4 and 5). For example, the individual selection unit 236 selects the selected individual analysis evaluation unit based on history information of past analysis evaluation results included in the game situation information 700 (for example, the processing after step S60 in FIG. 15).

[0172] The evaluation information provision control unit 240 controls the provision of evaluation information based on the analysis and evaluation results by the analysis and evaluation unit 230 to the player. The evaluation information provision control unit 240 is a machine learning model that has been trained with training data and that outputs evaluation information in a predetermined data format, using, for example, the analysis and evaluation results of the primary analysis and evaluation 10 and the secondary analysis and evaluation 20 as input. When the evaluation information is words such as report sentences, the machine learning model may include an LLM in the output stage.

[0173] The evaluation information provision control unit 240 includes a second individual selection unit 241 . The second individual selection unit 241 selects a selected individual analysis and evaluation unit from among a plurality of individual analysis and evaluation units (the primary individual analysis and evaluation unit 231, the secondary individual analysis and evaluation unit 232). For example, the second individual selection unit 241 may select a selected individual analysis and evaluation unit based on setting information regarding the player's requests for evaluation during the game (for example, steps S50 to S52 in FIG. 15). Alternatively, the second individual selection unit 241 may select a selected individual analysis and evaluation unit based on the results of analysis and evaluation performed by each of the plurality of individual analysis and evaluation units (the primary individual analysis and evaluation unit 231, the secondary individual analysis and evaluation unit 232), and provide the player with evaluation information based on the analysis and evaluation of the selected individual analysis and evaluation unit.

[0174] The feedback information reception control unit 242 receives feedback information based on the player's operation in response to the evaluation information.

[0175] The AI ​​model update unit 244 performs control to update the AI ​​model based on the feedback information.

[0176] The timekeeping unit 280 uses a system clock to measure various times such as the current date and time 900 (see FIG. 9) and time limits.

[0177] Returning to Figure 9, primary individual analysis and evaluation AI model data 512 is prepared for each primary individual analysis and evaluation AI model 6 (see Figure 4). One primary individual analysis and evaluation AI model data 512 includes an AI model ID, applicable requirements, evaluation criteria, evaluation type, evaluation items, tone type, and machine learning model data. Of course, other data may also be included as appropriate. The evaluation item list indicates the evaluation items that can be output by the AI ​​model, and selecting and executing the AI ​​model will display the evaluation items selected from multiple evaluation items.

[0178] Secondary individual analysis evaluation AI model data 514 is prepared for each secondary individual analysis evaluation AI model 7 (see Figure 5). One secondary individual analysis evaluation AI model data 514 includes an AI model ID, application requirements, evaluation criteria, evaluation type, evaluation items, tone type, and machine learning model data. Of course, other data may also be included as appropriate. The evaluation item list indicates the evaluation items that can be output by the AI ​​model, and selecting and executing the AI ​​model will display the evaluation items selected from multiple evaluation items.

[0179] The first selection game situation information selection criteria data 516 is criteria data for determining which information or data to select as the first selection game situation information 751 (see FIG. 3), and is prepared for each selection method. One first selection game situation information selection criteria data 516 includes one or more types of evaluation items used in the first analysis and evaluation, and a list of information types and data types included in the game situation information 700 selected as the first selection game situation information 751.

[0180] The secondary selection game situation information selection criteria data 518 is criteria data for determining which information or data to select as the secondary selection game situation information 752 (see FIG. 3), and is prepared for each selection method. One secondary selection game situation information selection criteria data 518 includes one or more types of evaluation items used in the secondary analysis evaluation, and a list of information types and data types included in the game situation information 700 to be selected as the secondary selection game situation information 752.

[0181] The primary evaluation information definition data 520 is prepared for each type of primary evaluation information 14 and is data for determining or generating the primary evaluation information 14 based on the primary analysis and evaluation result 12 . One primary evaluation information definition data 520 stores, for example, application requirements indicating the conditions that must be met to select and apply the definition data, in association with primary evaluation information. The application requirements are described by combining one or more sub-conditions with AND or OR. The sub-conditions are described by the type, number, and evaluation results of the evaluation items that are to be used for the primary analysis evaluation 10.

[0182] The secondary evaluation information definition data 522 is prepared for each type of secondary evaluation information 24 and is data for determining or generating the secondary evaluation information 24 based on the secondary analysis evaluation result 22. A single secondary evaluation information definition data 522 stores, for example, application requirements indicating the conditions that must be met to select and apply the definition data, in association with secondary evaluation information. The application requirements are described by combining one or more sub-conditions using AND or OR. The sub-conditions are described using the type, number, and evaluation results of the evaluation items that are to be considered as the secondary analytical evaluation 20. The sub-conditions may also be described using the type, number, and evaluation results of the evaluation items that are to be considered as the primary analytical evaluation 10.

[0183] User registration data 600 is prepared for each user who has completed the registration procedure, and includes a user account, past game performance data, game save data, etc. Game save data includes, for example, the save date and time, player character setting data, a list of owned items, player level, etc. Of course, data other than these may also be included as appropriate.

[0184] 12 is a diagram showing an example of the data configuration of the game situation information 700. FIG.

[0185] Game status information 700 is prepared for each game play and stores various data related to that game play. One piece of game status information 700 includes, for example, a play ID 701, a player count list 702, player request setting data 704, and distribution control data 706.

[0186] The distribution control data 706 stores various data for realizing the distribution of gameplay videos, the acceptance of viewer comments, the sharing and display of comments, etc. For example, the distribution control data 706 may store distribution website data, gameplay video data, comment sharing control data, viewer accounts, the number of viewers, etc.

[0187] The game status information 700 also includes basic information 710, game parameter history data 712, viewer comment information 716, other information 718, past primary analysis and evaluation results 731, and past secondary analysis and evaluation results 732. The past primary analysis and evaluation results 731 and past secondary analysis and evaluation results 732 correspond to history information of past analysis and evaluation performance.

[0188] The game status information 700 also includes analytical evaluation control data 720. One analytical evaluation control data 720 includes a player count 722 of the player who is the subject of analytical evaluation, a primary analytical evaluation used AI model ID 741, a secondary analytical evaluation used AI model ID 742, primary selection game status information 751, and secondary selection game status information 752. Of course, data other than these may also be included as appropriate.

[0189] FIG. 13 is a flowchart for explaining the flow of processing executed by the server system 1100 in relation to one game play. The server system 1100 performs a player reception process before the start of the game, and sets the accounts of the players participating in the game in the player count list 702 of the game status information 700 (step S10; see FIG. 12).

[0190] When a predetermined player request setting request operation is input by the player (YES in step S12), the server system 1100 executes player request reception control. Then, the server system 1100 displays a request reception screen W2 (see FIG. 11) on the user terminal 1500 that sent the request, and sets the player request setting data 704 in the game situation information 700 (step S14).

[0191] The server system 1100 initializes the basic information 710 of the game situation information 700, starts controlling the game progress (step S20), and starts sequentially updating the basic information 710 and recording the game parameter history data 712 (step S22).

[0192] The server system 1100 also starts distributing the gameplay video, accepting comments from viewers, and sharing and displaying the comments (step S24).

[0193] During the game, the server system 1100 monitors whether the primary timing has occurred. If the primary timing has occurred (YES in step S30), the server system 1100 executes loop A for each player (step S32 to step S102 in FIG. 14).

[0194] In loop A, the server system 1100 executes the AI ​​model selection process with "n=1" (step S34).

[0195] FIG. 15 is a flowchart for explaining the flow of the AI ​​model selection process. "n" represents the order of the analytical evaluation. The AI ​​model selection process with "n=1" selects the first individual analytical evaluation AI model 6 to be used in the first analytical evaluation 10. In other words, the AI ​​model selection process can be said to function as the individual selection unit 236.

[0196] In the AI ​​model selection process where n=1, the server system 1100 first checks whether or not the player's request setting data 704 (see FIG. 12) exists for the player who is the object of the loop process.

[0197] If there is player request setting data 704 for the player subject to the loop processing (YES in step S50), the server system 1100 searches for the primary individual analytical evaluation AI model 6 that best satisfies the request from among the primary individual analytical evaluation AI models 6. Then, the server system 1100 sets it as the primary individual analytical evaluation AI model 6 to be used in the primary analytical evaluation (step S52; see FIG. 4).

[0198] For example, focusing on evaluation items, the server system 1100 sets the evaluation items indicated in the desired evaluation item list of the player desired setting data 704 as "selected evaluation items" and searches for a primary individual analysis evaluation AI model 6 that can evaluate the most selected evaluation items. The server system 1100 similarly searches for the evaluation criteria, evaluation type, and tone of voice of the tone of voice, searching for a primary individual analysis evaluation AI model 6 that best satisfies each of the requests. The server system 1100 then sets the AI ​​model that is most frequently searched for as the primary individual analysis evaluation AI model 6 to be used in the primary analysis evaluation 10.

[0199] If the player request setting data 704 is not set (NO in step S50), the server system 1100 refers to the past primary analysis evaluation results 731 (see FIG. 12). Then, the server system 1100 determines whether a statistically significant number of past primary analysis evaluation results 731 are stored, and whether the evaluation items are statistically processed by type to determine whether a bias greater than a predetermined standard has occurred.

[0200] If there is a bias (YES in step S60), the server system 1100 selects an evaluation item from all types of evaluation items so as to reduce the bias (step S62).Then, the server system 1100 sets the primary individual analytical evaluation AI model 6 capable of evaluating the selected evaluation item as the primary individual analytical evaluation AI model 6 to be used in the primary analytical evaluation (step S64).

[0201] If there is no bias (NO in step S60), the server system 1100 determines by lottery processing whether to select the primary individual analytical evaluation AI model 6 to be used in the primary analytical evaluation based on the application requirements or by random selection (step S70).

[0202] If the lottery result is "based on application requirements" (based on application requirements in step S72), the server system 1100 searches for a primary individual analytical evaluation AI model 6 that satisfies the application requirements from among the primary individual analytical evaluation AI models 6. Then, the server system 1100 sets it as the primary individual analytical evaluation AI model 6 to be used in the primary analytical evaluation (step S74; see FIG. 4).

[0203] If the result of the lottery is "random" (random in step S72), the server system 1100 randomly selects the evaluation items to be adopted in the primary analysis and evaluation (step S76). Specifically, the server system 1100 searches the evaluation item selection pattern data 510 (see FIG. 9) for pattern data that meets the applicable requirements, and randomly selects the selected evaluation item list as a candidate from the candidates. Alternatively, the server system 1100 may randomly select from among all types of evaluation items regardless of the evaluation item selection pattern data 510.

[0204] Then, the server system 1100 sets the primary individual analytical evaluation AI model 6 capable of evaluating the selected evaluation item as the primary individual analytical evaluation AI model 6 to be used in the primary analytical evaluation (step S78).

[0205] Returning to FIG. 13, the server system 1100 selects and determines the primary selection game situation information 751 from the game situation information 700 (step S90). Specifically, the server system 1100 references the evaluation item list from the primary individual analysis and evaluation AI model data 512 (see FIG. 9) of the primary individual analysis and evaluation AI model 6 used in the primary analysis and evaluation results. The evaluation item list indicates the types and number of evaluation items that the AI ​​model can evaluate. The server system 1100 searches the referenced evaluation item list for primary selected game status information selection criteria data 516 that meets the application requirements, and selects information and data to be used as primary selected game status information 751 according to the searched criteria data.

[0206] Next, the server system 1100 inputs the first selection game situation information 751 into the first individual analysis evaluation AI model 6 used in the first analysis evaluation 10 to obtain the first analysis evaluation result 12 (see FIG. 2) (step S92). The new first analysis evaluation result is added to the past first analysis evaluation result 731 (see FIG. 12).

[0207] Next, the server system 1100 searches for primary evaluation information definition data 520 that satisfies the applicable requirements based on the primary analysis evaluation result 12, and generates and determines primary evaluation information 14 in accordance with the definitions of the searched primary evaluation information definition data 520 (step S94). Then, the server system 1100 performs control to provide the primary evaluation information 14 to the player (step S96).

[0208] 14, the server system 1100 displays a predetermined feedback information input screen on the user terminal 1500 of the loop processing target player and accepts this (step S98). Then, the server system 1100 updates the primary individual analysis and evaluation AI model 6 used in step S92 based on the accepted feedback information (step S100), and ends loop A (step S102).

[0209] During the game, the server system 1100 monitors whether the secondary timing has occurred. If the secondary timing has occurred (YES in step S110), the server system 1100 executes loop B for each player (steps S112 to S128).

[0210] In loop B, the server system 1100 executes the AI ​​model selection process (see FIG. 15) for n=2 (step S114). Since "n" represents the order of analytical evaluation, the AI ​​model selection process for n=2 selects the secondary individual analytical evaluation AI model 7 to be used in the secondary analytical evaluation 20.

[0211] Next, the server system 1100 selects and determines the secondary selection game situation information 752 from the game situation information 700 (step S116). Specifically, the server system 1100 references the evaluation item list in the secondary individual analysis evaluation AI model data 514 (see FIG. 9) of the secondary individual analysis evaluation AI model 7 used in the secondary analysis evaluation 20. The evaluation item list indicates the types and number of evaluation items that the AI ​​model can evaluate. The server system 1100 selects information and data to be used as the secondary selection game status information 752 in accordance with the secondary selection game status information selection criteria data 518 that matches the referenced items, and determines the secondary selection game status information 752.

[0212] Next, the server system 1100 inputs the secondary selection game situation information 752 to the secondary individual analysis evaluation AI model 7 used in the secondary analysis evaluation result to obtain the secondary analysis evaluation result 22 (see FIG. 2) (step S118). The new secondary analysis evaluation result is added to the past secondary analysis evaluation result 732 (see FIG. 12).

[0213] Next, the server system 1100 searches for secondary evaluation information definition data 522 that meets the applicable requirements based on the secondary analysis evaluation results 22, and generates and determines secondary evaluation information 24 in accordance with the definitions of the searched secondary evaluation information definition data 522 (step S120).

[0214] Next, the server system 1100 performs control to provide the secondary evaluation information 24 to the player (step S122). Next, the server system 1100 displays a predetermined feedback information input screen on the user terminal 1500 of the player to be processed in loop B and accepts the feedback information (step S124). Then, based on the accepted feedback information, the server system 1100 updates the secondary individual analysis evaluation AI model 7 used in step S118 this time (step S126), and ends loop B (step S128).

[0215] The server system 1100 repeats steps S30 to S128 until the game end condition is met (NO in step S140). If the game end condition is met (YES in step S140), the server system 1100 executes a game end process that displays the game results, saves game save data, etc. (step S142), and ends the series of processes.

[0216] As described above, according to this embodiment, it is possible to provide a new technique that allows various evaluations of game play to be obtained.

[0217] In other words, if one wishes to obtain evaluations of a variety of gameplay, it is necessary to have multiple analytical evaluation means with different standards for analysis and evaluation. The game system 1000 is capable of providing a variety of evaluations to the player, and performs the analysis and evaluation based on selected game situation information selected from the game situation information, thereby reducing the processing load associated with the analysis and evaluation and enabling the efficient provision of a variety of evaluations.

[0218] Furthermore, the game system 1000 can select and operate an individual analysis evaluation to be performed from among a plurality of individual analysis evaluations. This allows for a variety of analysis results based on a variety of evaluation criteria to be obtained depending on the situation, and a variety of evaluation information to be provided to the player, preventing the game from becoming stale.

[0219] Furthermore, the game system 1000 can use the analysis and evaluation results on the input layer side as input on the output layer side to obtain further analysis and evaluation results on the output layer side, thereby enabling the computer system to provide hierarchical evaluation information to the player.

[0220] [Modification] Although examples of embodiments to which the present invention is applied have been described above, the forms to which the present invention can be applied are not limited to the above forms, and constituent elements can be added, omitted, or modified as appropriate.

[0221] (Variation 1) The game system 1000 is not limited to a client-server system. As shown in FIGS. 16 and 17, the game system 1000B may be a stand-alone user terminal 1500B. In this case, the user terminal 1500B executes a game program 503 (see FIG. 16) to realize the function of a processing unit 200 (see FIG. 17). The processing unit 200 has the same functional configuration as the server processing unit 200s (see FIG. 10). With this configuration, the same effects as those of the above embodiment can be obtained.

[0222] Furthermore, for example, the game system 1000 may be realized by a P2P (Peer to Peer) architecture for multiple user terminals 1500. In this case, programs and data according to the allocation of functions are stored in the user terminals 1500, and the functions of the processing unit 200 (see FIG. 17) are distributed and realized by the user terminals 1500 that serve as P2P nodes. With this configuration, the same effects as those of the above embodiment can be obtained.

[0223] (Variation 2) The function of the primary individual analytical evaluation AI model 6 is not limited to executing the primary analytical evaluation 10 and obtaining the primary analytical evaluation result 12. As shown in FIG. 18 , the primary individual analytical evaluation AI model 6 may have the function of executing the primary analytical evaluation 10 and obtaining the primary evaluation information 14.

[0224] In this case, the primary individual analysis and evaluation AI model data 512 (see Figure 9) of the primary individual analysis and evaluation AI model 6 is model data machine-learned based on training data that takes as input the information and data selected as the primary selection game situation information 751 and outputs the primary evaluation information 14.

[0225] The function of the secondary individual analytical evaluation AI model 7 is not limited to executing the secondary analytical evaluation 20 and obtaining the secondary analytical evaluation result 22. The secondary individual analytical evaluation AI model 7 may have the function of executing the secondary analytical evaluation 20 and obtaining the secondary evaluation information 24.

[0226] In this case, the secondary individual analysis evaluation AI model data 514 (see Figure 9) of the secondary individual analysis evaluation AI model 7 is model data machine-learned based on training data that takes as input the information or data selected as the secondary selection game situation information 752 and outputs the secondary evaluation information 24. [Explanation of symbols]

[0227] 6…Primary individual analysis and evaluation AI model 7…Secondary individual analysis and evaluation AI model 10...Primary analysis and evaluation 12...First analysis and evaluation results 14...Primary evaluation information 16...Provision of primary evaluation information 18…First feedback information acceptance 20...Secondary analysis and evaluation 22... Secondary analysis and evaluation results 24...Secondary evaluation information 26...Provision of secondary evaluation information 28...Second feedback information acceptance 200s...Server processing section 210...Player request reception control unit 220...Evaluation Item Selection Section 224...Information selection section 230...Analysis and Evaluation Department 231…First Individual Analysis and Evaluation Department 232…Secondary Individual Analysis and Evaluation Department 233...Input layer side individual analysis and evaluation unit 234...Output layer side individual analysis and evaluation unit 236…Individual Selection Section 240...Evaluation information provision control unit 241...Second Individual Selection Section 242...Feedback information reception control unit 244…AI Model Update Department 501...Server program 512…Primary individual analysis evaluation AI model data 514…Secondary individual analysis evaluation AI model data 516...First selection game situation information selection criteria data 518... Secondary selection game situation information selection criteria data 520...Primary evaluation information definition data 522...Secondary evaluation information definition data 700...Game status information 704...Player request setting data 731... Past primary analysis and evaluation results 732... Past secondary analysis evaluation results 741: Secondary analysis evaluation used AI model ID 742… Secondary analysis evaluation used AI model ID 751...First selection game status information 752... Second selection game status information 1000...Game System 1100...Server system 1500...User terminal

Claims

1. an analysis and evaluation means for analyzing and evaluating given game situation information; evaluation information provision control means for controlling the analysis and evaluation means to analyze and evaluate game situation information in which a player has played a game, and for providing evaluation information based on the analysis and evaluation results to the player; an information selection means for selecting selected game status information from the game status information of the player; Equipped with the evaluation information provision control means provides the player with the evaluation information based on the analysis and evaluation result obtained by causing the analysis and evaluation means to analyze and evaluate the selected game situation information. Computer system.

2. an analysis and evaluation means for analyzing and evaluating given game situation information; evaluation information provision control means for controlling the analysis and evaluation means to analyze and evaluate game situation information in which a player has played a game, and for providing evaluation information based on the analysis and evaluation results to the player; Equipped with The analytical evaluation means includes a plurality of individual analytical evaluation means, the evaluation information provision control means has an individual selection means for selecting a selected individual analysis and evaluation means from the plurality of individual analysis and evaluation means, and provides the evaluation information based on the analysis and evaluation results by the selected individual analysis and evaluation means to the player. Computer system.

3. an analysis and evaluation means for analyzing and evaluating given game situation information; evaluation information provision control means for controlling the analysis and evaluation means to analyze and evaluate game situation information in which a player has played a game, and for providing evaluation information based on the analysis and evaluation results to the player; Equipped with The analytical evaluation means includes: a plurality of individual analytical evaluation means; an individual selection means for selecting a selected individual analysis and evaluation means from the plurality of individual analysis and evaluation means based on the game situation information; wherein the analysis and evaluation of the game situation information is performed by the selective individual analysis and evaluation means. Computer system.

4. an analysis and evaluation means for analyzing and evaluating given game situation information; evaluation information provision control means for controlling the analysis and evaluation means to analyze and evaluate game situation information in which a player has played a game, and for providing evaluation information based on the analysis and evaluation results to the player; Equipped with the analysis and evaluation means has individual analysis and evaluation means for each layer from an input layer to an output layer, and performs hierarchical analysis and evaluation in which the individual analysis and evaluation results of the individual analysis and evaluation means on the input layer side are individually analyzed and evaluated by the individual analysis and evaluation means on the output layer side; Computer system.

5. the analysis and evaluation means is capable of analyzing and evaluating the game situation information based on a given selected evaluation item from a plurality of evaluation items; further comprising an evaluation item selection means for selecting the selected evaluation item from the plurality of evaluation items; the evaluation information provision control means controls the analysis and evaluation means to analyze and evaluate the game situation information based on the selected evaluation items; A computer system according to any one of claims 1 to 4.

6. the evaluation item selection means selects the selected evaluation item based on an operation by the player; 6. The computer system of claim 5.

7. the analysis and evaluation means is capable of analyzing and evaluating the game situation information based on a given selected evaluation item from a plurality of evaluation items; further comprising an evaluation item selection means for selecting the selected evaluation item from the plurality of evaluation items; the individual selection means selects the selected individual analysis and evaluation means based on the selected evaluation items; The evaluation information provision control means controls the selected individual analysis and evaluation means to perform analysis and evaluation based on the selected evaluation items.

3. The computer system of claim 2.

8. the evaluation item selection means selects the selected evaluation item based on an operation by the player; 8. The computer system of claim 7.

9. the individual selection means selects the selected individual analysis and evaluation means based on an operation by the player; 3. The computer system of claim 2.

10. further comprising an evaluation item selection means for selecting a selected evaluation item from a plurality of evaluation items; the analysis and evaluation means analyzes and evaluates the game situation information based on the selected evaluation items; A computer system according to any one of claims 1 to 4.

11. the evaluation item selection means selects the selected evaluation item based on the game situation information.

11. The computer system of claim 10.

12. the analysis and evaluation means performs the analysis and evaluation using an AI (Artificial Intelligence) model for analyzing and evaluating the game situation information. A computer system according to any one of claims 1 to 4.

13. feedback information receiving means for receiving feedback information based on an operation of the player in response to the evaluation information; an AI model update means for updating the AI ​​model based on the feedback information; The computer system of claim 12 further comprising:

14. The individual analytical evaluation means of the output layer are plural, the evaluation information provision control means generates the evaluation information based on analysis and evaluation results of the plurality of individual analysis and evaluation means in the output layer.

5. The computer system of claim 4.

15. the individual analysis and evaluation means performs the individual analysis and evaluation using an AI (Artificial Intelligence) model for individually analyzing and evaluating the game situation information; feedback information receiving means for receiving feedback information based on an operation of the player in response to the evaluation information; an AI model update means for updating the AI ​​model of the individual analysis and evaluation means based on the feedback information; Further comprising:

15. A computer system according to claim 4 or 14.

Images