Game delivery system, game delivery method, and game delivery program

Abstract

This system provides a game delivery system, a game delivery method, and a game delivery program that can provide appropriate opponents according to the user's situation and maintain the user's motivation. [Solution] The game provision system 1 includes a storage unit that stores the user's battle history and generation conditions, a generation unit that generates AI users adjusted for battles against the user based on the battle history, a storage unit that holds the AI ​​users, and a control unit that prioritizes battling AI users when a matching request is made by the user. This makes it possible to provide an appropriate opponent according to the user's situation.

Description

Technical Field

[0001] This embodiment relates to a game providing system, a game providing method, and a game providing program.

Background Art

[0002] In recent years, online battle games in which users located at distant places through a network play against each other have become widely popular. In such battle games, for example, fighting games and FPS (First Person Shooter), the movement of the characters operated by users greatly affects the victory or defeat.

[0003] Movement refers to a series of action guidelines for performing optimal operations according to the situation, such as the movement of characters, the timing of using attack skills, the judgment of defense and avoidance, and positioning. Users refine this movement through daily battles and find joy in improving their own skill levels.

[0004] In particular, battles against opponents whose strength is comparable to one's own, that is, opponents who are not too strong or too weak, bring about an appropriate sense of tension and concentration to the user. In battles against such opponents, it is difficult to have a one-sided development, and it is easy to discover the problems and improvement points of one's own movement. Therefore, it becomes a very important factor for improvement.

[0005] When winning a battle against a comparable opponent, the user can confirm that their movement was correct and can obtain a very great sense of achievement and motivation. Also, even in the case of a narrow defeat, the user can feel their own growth, and a desire to win next emerges, which serves as a motivation for continuous play.

[0006] Thus, in order to maintain the motivation of users and promote skill improvement in battle games, it is essential to match with opponents having an appropriate difference in strength. However, conventional matching systems mainly simply mechanically match users with similar ratings or ranks.

[0007] Therefore, depending on the user's performance and the compatibility between characters, matches could sometimes become one-sided. For example, there are systems that adjust the user's abilities after a losing streak, but these directly alter the game balance and do not adjust the relationship between the player and their opponent.

[0008] In the prior art described in Patent Document 1, at predetermined timings, operation data related to the techniques performed by the character in response to the operation of the input unit, and screen state data related to the screen display are collected and written to the learning data storage unit, and deep learning computation processing is performed based on the operation data and the screen state data to optimize the weights of the learning results. When a first character controlled by the player and a second character controlled by the computer are in battle, key data of the inference result is obtained for the second character, added to the key history, and the action (or technique) is determined by referring to the key history and command table. [Prior art documents] [Patent Documents]

[0009] [Patent Document 1] Japanese Patent Publication No. 2019-195512 [Overview of the project] [Problems that the invention aims to solve]

[0010] As mentioned above, in competitive games, opportunities to encounter worthy opponents of similar skill levels, with whom one can learn from one another, are limited as one's skill level improves. In particular, users who have moved beyond the beginner stage often find themselves overwhelmed and unilaterally defeated by advanced players with sophisticated tactics, while feeling unsatisfied when playing against beginners.

[0011] In such situations, it is difficult to gain a sense of accomplishment from victory or to learn from defeat, leading to a loss of opportunities to improve one's gameplay and a loss of motivation for the matches themselves. Therefore, this embodiment aims to provide a game provision system that can maintain user motivation by offering appropriate opponents according to the user's situation. [Means for solving the problem]

[0012] A game provision system according to a first aspect of this disclosure is a game provision system that generates an AI user to play against a user on behalf of another user in a game that allows users to play against each other using user terminals, and comprises: a storage unit that stores the in-game battle history for each user and generation conditions for determining whether or not to generate the AI ​​user; a generation unit that generates the AI ​​user adjusted for playing against the target user based on the battle history if there is a target user whose battle history satisfies the generation conditions; a storage unit that holds the AI ​​user generated by the generation unit in the storage unit as a target for playing against the target user until a matching request is received from the target user; a battle unit that causes the AI ​​user to play against the target user; and a control unit that receives a matchmaking request for a battle from the target user and, if the storage unit holds the AI ​​user corresponding to the target user, causes the battle unit to cause the AI ​​user to play against the target user with priority over other users.

[0013] A game provision method according to a second aspect of this disclosure is a game provision method executed by a computer that generates an AI user to play against a user on behalf of another user in a game that allows users to play against each other's user terminals, and includes the steps of: storing the in-game battle history for each user and generation conditions for determining whether or not to generate the AI ​​user; if there is a target user whose battle history satisfies the generation conditions, generating the AI ​​user adjusted for playing against the target user based on the stored battle history; holding the generated AI user in the storage unit as a target for playing against the target user until a matching request is received from the target user; and when a matching request for a match is received from the target user, if an AI user corresponding to the target user is held, having the AI ​​user play against the target user with priority over other users.

[0014] A game provision program according to a third aspect of this disclosure provides a computer that generates AI users to play against other users in a game where users' terminals can play against each other. This computer functions as follows: a storage unit that stores the in-game battle history for each user and generation conditions for determining whether or not to generate the AI ​​user; a generation unit that generates an AI user adjusted for playing against the target user based on the battle history stored in the storage unit, if there is a target user whose battle history satisfies the generation conditions; a storage unit that holds the AI ​​user generated by the generation unit in the storage unit as a target for playing against the target user until a matching request is received from the target user; a battle unit that causes the AI ​​user to play against the target user; and a control unit that receives a matchmaking request from the target user and, if the storage unit holds an AI user corresponding to the target user, causes the battle unit to prioritize the AI ​​user against the target user over other users. [Effects of the Invention]

[0015] According to this embodiment, it is possible to provide a game providing system that can provide an appropriate opponent for a user according to the user's situation and maintain the user's motivation.

Brief Description of Drawings

[0016] [Figure 1] It is a diagram showing the overall configuration of the game providing system 1 according to an embodiment of the present disclosure. [Figure 2] It is a block diagram showing the functional configuration of the game providing server 100 according to an embodiment of the present disclosure. [Figure 3] It is a flowchart showing the main loop of system-wide monitoring and matching control in the game providing system 1 according to an embodiment of the present disclosure. [Figure 4] It is a detailed flowchart of an internal process for generating an AI user according to an embodiment of the present disclosure. [Figure 5] It is a diagram showing a configuration example of the battle history data 500 for managing the statistical results of the game play of each user according to an embodiment of the present disclosure. [Figure 6] It is a diagram showing a configuration example of the generation condition data 510 for determining whether to generate an AI user according to an embodiment of the present disclosure, and the difficulty target at that time. [Figure 7] It is a diagram showing a configuration example of the battle detail data 520 according to an embodiment of the present disclosure. [Figure 8] It is a diagram showing a configuration example of the AI user data 530 according to an embodiment of the present disclosure. [Figure 9] It is a diagram showing an example of the matching screen 610 according to an embodiment of the present disclosure. [Figure 10] It is a diagram showing an example of the battle screen 620 according to an embodiment of the present disclosure. [Figure 11] It is a block diagram exemplifying the hardware configuration of a computer for realizing the game providing server 100 and the user terminal 200 according to an embodiment of the present disclosure.

Best Mode for Carrying Out the Invention

[0017] Hereinafter, the overall configuration of the game providing system 1 according to this embodiment and each device will be described in detail with reference to the drawings.

[0018] FIG. 1 is a diagram showing the overall configuration of the game providing system 1 according to this embodiment. The game providing system 1 of this embodiment is a distributed game providing system composed of a plurality of nodes connected via a communication network, and provides an online battle game environment for users. The entire system is configured by connecting a game providing server 100 and a plurality of user terminals 200 operated by users who are players so as to be able to communicate with each other via a network NW.

[0019] The game provided by the game providing system 1 may be any game as long as it enables battles between users. The game may include a command input type, a so-called fighting game type, a card game that battles based on a pre-constructed deck, a real-time strategy that progresses in real time, and PvP (Player vs Player) in MMO (Massively Multiplayer Online).

[0020] The network NW is an arbitrary communication network that enables information transmission. For example, it is constructed by the Internet, a dedicated line network, a VPN, a mobile communication network such as LTE or 5G, a wireless LAN, etc., or any combination thereof. The network NW has sufficient bandwidth and redundancy to support high-capacity and low-latency real-time communication occurring between the game providing server 100 and a large number of user terminals 200.

[0021] The user terminal 200 is a client device for a user to play a competitive game according to this embodiment. The user terminal 200 can be implemented as, for example, a smartphone, a tablet computer, a personal computer, a dedicated home game console, or a portable game console.

[0022] The user terminal 200 executes the game client program using its built-in processor. It then performs tasks such as rendering the game space, receiving user input, sending and receiving data with the game server 100, and local calculations for matchmaking. The game server 100 oversees the entire game service, managing numerous user accounts, synchronizing game progress, and performing matchmaking.

[0023] Furthermore, it is a backend system that generates and controls AI users, which is a key function of this embodiment. The game provision server 100 may be configured as a single physical server. Alternatively, to enhance scalability and availability, it may be virtually built on a cloud computing environment consisting of a group of servers located behind a load balancer.

[0024] Figure 2 is a block diagram showing the functional configuration of the game provision server 100 according to this embodiment. The game provision server 100 comprises a communication unit 110, a storage unit 120, and a processing unit 130 as functional units. The communication unit 110 transmits and receives various data necessary for the progress of the game to and from the user terminal 200 via the network NW. Specifically, it transmits and receives matching requests, user operation inputs, game status update data, and match results, etc.

[0025] The memory unit 120 holds the application program 121 and the database 122. The application program 121 is a set of instructions for realizing the various functions of the game provision server 100. The database 122 is constructed using a data management system such as a relational database or a NoSQL database. It is structured to efficiently search and update profile information and play history of a large number of users.

[0026] The processing unit 130 includes a generation unit 131, a storage unit 132, a battle unit 133, a control unit 134, and a display control unit 135. These units work closely together via shared memory and inter-process communication mechanisms. This enables real-time and seamless execution of a series of advanced information processing tasks, from monitoring the user's battle history to dynamically generating AI users when conditions are met, intervening in matching, and executing battles.

[0027] The generation unit 131 generates an AI user adjusted for playing against a target user based on their match history, provided that the target user's match history meets the generation conditions. Here, "user" means player, operator, or account holder.

[0028] Furthermore, an AI user refers to a bot player, virtual opponent, or a character automatically controlled by a computer program that behaves like a human. A user targeted for generation refers to a real user who meets certain conditions (such as a slump or winning streak) and becomes the target of system intervention.

[0029] A match history refers to information that includes not only a record of wins and losses, but also detailed play logs such as the character used, the opponent's character, match duration, damage dealt, damage received, frequency of skill use, and movement patterns.

[0030] The generation unit 131 adjusts the AI ​​user based on the target win rate stored in the memory unit 120 and the defeat data included in the battle history, which includes at least one of the status, behavior patterns, and characters used by other users against whom the target user was defeated.

[0031] The defeat data contains the reasons why the target user lost, and by starting with information about the opponent who defeated the user, it is possible to efficiently generate AI users that the target user can narrowly win against.

[0032] The target win rate refers to the estimated win rate that the system intends for the generated user when the AI ​​user plays against the generated user. In addition, the generation unit 131 adjusts the frequency or precision of action patterns selected during a match, in addition to or instead of adjusting the numerical values ​​of the status when adjusting the AI ​​user. The numerical values ​​of the status refer to the basic ability values ​​of the character, such as HP (Hit Points), attack power, defense power, and speed.

[0033] The behavioral patterns during a match refer to tactical tendencies and movement habits, such as being attack-oriented, defense-oriented, or primarily using long-range attacks. The frequency or accuracy of selection refers to the probability of using a particular skill, the accuracy of hitting an attack, the success rate of guarding or dodging, etc. The generation unit 131 adjusts the AI ​​user within a predetermined range so that the target user does not feel any discomfort during a match.

[0034] Furthermore, the generation unit 131 generates the AI ​​user's name and icon during the generation of the AI ​​user. The AI ​​user's name refers to the username or handle name displayed in the game, and the icon refers to the avatar image or symbol image used to identify the user.

[0035] Furthermore, the generation unit 131 stores the match results between the AI ​​user and the target user in the storage unit 120 as a new match history for the target user. Then, when the target user again meets the generation conditions, the generation unit 131 adjusts the adjustment parameters of the AI ​​user, giving more weight to the most recent match results. The adjustment parameters refer to various variables used to determine the strength and behavioral characteristics of the AI ​​user.

[0036] In addition, the generation unit 131 may generate a virtual user that reproduces the numerical values ​​of the target user's status and their behavior patterns during matches, based on the target user's match history. The generation unit 131 adjusts the AI ​​user based on the win rate calculated by virtually having this virtual user play against the AI ​​user. A virtual user refers to a bot that imitates the abilities and habits of the target user for simulation purposes.

[0037] The generation unit 131 may increase the weighting of the results of winning and losing streaks in the match history. This can improve the accuracy of creating a close match between the target user and the AI ​​user.

[0038] The storage unit 132 stores the AI ​​users generated by the generation unit 131 in the memory unit 120. The battle unit 133 has the AI ​​users battle against the generated target users. Here, battle means the act of characters fighting each other based on game rules to determine a winner and loser.

[0039] The control unit 134 receives a matchmaking request from the user to be generated, and if the storage unit 132 has an AI user corresponding to the user to be generated, it instructs the match unit 133 to have the AI ​​user play against the user to be generated, prioritizing them over other users. A matchmaking request means a request that a user sends to the system to find an opponent.

[0040] Furthermore, the control unit 134 initiates a match between the AI ​​user and the target user without notifying the target user that the match is against an AI user. In addition, the control unit 134 causes the AI ​​user to send messages or stamps created using a pre-configured personality model or generated AI to the target user during or before / after the match.

[0041] When the battle unit 133 has the AI ​​user play against the target user, the display control unit 135 displays the AI ​​user on the user terminal 200 in a way that makes it indistinguishable from other users. Specifically, when playing against an AI user, the display control unit 135 displays the AI ​​user's name and icon in the same format as other users.

[0042] Figure 3 is a flowchart showing the main loop for overall system monitoring and matchmaking control in the game provision system. In step S102, the processing unit 130 of the game provision server 100 receives the win / loss result and match details each time a match for each user ends. The match details include the characters used and action patterns. The processing unit 130 then stores the received data as a new record in the match history data of the database 122 and accumulates it.

[0043] In step S104, the generation unit 131 scans the updated battle history data periodically or at specific triggers such as the end of a game. The generation unit 131 then determines whether the user's battle record satisfies any of the generation conditions defined in the generation condition data. The generation conditions may include the loss rate included in the battle history of the target user exceeding a predetermined loss threshold.

[0044] Here, the defeat threshold refers to a standard value used to determine if a user is losing excessively. Furthermore, the generation conditions may include at least one of the following: the target user has lost more than a predetermined number of times consecutively, or has won more than a predetermined number of times consecutively. The defeat threshold may include types that emphasize the number of most recent consecutive losses and types that emphasize the overall number of losses.

[0045] If a user who meets the generation conditions is detected in step S104, the process proceeds to step S106. In step S106, the generation unit 131 executes a subroutine to generate an AI user for that user. In the subsequent step S108, the storage unit 132 stores and retains the parameter set and profile information of the custom-made AI user generated in step S106 in the database 122.

[0046] In step S110, the control unit 134 monitors whether it has received a matchmaking request from any online user terminal 200 via the communication unit 110. If a matchmaking request is received, in step S112, the control unit 134 searches and determines whether the user who sent the request is a target user. That is, the control unit 134 determines whether dedicated AI user data associated with that user exists in the database 122.

[0047] If no corresponding AI user is found in step S112, the process proceeds to the normal matching process. On the other hand, if it is determined that an AI user is found, the process proceeds to step S114. In step S114, the control unit 134 instructs the battle unit 133 to assign the dedicated AI user that is found as the matching opponent, with priority given to other real users in the waiting queue. At this time, the control unit 134 may instruct the battle unit 133 to start a match between the AI ​​user and the target user without notifying the target user that the match is against an AI user.

[0048] Figure 4 is a detailed flowchart of the internal process for generating AI users. In step S122, the generation unit 131 retrieves the past match history of the identified target user from the database 122. At this time, it extracts not only win / loss statistics but also loss data that shows specifically who the user lost to.

[0049] In step S124, the generation unit 131 analyzes the types of characters and tactical tendencies that the user struggles with, based on the extracted defeat data. It then creates a prototype of the AI ​​user based on this. In step S126, the generation unit 131 obtains a target win rate corresponding to the activated generation conditions, that is, an estimate of the user's win probability, from the generation condition data.

[0050] In step S128, the generation unit 131 adjusts the numerical values ​​of the AI ​​user's status as the first step in approaching the target win rate. For example, if the target win rate is high, it performs calculations to reduce the original HP and attack power to 0.8 times the standard value. The memory unit 120 stores the target win rate, which is an estimate of the target user's win rate when the AI ​​user plays against the target user.

[0051] On the other hand, if the user's win rate exceeds the win threshold in the generation conditions (e.g., during a winning streak), the generation unit 131 may generate an AI user designed to cause the user to experience defeat or difficulty. In this case, the generation unit 131 sets a low target win rate (e.g., 30%), strengthens the AI ​​user's status to 1.2 times the baseline value, or adjusts it to frequently select behavioral patterns that the user has previously struggled with and lost to. This provides an appropriate obstacle to prevent the user from becoming overconfident and stimulates their desire to challenge themselves.

[0052] In step S130, the generation unit 131 further adjusts the behavior patterns for more human-like behavior and precise difficulty control. Specifically, in adjusting the AI ​​user, the generation unit 131 may adjust the frequency or precision of selecting behavior patterns during a match, in addition to or instead of adjusting the numerical values ​​of the status. For example, the generation unit 131 sets variables in the AI's decision-making logic, such as lowering the probability of guarding an attack or increasing the probability of missing a powerful move.

[0053] In step S132, the generation unit 131 determines whether the AI ​​user that has undergone adjustments up to this point is truly at a level that meets the target win rate. Preferably, this determination is made by running a high-speed virtual battle simulation in the background against a virtual user that mimics the user's behavior and evaluating the convergence value of the statistical win rate. That is, the generation unit 131 generates a virtual user that reproduces the numerical values ​​of the target user's status and the behavior patterns during battle.

[0054] The generation unit 131 may then adjust the AI ​​user based on the win rate calculated by virtually having this virtual user play against the AI ​​user. If the judgment in step S132 determines that the target win rate has not been met, the process loops back to step S126 or S128, and further fine-tuning of the status and behavior patterns is repeated.

[0055] If it is determined that the target win rate falls within an acceptable range, in step S134, the generation unit 131 automatically generates and assigns a human-like display name and icon image to the completed internal AI logic. This creates a dataset of completed AI users that are visually indistinguishable from real users.

[0056] Thus, according to one embodiment of the present invention, a game provision system is provided comprising: a storage unit 120 that stores the in-game battle history of each user and the generation conditions for determining whether or not to generate an AI user; a generation unit 131 that generates an AI user adjusted for playing against a target user based on the battle history when there is a target user whose battle history satisfies the generation conditions; a storage unit 132 that stores the AI ​​user generated by the generation unit 131 in the storage unit 120; a battle unit 133 that makes the AI ​​user play against the target user; and a control unit 134 that receives a matchmaking request for a battle from the target user and, if the storage unit 132 holds an AI user corresponding to the target user, makes the battle unit 133 prioritize playing the AI ​​user against the target user over other users.

[0057] This allows the system to automatically detect user win / loss records and biases in their performance, and at the appropriate time, generate a dedicated AI user matched to that user's level, prioritizing that user for matchmaking.

[0058] Furthermore, according to this embodiment, the generation conditions are at least one of the following: the loss rate included in the match history of the target user exceeds a predetermined loss threshold, and the win rate exceeds a predetermined win threshold. This makes it possible to objectively evaluate the fluctuations in the user's performance during gameplay using a clear and quantitative indicator such as the win-loss ratio, and to accurately determine the optimal timing for intervening with the AI ​​user.

[0059] Furthermore, according to this embodiment, the memory unit 120 stores a target win rate, which is an estimate of the target user's win rate when the AI ​​user plays against the target user, serving as a basis for adjusting the AI ​​user. The generation unit 131 adjusts the AI ​​user based on the target win rate and defeat data, which includes at least one of the status, behavior patterns, and characters used by other users that the target user lost to in battle, as included in the battle history. This makes it possible to control the game balance precisely based on a concrete indicator such as the win rate, rather than making vague difficulty adjustments for the user.

[0060] Furthermore, according to this embodiment, when the battle unit 133 has an AI user play against a target user, the system further includes a display control unit 135 that displays the AI ​​user on the user terminal 200 in a way that makes it indistinguishable from other users. As a result, the user can immerse themselves in the game believing that they are playing against a genuine human opponent, without realizing that their opponent is a program generated by the system.

[0061] Furthermore, according to this embodiment, the control unit 134 initiates a match between the AI ​​user and the target user without notifying the target user that they are playing against an AI user. This ensures that the user's immersion in the player-versus-player match is maintained without any interruption during the seamless transition process from successful matchmaking to the start of the match.

[0062] Furthermore, according to this embodiment, in adjusting the AI ​​user, the generation unit 131 adjusts the frequency or precision of selecting action patterns during battle, in addition to or instead of adjusting the numerical values ​​of the status. This makes it possible to adjust the difficulty of the AI ​​user not only through mechanical methods such as increasing or decreasing parameters, but also through a more human-like approach such as making mistakes or frequently using certain techniques, thereby providing a more realistic battle experience.

[0063] Furthermore, according to this embodiment, the generation unit 131 adjusts the AI ​​user within a predetermined range that does not cause discomfort to the user during a match. This prevents the game balance from being disrupted or the world view from being destroyed by drastically changing the strength of the AI ​​user, and protects the user's perception that "the opponent is a human."

[0064] Furthermore, according to this embodiment, the generation unit 131 generates the name and icon of the AI ​​user when generating the AI ​​user, and the display control unit 135 displays the name and icon of the AI ​​user in the same format as other users when playing against the AI ​​user. This allows the AI ​​user to have a personality that blends into the game's worldview and the culture of other users, thereby enhancing the user's sense of realism towards their opponent.

[0065] Furthermore, according to this embodiment, the generation conditions include at least one of the following: the target user has lost a predetermined number of times consecutively, or has won a predetermined number of times consecutively. This allows for real-time detection and immediate intervention of sudden spikes in user frustration or signs of rapid boredom, which cannot be captured by long-term statistical indicators such as overall win rates.

[0066] Furthermore, according to this embodiment, the control unit 134 causes the AI ​​user to send messages or stamps created using a pre-set personality model or generating AI to the target user during or before / after a match. This extends the human-like qualities of the AI ​​user not only to in-game movements but also to the level of communication, thereby promoting deeper emotional engagement with the user.

[0067] Furthermore, according to this embodiment, the generation unit 131 stores the match results between the AI ​​user and the target user in the storage unit 120 as a new match history for the target user, and when adjusting the adjustment parameters of the AI ​​user to be generated when the target user again meets the generation conditions, it gives more weight to the most recent match results. This allows the system to learn the impact of matches with the AI ​​user it has generated and form a self-adaptive feedback loop that continuously improves the accuracy of subsequent interventions.

[0068] Furthermore, according to this embodiment, the generation unit 131 generates a virtual user that reproduces the target user's status values ​​and behavior patterns during matches based on the target user's match history, and adjusts the AI ​​user based on the win rate calculated by virtually having the virtual user play against the AI ​​user. This allows the system to perform a simulation before actually having the user and the AI ​​user play against each other, dramatically improving the quality and reliability of difficulty adjustment.

[0069] Figure 5 shows an example of the configuration of the battle history data 500, which manages the statistical performance of each user's gameplay. The battle history data 500 includes the following items: user ID, total number of battles, win rate, loss rate, consecutive wins, consecutive losses, and recent battle history pointer.

[0070] The User ID is the primary key used to uniquely identify a user within the system. Total number of matches indicates the total number of matches a user has played from the time their account was created until the present. Win rate and loss rate indicate the proportion of wins and losses relative to the total number of matches, respectively. Consecutive wins and consecutive losses are indicators used to measure a user's short-term psychological state; these values ​​are incremented if the current result is the same as the previous one and reset if it reverses. The recent match history pointer provides a list of references to detailed log data for individual matches.

[0071] This allows us to trace not only statistical figures, but also the specific micro-context of who a user fought against and how. For example, if a user is on a losing streak against a particular character, their difficulty against that character is accumulated as data and becomes an important indicator when generating the next AI user. Furthermore, by managing data with pointers, it becomes possible to quickly retrieve necessary data from a vast amount of battle logs.

[0072] Figure 6 shows an example of the configuration of generation condition data 510, which defines the generation conditions for determining whether or not to generate an AI user, and the difficulty target in that case. The generation condition data 510 has the following items: condition ID, defeat threshold, win threshold, consecutive defeat threshold, consecutive win threshold, and target win rate.

[0073] The Condition ID is a key that uniquely identifies each generation condition. The Loss Threshold indicates a threshold value for determining whether a user's loss rate is excessive. The Win Threshold indicates a threshold value for determining whether a user's win rate is excessive. The Consecutive Loss Threshold and Consecutive Win Threshold indicate threshold values ​​for determining whether to generate an AI user based on the number of consecutive losses or wins. The Target Win Rate is an indicator of the strength of the generated AI user, and indicates the probability that a user will win against the AI ​​user if the condition is met.

[0074] For example, condition "CND-001" indicates that the game will be triggered when the loss rate exceeds 60% and the number of consecutive losses reaches 3, with a target win rate set at 70% to allow the user to experience a victory. Conversely, condition "CND-004" is triggered when the win rate reaches 90% and the user has had 10 consecutive wins, a situation that could bore the user, with a target win rate set at 10% to allow the user to experience a defeat.

[0075] Figure 7 shows an example of the structure of the battle details data 520, which records the specific details of an individual match. The battle details data 520 includes the following items: battle ID, user ID, opponent status, opponent's character, and opponent's action pattern.

[0076] The Match ID is a unique identifier for each match. The User ID indicates the user who played that match. Opponent Status shows the specific ability values ​​such as HP and ATK (attack power) of the enemy character the user faced in that match. Opponent's Character Used indicates the type of character the opponent selected. Opponent's Action Pattern shows the opponent's tactical characteristics and play style during the match, such as the frequency of skill use and how they maintained distance.

[0077] For example, in the match with ID "DEF-8812," it can be seen that the user lost to an enemy with a pattern of frequent skill usage. The generation unit 131 refers to this record and fine-tunes the overall parameters while reproducing the user's weaknesses, such as making the next AI user's behavior pattern also have a higher skill usage frequency.

[0078] Figure 8 shows an example of the configuration of AI user data 530, which is dynamically generated by the system and awaits its turn for matching. The AI ​​user data 530 includes the following items: AI user ID, target user ID, display name, icon ID, status adjustment value, and behavior pattern parameter.

[0079] The AI ​​User ID is an ID that uniquely identifies the generated hypothetical opponent. The Generation Target User ID is linking information indicating which user this AI user was created for. The Display Name and Icon ID are profile information used to disguise the AI ​​as indistinguishable from other real users. The Status Adjustment Value indicates a multiplier to the parameters of the base character. The Behavior Pattern Parameters indicate the variables passed to the AI's decision-making engine.

[0080] For example, "AIU-9901" has its evasion rate for Skill X set to low, indicating that it has been intentionally adjusted to make it easier for the user's special moves to hit. These databases work together organically to enable a series of advanced personalization processes, from detecting macro-level statistical conditions to reproducing micro-level tactics and deploying completely disguised opponents.

[0081] Figure 9 shows an example of the matching screen 610 displayed on the user terminal 200 just before the match begins after the opponent has been determined. The words "MATCH FOUND!" and "VS" are prominently displayed in the center of the screen, indicating that the match has been established.

[0082] On the left side of the screen, the user's own information is displayed, including the user's icon 612, name, and current rank. On the right side of the screen, the opponent's information is displayed. What is important here is that even if this opponent is actually an AI user generated by the system, the display format uses the exact same layout, design, and font as the user's own information.

[0083] The opponent's icon 614 uses a typical illustration, and the name displayed is a natural, human-like handle. In this way, when the battle unit 133 has the AI ​​user play against the generated target user, the display control unit 135 displays the AI ​​user on the user terminal 200 in a way that makes it indistinguishable from other users.

[0084] The rank information is displayed as Bronze, a plausible-looking spoof created by the system to match the actual user's rank range. Text and a progress bar are displayed at the bottom of the screen, functioning as a visual effect to make it appear as if the connection with other users over the network is synchronized.

[0085] Figure 10 shows an example of a match screen 620 during an actual match. The screen is divided vertically, with the opponent's HP gauge and character model at the top and the user's own HP gauge and character model at the bottom. At the very bottom of the screen is an operation section 626 for user input.

[0086] When a user taps a button, the operation data is sent to the game server 100, and the battle unit 133 calculates the win / loss and damage in real time. In this battle screen, the AI ​​user acts at regular intervals just like a human, sometimes making mistakes and sometimes cleverly dodging the user's attacks.

[0087] When the opponent's HP is significantly reduced by the user's attack, an icon 622 and a speech bubble message 624 will suddenly appear in the opponent's area. Icon 622 is a stamp image representing the AI ​​user's current emotional state, such as a confused face or a crying face.

[0088] Message 624 is a specific text that corresponds to the battle situation, such as "I can't believe it's this strong...!". In connection with this configuration, the control unit 134 may have the AI ​​user send messages or stamps created using a pre-set personality model or generating AI to the target user during or before / after a match.

[0089] This makes it possible to create the illusion that the AI ​​user is playing with emotions on the other side of the screen. For example, it can send a message of praise like "Amazing!" immediately after the user performs a great combo, or congratulate the AI ​​user on a good fight with "That was a good match" when they achieve a come-from-behind victory, thereby satisfying the user's need for recognition.

[0090] During a match, receiving such human-like reactions from the opponent gives the user an overwhelming sense of superiority as they corner their opponent, and a strong feeling of the raw, visceral impact unique to player-versus-player combat.

[0091] <Another Embodiment 1> According to one embodiment of the present invention, the generation unit 131 may adjust the AI ​​user based on a target win rate and victory data which includes at least one of the status, behavior patterns, and characters used by other users that the target user has defeated in battles, as included in the battle history.

[0092] For example, instead of generating an AI user with the same behavioral patterns as opponents that the user has previously defeated—so-called "easy targets"—for a user who is on a winning streak, the system generates an AI user with characteristics not found in the user's winning data, or opponents that the user struggled against. This ensures that the user always has fresh and challenging matches.

[0093] <Another Embodiment 2> According to one embodiment of the present invention, the generation unit 131 may generate an AI user based on a user who is strong in relative terms from the perspective of all users, but who is incompatible with the character used and control habits of the target user, and who is easy to win against from the perspective of the target user.

[0094] For example, an AI user can be created by basing it on the behavioral models of "strong characters" or "advanced players" who are generally high-ranking, while also incorporating characteristics such as weaker defenses against attack patterns that the target user excels at, or by selecting characters with attributes that give the target user's character an advantage. This allows the user to experience a great sense of accomplishment from "defeating a superior opponent through skill," and can boost their motivation without even realizing the existence of an AI user.

[0095] <Other Embodiment 3> According to one embodiment of the present invention, the generation unit 131 may use a stress value estimated from the user's biometric information or operation log as a generation condition. For example, if the generation unit 131 determines that the user is irritated or anxious based on information such as the strength and irregularity of taps acquired from the user terminal 200, or the heart rate acquired from a wearable device, it immediately generates an AI user (a friendly AI user) that the user can easily win against. This allows for consideration of the user's emotions before the outcome of the game is determined, and prevents the user from abandoning the game.

[0096] <Other Embodiments 4> According to one embodiment of the present invention, if the game provided is in a format in which teams consisting of multiple users compete against each other, the generation unit 131 may generate AI users not as opponents, but as allies belonging to the same team as the user being generated.

[0097] For example, if it is estimated that the reason a user is on a losing streak is due to "insufficient coordination with team members" or "insufficient skills on teammates," the generation unit 131 generates an AI user with excellent coordination abilities that can accurately perform support actions such as healing and defense (tanking). Then, the control unit 134 places this AI user as an ally of the generated user during matchmaking and has them play a match in the battle unit 133. As a result, the user can get the feeling that "they played a key role in leading the team to victory" (the so-called feeling of carrying the team), which can alleviate the frustration and stress that is unique to team battles regarding teammates and help maintain motivation.

[0098] <Another Embodiment 5> According to one embodiment of the present invention, the generation unit 131 may refer to the possession information of the target user of in-game items or skins stored in the storage unit 120 and generate an AI user that is intentionally equipped with a specific item or skin that the target user does not possess.

[0099] For example, the generation unit 131 generates an AI user equipped with the latest powerful weapon that is compatible with the character used by the target user, and has them battle in the battle unit 133. In this battle, the control unit 134 creates a scenario in which the AI ​​user uses the unique powerful skill of that item, or in which the user is defeated after a close battle due to the power of that item. This not only adjusts the outcome of the battle, but also allows the user to experience the appeal and usefulness of items they do not yet possess in actual battle, and gives the AI ​​user the role of an advertising tool that naturally stimulates the desire to purchase (motivation to pay).

[0100] <Other Embodiments 6> According to one embodiment of the present invention, when the battle unit 133 has the AI ​​user and the generated target user compete against each other, the control unit 134 may perform processing to intentionally simulate network communication delays (lag) and packet loss in relation to the actions and displays of the AI ​​user.

[0101] Unlike humans, computer-controlled AI users have extremely fast reaction times and completely stable communication, which poses a risk of users suspecting that the other party is a system (AI). Therefore, the control unit 134 adds random delays to the AI ​​user's input timing and intentionally causes behaviors characteristic of poor network conditions, such as instantaneous teleportation (warping), to occur during character movement. This reinforces the perception that the other party is a real user playing under a human-specific communication environment, completely preventing the AI ​​user from being exposed.

[0102] <Other Embodiments 7> According to one embodiment of the present invention, the generation unit 131 may store in the storage unit 120 the identification information and profile of a specific AI user that the user to be generated has previously defeated in a close battle, and may bring them back as a continuous "rival".

[0103] Specifically, using the increase in the rank and level of the target user as a new generation condition, the generation unit 131 reads data of past AI users from the memory unit 120, enhances (promotes) the status and behavior patterns to match the current skill level of the target user, and then regenerates an AI user with the same name and icon. The control unit 134 then matches the user with this enhanced rival AI user again. This allows the user to feel a sense of rivalry, as if "a former opponent has grown stronger and is challenging me again," providing a deep sense of immersion in the game world and a desire to continue playing.

[0104] <Other Embodiments 8> According to one embodiment of the present invention, the generation unit 131 may generate an AI user that has the characteristics of a mentor (instructor) that helps the user improve their playing skills, rather than simply being an opponent.

[0105] Specifically, the generation unit 131, when it identifies from the battle history data that the target user has lost due to being unable to deal with a particular attack pattern or combo, generates an AI user that frequently uses that particular attack action but is adjusted to either "make the preparatory movements clearer (larger)" or "slightly delay the timing of the attack" compared to a normal player. By playing against this AI user in the battle unit 133, the user can naturally learn the timing of dodging and defending against attacks they are weak against in actual battle. As a result, the user can experience a strong sense of accomplishment and growth from devising countermeasures and winning with their own skills, and it is possible to maintain motivation through more advanced friendly play.

[0106] Figure 11 is a block diagram illustrating the hardware configuration of a computer 900 for realizing a game provision server 100 and a user terminal 200. The computer 900 comprises a control unit 901, a storage unit 902, a communication unit 903, an input unit 904, and an output unit 905.

[0107] These components are electrically connected via bus 910 and exchange data and control signals with each other. The control unit 901 is a processor that provides overall control over the operation of the entire computer 900.

[0108] The control unit 901 reads and executes the operating system and the application program for providing the game according to this embodiment from the storage unit 902, thereby realizing various functional blocks.

[0109] The memory unit 902 is a storage device that holds programs executed by the control unit 901 and data to be processed, and includes main memory and auxiliary storage. The communication unit 903 is a communication interface that connects to a network NW and transmits and receives data with other computer devices according to a communication protocol. The input unit 904 is a device that receives instructions and data input from users and system administrators. The output unit 905 is a device that outputs processing results, game screens, etc.

[0110] • Note 1 In a game that allows users to compete against each other using their respective user terminals, in a game provision system that generates an AI user to compete against the user on behalf of another user, A storage unit that stores the in-game battle history for each user and the generation conditions for determining whether or not to generate an AI user, If there is a target user whose battle history satisfies the generation conditions, the generation unit generates an AI user adjusted for battles against the target user based on the battle history, A storage unit that stores the AI ​​user generated by the generation unit in the storage unit as a target user to be matched with until a matching request is received from the target user, A battle unit that has the aforementioned AI user play against the aforementioned target user, If the system receives a matchmaking request from the user to be generated, and the storage unit holds the AI ​​user corresponding to the user to be generated, the battle unit will have a control unit that prioritizes the AI ​​user over other users in battles with the user to be generated. A game delivery system equipped with these features.

[0111] • Appendix 2 The game provision system as described in Appendix 1, wherein the generation condition is at least one of the following: the loss rate included in the match history of the target user exceeds a predetermined loss threshold, and the win rate exceeds a predetermined win threshold.

[0112] • Appendix 3 The memory unit stores a target win rate, which is an estimate of the win rate of the target user when the AI ​​user plays against the target user, and serves as a basis for adjusting the AI ​​user. The game provision system according to any one of the appendices 1 to 2, wherein the generation unit adjusts the AI ​​user based on the target win rate and defeat data which includes at least one of the status, behavior patterns, and characters used by other users against whom the target user has lost in the match history.

[0113] • Appendix 4 The game provision system according to any one of the appendices 1 to 3, further comprising a display control unit that displays the AI ​​user on the user terminal in a manner that makes it indistinguishable from other users when the battle unit has the AI ​​user and the generated target user compete against each other.

[0114] • Appendix 5 The game provision system according to any one of the appendices 1 to 4, wherein the control unit causes the battle unit to start a battle between the AI ​​user and the target user without notifying the target user that the battle is against the AI ​​user.

[0115] • Appendix 6 The generation unit, in addition to or instead of adjusting the numerical values ​​of the status, adjusts the frequency or precision of selecting action patterns during a match, as described in any of Appendix 1 to 5 of the game provision system.

[0116] • Appendix 7 The generation unit performs adjustments to the AI ​​user within a predetermined range so that the user to be generated does not feel any discomfort during the match, as described in any of Appendix 1 to 6 of the game provision system.

[0117] • Appendix 8 The generation unit generates the name and icon of the AI ​​user in the generation of the AI ​​user. The display control unit displays the name and icon of the AI ​​user in the same format as other users when playing against the AI ​​user, as described in any of Appendix 1 to 7 of the game provision system.

[0118] • Appendix 9 The game provision system described in any of Appendix 1 to 8, wherein the generation conditions include at least one of the following: the user to be generated has lost a predetermined number of times in a row, and has won a predetermined number of times in a row.

[0119] • Appendix 10 The game provision system according to any one of the appendices 1 to 9, wherein the control unit causes the AI ​​user to send messages or stamps created using a pre-set personality model or generating AI to the target user during or before / after a match.

[0120] • Appendix 11 The game provision system according to any one of the appendices 1 to 10, wherein the generation unit stores the match results between the AI ​​user and the target user as a new match history for the target user in the storage unit, and when the target user again satisfies the generation conditions, it increases the weighting of the most recent match results when adjusting the adjustment parameters of the AI ​​user that are generated.

[0121] • Appendix 12 The game provision system according to any one of the appendices 1 to 11, wherein the generation unit generates a virtual user that reproduces the numerical status and behavioral patterns of the target user based on the target user's battle history, and adjusts the AI ​​user based on the win rate calculated by virtually having the virtual user battle the AI ​​user.

[0122] • Appendix 13 A game provision method performed by a computer that generates an AI user to play against a user on behalf of another user in a game that allows users to compete against each other using user terminals, The steps include storing in the storage unit the in-game battle history for each user and the generation conditions for determining whether or not to generate the AI ​​user, If there is a target user whose battle history satisfies the generation conditions, the step of generating an AI user adjusted for battles against the target user based on the stored battle history, The steps include: storing the generated AI user in the storage unit as a target user for matching against the target user until a matching request is received from the target user; When a matchmaking request for a match is received from the user to be generated, and an AI user corresponding to the user to be generated is held, the step is to have the AI ​​user play against the user to be generated with priority over other users. A method of providing games that includes this.

[0123] • Appendix 14 In a game that allows users to compete against each other using their respective user terminals, a computer that generates an AI user to compete against the user on behalf of other users is defined as follows: A storage unit that stores the in-game battle history for each user and the generation conditions for determining whether or not to generate an AI user. If there is a target user whose battle history satisfies the generation conditions, the generation unit generates an AI user adjusted for battle against the target user based on the battle history stored in the storage unit. A storage unit that stores the AI ​​user generated by the generation unit in the storage unit as a target user to be played against until a matching request is received from the target user. A battle unit that causes the aforementioned AI user to play against the aforementioned target user, If the system receives a matchmaking request from the user to be generated, and the storage unit holds the AI ​​user corresponding to the user to be generated, the battle unit will have the control unit prioritize the AI ​​user over other users in a battle against the user to be generated. A game provision program designed to function as such. [Explanation of symbols]

[0124] 1. Game delivery system 100 game server 110 Communications Department 120 Storage section 121 Application Programs 122 Databases 130 Processing Unit 131 Generation part 132 Holding part 133 Match Division 134 Control Unit 135 Display Control Unit 200 user terminals 500 match history data 510 Generation Condition Data 520 Match Details 530 AI user data 610 Matching screen 612 icons 614 icons 620 Battle screen 622 icons 624 messages 626 Operation section 900 Computers 901 Control Unit 902 Storage section 903 Communications Department 904 Input Section 905 Output section 910 Bus NW Network

Claims

1. In a game that allows users to compete against each other using their respective user terminals, in a game provision system that generates an AI user to compete against the user on behalf of other users, A storage unit that stores the in-game battle history for each user and the generation conditions for determining whether or not to generate an AI user, If there is a target user whose battle history satisfies the generation conditions, the generation unit generates the AI ​​user adjusted for battles against the target user based on the battle history, A storage unit that stores the AI ​​user generated by the generation unit in the storage unit as a target for matching with the target user until a matching request is received from the target user, A battle unit that has the AI ​​user play against the generated target user, If the system receives a matchmaking request from the user to be generated, and the storage unit holds the AI ​​user corresponding to the user to be generated, the battle unit will have a control unit that prioritizes the AI ​​user over other users in battles with the user to be generated. A game delivery system equipped with these features.

2. The game provision system according to claim 1, wherein the generation condition is at least one of the following: the loss rate included in the match history of the target user exceeds a predetermined loss threshold, and the win rate exceeds a predetermined win threshold.

3. The memory unit stores a target win rate, which is an estimate of the win rate of the target user when the AI ​​user plays against the target user, and serves as a basis for adjusting the AI ​​user. The game provision system according to claim 1, wherein the generation unit adjusts the AI ​​user based on the target win rate and defeat data which includes at least one of the status, behavior patterns, and characters used of other users against whom the target user has lost in the match history.

4. The game provision system according to claim 1, further comprising a display control unit that, when the battle unit has the AI ​​user and the generated target user compete against each other, displays the AI ​​user on the user terminal in a manner that makes it indistinguishable from other users.

5. The game provision system according to claim 1, wherein the control unit causes the battle unit to start a battle between the AI ​​user and the target user without notifying the target user that the battle is against the AI ​​user.

6. The game provision system according to claim 1, wherein the generation unit, in addition to or instead of adjusting the numerical values ​​of the status, adjusts the frequency or precision of selecting action patterns during a match when adjusting the AI ​​user.

7. The game provision system according to claim 1, wherein the generation unit adjusts the AI ​​user within a predetermined range so that the user to be generated does not feel any discomfort during the match.

8. The generation unit generates the name and icon of the AI ​​user in the generation of the AI ​​user. The game provision system according to claim 4, wherein the display control unit displays the name of the AI ​​user and the icon of the AI ​​user in the same format as other users when playing against the AI ​​user.

9. The game provision system according to claim 1, wherein the generation conditions include at least one of the following: the user to be generated has lost a predetermined number of times in a row, and has won a predetermined number of times in a row.

10. The game provision system according to claim 1, wherein the control unit causes the AI ​​user to send a message or stamp created using a pre-set personality model or generating AI to the target user during or before / after a match.

11. The game provision system according to claim 1, wherein the generation unit stores the match results between the AI ​​user and the target user as a new match history for the target user in the storage unit, and when the target user again satisfies the generation conditions, it increases the weighting of the most recent match results when adjusting the adjustment parameters of the AI ​​user that are generated.

12. The game provision system according to claim 1, wherein the generation unit generates a virtual user that reproduces the numerical values ​​of the status of the target user and the behavioral patterns during matches based on the match history of the target user, and adjusts the AI ​​user based on the win rate calculated by having the virtual user virtually play against the AI ​​user.

13. A game provision method performed by a computer that generates an AI user to play against a user on behalf of another user in a game that allows users to compete against each other using user terminals, The steps include storing in the storage unit the battle history within the game for each user, and the generation conditions for determining whether or not to generate the AI ​​user, If there is a target user whose match history satisfies the generation conditions, the step of generating an AI user adjusted for playing against the target user based on the stored match history, The steps include: storing the generated AI user in the storage unit as a target user for matching against the generated target user until a matching request is received from the target user; When a matchmaking request for a match is received from the user to be generated, and the AI ​​user corresponding to the user to be generated is held, the AI ​​user is given priority over other users to play against the user to be generated. A method of providing games that includes this.

14. In a game that allows users to compete against each other using their respective user terminals, a computer that generates an AI user to compete against the user on behalf of other users is provided. A storage unit that stores the in-game battle history for each user and the generation conditions for determining whether or not to generate an AI user. If there is a target user whose battle history satisfies the generation conditions, the generation unit generates the AI ​​user adjusted for battles against the target user based on the battle history stored in the storage unit. A storage unit that stores the AI ​​user generated by the generation unit in the storage unit as a target for matching with the target user until a matching request is received from the target user. A battle unit that causes the AI ​​user to play against the generated target user, If the system receives a matchmaking request from the user to be generated, and the storage unit holds the AI ​​user corresponding to the user to be generated, the battle unit will have a control unit that prioritizes the AI ​​user over other users in a battle against the user to be generated. A game provision program designed to function as such.

Images