Computer system and program

Abstract

To provide technology capable of improving amusement as technology for utilizing an AI in an entertainment field.SOLUTION: A support character 4 with a character AI 8 associated therewith is set to be a reception character 6 for receiving an order. When an order is inputted, an initial answer is generated by using the character AI 8 of the reception character 6, and is presented to a user. If the initial answer does not pass, another support character 4 different from the reception character 6 is made to be a secondary answer character 7. Then, a secondary answer is generated by using the character AI 8 of the secondary answer character 7, and is presented to the user. A process in which the secondary answer is generated and presented to the user is accompanied with a search execution performance 20, a counseling performance 24, and a re-answer presentation performance 26.SELECTED DRAWING: Figure 5

Description

[Technical field]

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

[0002] In recent years, artificial intelligence (AI) based on machine learning has come to be used in various situations. For example, Patent Document 1 discloses a technology for an AI service usage support system that uses multiple AIs. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2019-28950 A Summary of the Invention [Problem to be solved by the invention]

[0004] AI is often used for practical purposes, but it should also be possible to use AI for entertainment purposes. However, if conventional AI technology is applied to mechanical response control, the answer will be truly "mechanical," and will lack entertainment value.

[0005] The problem that the present invention aims to solve is to provide a technology that utilizes AI in the entertainment field and can improve interest. [Means for solving the problem]

[0006] The first invention for solving the above problem includes an answerer object setting means (e.g., the game management unit 210, the answerer object setting unit 212, the support character setting data 610 in FIG. 14, step S14 in FIG. 17) for setting a plurality of answerer objects (e.g., the answerer object 5 in FIG. 2) capable of answering a given order for which an answer is required in a given virtual world in association with an answer automatic generation model (e.g., the character AI8 in FIG. 2) that automatically generates an answer for the answerer object; a reception means (e.g., the server processing unit 200s in FIG. 10, the game management unit 210 in FIG. 11, the order reception control unit 214 in FIG. 11, the received order data 710 in FIG. 15, step S42 in FIG. 17) for receiving an order from a user to a reception object (e.g., the reception character 6 in FIG. 2) which is one of the plurality of answerer objects; an initial response generation control means (for example, the initial response generation control unit 218 in FIG. 11, the response data 720 in FIG. 15, and step S52 in FIG. 17) for generating an initial response to the reception order by using the automatic response generation model of the reception object; a secondary answer generation control means (e.g., the secondary answer generation control unit 220 in FIG. 11, the answer data 720 in FIG. 15, step S162 in FIG. 18) for selecting a secondary answer object that is an answerer object other than the reception object when the initial answer does not satisfy a given passing condition, and generating a secondary answer for the reception order using the answer automatic generation model of the secondary answer object; A computer system comprising:

[0007] An "answer object" may be, for example, an object designed as a character such as a human, animal, or robot, and visible in a specific display form on the display screen, or it may be a fictional object that does not have a specific display form and outputs answers in the form of text, voice, etc.

[0008] The "automatic answer generation model" is data for establishing and functioning an AI. For example, it may be a machine-learned AI model or a rule-based AI model.

[0009] The computer system of the first invention receives an order from a user using one of the answerer objects as a reception object. An initial answer to the reception order is generated using an automatic answer generation model corresponding to the reception object. If the initial answer does not satisfy the pass condition, a secondary answer is generated using the automatic answer generation model of the secondary answer object.

[0010] From the user's point of view, this flow seems to show that the reception object re-handles the reception order that it was unable to provide an answer that meets the criteria for passing by relying on the secondary answer object. This creates an interesting feeling of "human-like" behavior, as if the user was stuck for an answer and went to ask someone else for the answer. In other words, this technology can be used to utilize AI in the entertainment field to improve the level of interest.

[0011] A second invention is a computer system in which, in the above-mentioned computer system, the respondent object is a character appearing in the virtual world, and the reception means receives the reception order in the form of a conversation between the character of the reception object and the user (for example, reception through the order input field 13 of Figure 4).

[0012] The second invention is suitable for realizing visually “human-like” and interesting responses by the reception object in the entertainment field such as games in which characters appear in a virtual space or the metaverse.

[0013] A third invention is a computer system in which, in the above-mentioned computer system, attributes are set for the character, and the automatic answer generation model of the respondent object is set based on the attributes of the respondent character, which is the character of the respondent object.

[0014] The contents of the "attributes" can be set appropriately depending on the setting of the worldview of the virtual world expressed in the virtual space, the role of the answerer object in the virtual world, etc. For example, if the virtual world is a fantasy world with magic and monsters, the answerer object will be a character that appears in the virtual world, so the attributes can be set to the character's occupation (e.g., magician, warrior, healer, etc.), level, sex, race, etc.

[0015] According to the third invention, the answer automatic generation model is set based on the attribute of the corresponding answerer character. For example, the answer automatic generation model of a magician attribute character can be set to be good at answering based on the knowledge and experience of the magician attribute, but not good at answering based on the knowledge and experience of other attributes (for example, weapon craftsman). The content of the answer will take on the characteristics unique to the attribute of that character, and a technology that can further increase interest as a technology that utilizes AI in the entertainment field can be realized.

[0016] A fourth invention is a computer system in which the secondary answer generation control means selects the secondary answer object based on an attribute of the character of the secondary answer object (for example, attribute point Pz as a variable of the suitable person selection point in Figure 6).

[0017] According to the fourth invention, the computer system selects a secondary answer object reflecting the attributes of the character. From the user's point of view, it looks like the reception object has obtained a secondary answer by consulting a secondary answer object with an appropriate attribute. In other words, it looks like the person to consult has been properly selected. It is possible to express "human-likeness" in a series of processes related to the answer, which can increase the interest.

[0018] A fifth aspect of the present invention is the computer system further comprising a first relationship determination means (e.g., UC relationship point Puc in FIG. 7, first relationship determination unit 222 in FIG. 11, UC relationship data 630 in FIG. 14, step S22 in FIG. 17) for determining a relationship between the user and a reception character that is the character of the reception object, The initial answer generation control means is a computer system that generates a primitive answer for the reception order using the automatic answer generation model of the reception object (e.g., primitive answer generation process 34 in FIG. 7), and generates the initial answer by processing the primitive answer based on the relationship (e.g., primitive answer processing process 36 in FIG. 7).

[0019] "Relationships" include, for example, friendly relationships, master-servant relationships, hostile relationships, parent-child relationships, etc. According to the fifth invention, the computer system can generate an initial answer by processing the original answer based on the relationship between the user and the reception character. In other words, even if the reception order has the same content, the content of the initial answer can be changed depending on the relationship between the user and the reception character. Depending on the relationship with the person with whom they are talking (for example, the quality of their friendship), people may talk politely or carelessly. According to the fifth invention, it is possible to simulate such a person's conversation tendency. It is possible to express "human-likeness" in a series of processes related to the answer, thereby improving interest.

[0020] A sixth invention is a computer system further comprising a first relationship determination means for determining a relationship between the user and a reception character which is the character of the reception object, and the initial answer generation control means processes the reception order based on the relationship (e.g., order processing process 32 in FIG. 7) and generates an initial answer to the processed reception order using the automatic answer generation model of the reception object (e.g., original answer generation process 34 in FIG. 7).

[0021] According to the sixth invention, the computer system processes the reception order based on the relationship between the user and the reception character, and can generate an initial answer from the processed reception order. In other words, even if the reception order has the same content, the content of the initial answer can be changed depending on the relationship between the user and the reception character. Depending on the relationship with the person they are talking to (for example, the quality of their friendship), people may talk politely or carelessly. According to the sixth invention, it is possible to simulate such a person's conversation tendency. It is possible to express "human-likeness" in a series of processes related to the answer, and it is possible to improve interest.

[0022] A seventh invention is a computer system further comprising a first relationship determination means for determining a relationship between the user and a reception character which is the character of the reception object, and a pass condition setting change means for changing the strictness of the pass condition based on the relationship to set the pass condition (for example, the pass condition setting change unit 230 in FIG. 11, the pass condition data 716 in FIG. 15, step S106 in FIG. 18).

[0023] According to the seventh invention, the computer system can change the strictness of the pass condition based on the relationship between the user and the reception character. Even for the same reception order, the initial answer may be accepted or rejected depending on the relationship between the user and the reception character. People may have different tolerances for answers depending on their relationship with the other person. According to the seventh invention, it is possible to simulate such human tendencies. It is possible to express "human-likeness" in a series of processes related to answers, which can increase interest.

[0024] An eighth invention is a computer system further comprising a first relationship determination means for determining a relationship between the user and a reception character which is the character of the reception object, and the secondary answer generation control means selects the character of the secondary answer object based on the relationship.

[0025] According to the eighth invention, the computer system can select the character of the secondary answer object based on the relationship between the user and the reception character. The selection of the character of the secondary answer object affects the quality of the secondary answer. A person may select who to ask for help in answering the content of a consultation based on the relationship (for example, intimacy) with the person consulted. Specifically, if a consultation is from a close person, the person may choose a collaborator with specialized knowledge who can provide the best possible answer because it is an intimate person and it is okay to take time, but if a consultation is from a distant person, the person may choose a collaborator who is close to them so that they can provide an answer quickly. According to the eighth invention, such a tendency of a person can be simulated. A series of processes related to the answer can be expressed as "human-like," and interest can be improved.

[0026] A ninth aspect of the present invention is the computer system further comprising a second relationship determination means (for example, the second relationship determination unit 224 in FIG. 11, the CC relationship data 632 in FIG. 14, step S22 in FIG. 17) for determining 1) the relationship between the user and a secondary answer character which is the character of the secondary answer object (for example, the UF relationship point Puf in FIG. 6), or 2) the relationship between a reception character which is the character of the reception object and the secondary answer character (for example, the CC relationship point Pcc in FIG. 6), The secondary answer generation control means is a computer system that generates an original answer for the received order using the automatic answer generation model of the secondary answer object (for example, original answer generation process 34f in Figure 8), and generates the secondary answer by processing the original answer based on the relationship determined by the second relationship determination means (for example, original answer processing process 36f in Figure 8).

[0027] According to the ninth aspect of the present invention, the computer system can process the original answer based on the relationship between the user and the secondary answer character, or the relationship between the reception character and the secondary answer character, to generate a secondary answer. In other words, even if the reception order has the same content, the content of the secondary answer can be changed depending on these relationships.

[0028] Depending on the relationship with the other person (e.g., the quality of the friendship), a person may respond politely or carelessly, and this is reflected in the answer to the consultation. According to the ninth invention, such human tendencies can be simulated. This makes it possible to express "human-likeness" in the series of processes related to the answer, thereby improving interest.

[0029] A tenth aspect of the present invention is the computer system further comprising a second relationship determination means (for example, the second relationship determination unit 224 in FIG. 11, the CC relationship data 632 in FIG. 14, step S22 in FIG. 17) for determining 1) a relationship between the user and a secondary answer character that is the character of the secondary answer object, or 2) a relationship between a reception character that is the character of the reception object and the secondary answer character, The secondary answer generation control means is a computer system that processes the received order based on the relationship determined by the second relationship determination means (e.g., order processing process 32f in FIG. 8), and generates the secondary answer for the processed received order using the automatic answer generation model of the secondary answer object (e.g., original answer generation process 34f in FIG. 8).

[0030] According to the tenth aspect of the present invention, the computer system processes the received order based on the relationship between the user and the secondary answer character, or the relationship between the receiving character and the secondary answer character, and generates a secondary answer from the processed received order. In other words, even if the received order has the same content, the content of the secondary answer can be changed depending on these relationships. This makes it possible to express "human-likeness" in the series of processes related to the answer, thereby improving interest.

[0031] In an eleventh aspect of the present invention, in the computer system, the secondary answer generation control means, when the secondary answer for the Xth time does not satisfy the pass condition, selects the secondary answer object for the (X+1)th time, and repeats a process of generating the secondary answer for the (X+1)th time using the answer automatic generation model of the secondary answer object for the (X+1)th time (for example, a loop realized from step S140 in FIG. 18 to NO in step S194 in FIG. 19), A second relationship determination means for determining 1) a relationship between the user and an X-th secondary answer character that is the character of the secondary answer object for the Xth time, or 2) a relationship between a reception character that is the character of the reception object and the X-th secondary answer character; A pass condition setting change means (for example, the pass condition setting change unit 230 in FIG. 11, the pass condition data 716 in FIG. 15, and step S184 in FIG. 19) that changes the strictness of the pass condition based on the relationship determined by the second relationship determination means to set the pass condition; The computer system further comprises:

[0032] According to the eleventh invention, the computer system can change the strictness of the pass condition based on the relationship between the user and the secondary answer character related to the Xth secondary answer, and the relationship between the reception character and the secondary answer character related to the Xth secondary answer. Even for the same reception order, depending on these relationships, the secondary answer may be passed or rejected. People may have different tolerances for answers depending on their relationship with the other person. According to the eleventh invention, it is possible to simulate such human tendencies. It is possible to express "humanness" in a series of processes related to answers, thereby improving interest.

[0033] A twelfth aspect of the present invention is the computer system described above, further comprising a third relationship determination means (for example, the third relationship determination unit 226 in FIG. 11, the CC relationship data 632 in FIG. 14, step S22 in FIG. 17) for determining 1) a relationship between the user and each of the characters (for example, UF relationship points Puf and UC relationship points Puc in FIG. 6), or 2) a relationship between a reception character that is the character of the reception object and each of the characters other than the reception character (for example, CC relationship points Pcc in FIG. 6), The secondary answer generation control means is a computer system that selects a secondary answer character, which is the character to be the secondary answer object, based on the relationship determined by the third relationship determination means.

[0034] According to the twelfth aspect of the present invention, the computer system can select a character to be a secondary answer object based on the relationship between the user and each character and the relationship between the receiving character and another character. According to the twelfth aspect of the present invention, it is possible to improve the interest of a series of processes related to an answer.

[0035] A thirteenth aspect of the present invention is the computer system as described above, wherein the user owns a plurality of the characters, and the computer system further comprises a reception character selection means (for example, the reception character selection unit 232 in FIG. 11, the reception character ID 618 in FIG. 14, step S54 in FIG. 17) for selecting a reception character to be the reception object from among the characters owned by the user based on an operation by the user, The secondary answer generation control means is a computer system that selects a secondary answer character, which is the character to be the secondary answer object, from among the characters owned by the user (for example, step S154 in FIG. 18).

[0036] According to the thirteenth aspect of the present invention, a receiving character can be selected from among the characters owned by the user to answer an initial answer, and a secondary answering character can be selected from among the characters owned by the user to answer a secondary answer. In other words, the process can be performed as if the characters owned by the user, which are like close friends, consult with each other to generate an answer, which can increase the interest. In addition, the effect of encouraging the user to acquire a new character can be expected.

[0037] A fourteenth invention is a computer system according to the above, wherein the reception object is associated with the user, and further comprising a model change control means (e.g., the model change control unit 234 in FIG. 11, the user information 602 in FIG. 14, the play performance data 620, step S202 in FIG. 19) for changing the automatic answer generation model of the reception object based on information of the user.

[0038] "User information" includes, for example, attribute information of the user, information about actions that the user has taken to date, and information about the results of those actions. According to the fourteenth aspect of the present invention, the computer system can change the automatic answer generation model of the reception object based on the user information. Even for the same order and the same reception object, the answer can change depending on the user information. According to the fourteenth aspect of the present invention, the interest of a series of processes related to the answer can be improved.

[0039] A fifteenth aspect of the present invention is the computer system according to the above, wherein the virtual world is accessible to a plurality of users, each of the users is associated with a respondent object, and the system further comprises a fourth relationship determination means (for example, the fourth relationship determination unit 228 in FIG. 11, the UU relationship data 634 in FIG. 14, step S22 in FIG. 17) for determining a relationship between the users; The secondary answer generation control means is a computer system that selects a user other than the user of the reception object based on the relationship between the user of the reception object and the other user, and selects the answerer object of the other user as the secondary answer object.

[0040] According to the fifteenth aspect of the present invention, the computer system selects another user based on the relationship between the user of the reception object and the other user, and selects the answerer object of the other user as a secondary answer object, thereby making it possible to improve the interest of a series of processes related to answers.

[0041] A 16th invention is a computer system further comprising: a secondary answer generation control means for selecting the secondary answer object for the (X+1)th time when the secondary answer for the Xth time does not satisfy the pass condition, and repeatedly performing a process of generating the secondary answer for the (X+1)th time using the automatic answer generation model of the secondary answer object for the (X+1)th time, and a pass condition setting change means for changing the strictness of the pass condition to set the pass condition based on the relationship between the user of the reception object and the user of the secondary answer object for the Xth time (for example, the pass condition setting change unit 230 in FIG. 11, the pass condition data 716 in FIG. 15, step S184 in FIG. 19).

[0042] According to the sixteenth aspect of the present invention, the computer system can change the strictness of the pass condition based on the relationship between the user and the user related to the Xth secondary answer, thereby making the series of processes related to the answer more interesting.

[0043] A seventeenth invention is a computer system in which the secondary answer generation control means generates an original answer to the reception order using the automatic answer generation model of the secondary answer object of the other user (for example, original answer generation process 34f in FIG. 8), and generates the secondary answer by processing the original answer based on the relationship between the user of the reception object and the other user.

[0044] According to the seventeenth invention, the computer system can generate a secondary answer by processing an original answer generated using an automatic answer generation model of a secondary answer object of another user based on a relationship between the user of the reception object and the other user. In other words, even if the reception order has the same content, the content of the secondary answer can be changed depending on the relationship. Depending on the relationship with the other person (for example, the quality of the friendship), a person may respond to the consultation politely or rudely, and this is reflected in the answer to the consultation. According to the seventeenth invention, such a person's tendency can be simulated. This can improve the interest of a series of processes related to the answer.

[0045] An 18th invention is a computer system in which the secondary answer generation control means processes the received order based on the relationship between the user of the reception object and the other user, and generates a primitive answer to the received order from the secondary answer to the processed received order using the automatic answer generation model of the secondary answer object of the other user (for example, order processing process 32f in Figure 8, primitive answer generation process 34f, step S164 in Figure 18).

[0046] According to the eighteenth aspect of the present invention, the computer system processes the reception order based on the relationship between the user of the reception object and other users, and generates a secondary answer from the processed reception order. In other words, even if the reception order has the same content, the content of the secondary answer can be changed depending on the relationship. This makes it possible to increase the interest of a series of processes related to the answer.

[0047] A 19th invention is a computer system further comprising an initial answer presentation control means (e.g., the answer presentation unit 14 in FIG. 4, the initial answer presentation control unit 240 in FIG. 11, step S100 in FIG. 18) for presenting the initial answer to the user as if the receiving object itself had devised it.

[0048] According to the nineteenth aspect of the present invention, the user feels as if the reception object itself is thinking of the answer and responding, which can increase the interest of a series of processes related to the answer.

[0049] A twentieth invention is a computer system further comprising a selected comment output control means (for example, the comment text in the re-answer presentation unit 27 in FIG. 5, the selected comment output control unit 242 in FIG. 11, step S168 in FIG. 19) for outputting a comment to the user as if the receiving object itself had selected the secondary answer object.

[0050] According to the twentieth aspect of the present invention, the user feels as if the reception object has selected a person to consult with in order to obtain a secondary answer, which can improve the interest of a series of processes related to an answer.

[0051] A 21st invention is a computer system further comprising a search execution comment output control means (for example, the comment text in the speech bubble 22 in the search execution performance 20 in Figure 5, the search execution comment output control unit 244 in Figure 11, step S148 in Figure 18) for causing the receiving object itself to output a search execution comment to the user requesting an answer from an answerer object other than the receiving object if the initial answer does not satisfy the pass condition.

[0052] According to the twenty-first aspect of the present invention, it is possible to improve the interest of a series of processes relating to an answer.

[0053] A twenty-second aspect of the present invention is the computer system, comprising: initial answer presentation means (for example, the answer presentation unit 14 in FIG. 4, the initial answer presentation control unit 240 in FIG. 11, step S100 in FIG. 18) for presenting the initial answer to the user; A first evaluation acquisition means (for example, the evaluation input unit 15 in FIG. 4, the first evaluation acquisition control unit 250 in FIG. 11, or step S104 in FIG. 18) for acquiring an evaluation of the user for the initial answer; The computer system further includes a pass / fail determination means (for example, pass / fail determination unit 254 in FIG. 11, step S110 in FIG. 18) that determines whether or not the pass condition is satisfied based on the evaluation.

[0054] According to the twenty-second aspect of the present invention, the computer system can present an initial answer to a user, obtain an evaluation thereof, and determine whether or not the pass condition is met based on the evaluation.

[0055] A 23rd aspect of the present invention is a computer system comprising: a second evaluation acquisition means (e.g., a second evaluation acquisition control unit 252 in FIG. 11, step S108 in FIG. 18) for presenting the received order and the initial response to a given survey target user and acquiring an evaluation of the initial response; The computer system further includes a pass / fail determination means (for example, pass / fail determination unit 254 in FIG. 11, step S110 in FIG. 18) that determines whether or not the pass condition is satisfied based on the evaluation.

[0056] According to the twenty-third aspect of the present invention, the computer system can entrust the evaluation of the initial response to a given survey target user.

[0057] A 24th invention is a program for causing a computer system to function as an answerer object setting means for setting, in a given virtual world, a plurality of answerer objects capable of answering a given order requesting an answer, in correspondence with an answer automatic generation model that automatically generates the answer for the answerer object, a reception means for receiving from a user a reception order, which is the order for a reception object that is one of the plurality of answerer objects, an initial answer generation control means for generating an initial answer to the reception order using the answer automatic generation model of the reception object, and a secondary answer generation control means for selecting a secondary answer object that is the answerer object other than the reception object when the initial answer does not satisfy a given passing condition, and generating a secondary answer to the reception order using the answer automatic generation model of the secondary answer object.

[0058] According to a twenty-fourth aspect of the present invention, it is possible to realize a program that enables a computer system to achieve the same effects as the above-mentioned aspects. [Brief description of the drawings]

[0059] [Figure 1] FIG. 1 is a system configuration diagram showing an example of the configuration of a game system. [Diagram 2] FIG. 1 is a diagram for explaining a game. [Diagram 3] FIG. 13 is a diagram for explaining the correspondence between support characters and character AI. [Figure 4] A diagram to explain orders, responses, and evaluations. [Diagram 5] FIG. 13 is a diagram for explaining the transition of the effect display related to a re-answer. [Figure 6] FIG. 11 is a diagram for explaining selection of a character to be used in generating an answer. [Figure 7] FIG. 13 is a diagram for explaining degradation of an initial response. [Figure 8] FIG. 13 is a diagram for explaining degradation of re-answers. [Figure 9] FIG. 13 is a diagram for explaining cancellation of generation of a re-answer. [Figure 10] FIG. 2 is a block diagram showing an example of the functional configuration of the server system. [Figure 11] FIG. 4 is a diagram showing the functional units of a game management unit. [Figure 12] FIG. 4 is a diagram showing examples of programs and data stored in a server storage unit. [Figure 13] FIG. 4 is a diagram showing an example of the data structure of character AI definition data. [Figure 14] FIG. 4 is a diagram showing an example of a data configuration of user management data. [Figure 15] FIG. 4 is a diagram showing an example of the data structure of play data. [Figure 16] FIG. 4 is a functional block diagram showing an example of the functional configuration of a user terminal. [Figure 17] 6 is a flowchart illustrating a flow of processing executed by the server system. [Figure 18] Flowchart continuing from Figure 17. [Figure 19] Flowchart continuing from Figure 18. [Figure 20] 11 is a flowchart for explaining the flow of a degradation response generating process. [Figure 21] 10 is a flowchart illustrating the flow of a second evaluation acquisition process. [Figure 22] A figure to explain the transition of the presentation display related to a re-answer when a user owns multiple support characters. [Diagram 23] FIG. 11 is a diagram showing an example of the functional configuration of a user terminal according to the second embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0060] 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.

[0061] [First embodiment] 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 for executing a given online game in which users 2 using user terminals 1500 play by controlling their own player characters.

[0062] The game system 1000 is a computer system including a server system 1100 and user terminals 1500 (1500a, 1500b, ...) connected to each other so as to be capable of data communication via a network 9. The user terminals 1500 serve as a man-machine interface (MMIF).

[0063] The network 9 refers to a communication path that allows data communication. In other words, the network 9 includes a dedicated line (dedicated cable) for direct connection, a LAN (Local Area Network) using Ethernet (registered trademark), a telephone communication network, a cable network, the Internet, and the like.

[0064] The server system 1100 is a game server that performs various processes such as providing a predetermined registration procedure, managing registered user information, and generating and managing various data for executing the game.

[0065] The server system 1100 has a control board 1150 mounted on a main unit. The control board 1150 is mounted 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. Some or all of the functions mounted on the control board 1150 may be realized by an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or a SoC (System on a Chip).

[0066] Although the server system 1100 is depicted as if it were a single server device, it may be realized by a configuration of multiple devices. For example, the server system 1100 may be configured with multiple blade servers that share functions and are connected to each other via an internal bus so that data communication is possible. The server system 1100 may also include a database and online storage.

[0067] The user terminal 1500 is a computer system that is used by the user 2 to play the game as a player, and serves as a man-machine interface in the game system 1000.

[0068] 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, tablet computers, and the like.

[0069] The user terminal 1500 includes an operation input device, an image display device, and a control board 1550. Examples of the operation input device include a touch panel 1506, a keyboard, a game controller, a mouse, etc. Examples of the image display device include a touch panel 1506, a head-mounted display, a glass-type display, etc.

[0070] 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 enable reading and writing of data and sending and receiving of signals. A part or all of the control board 1550 may be configured with an ASIC, an FPGA, or an SoC.

[0071] 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 a predetermined application program to realize the functions of a man-machine interface for the game system 1000 and as a game client.

[0072] The user terminal 1500 is configured to download the programs and various data necessary to play the game from the server system 1100, but it may also be configured to read them from a storage medium such as a memory card obtained separately by the user.

[0073] FIG. 2 is a diagram for explaining the game executed by the game system 1000. As shown in FIG. The game executed by the game system 1000 is a game in which a user 2 controls a user character 3 and travels through a fantasy virtual world together with a support character 4. Specifically, the game system 1000 is a multiplayer online role playing game (MORPG) in which the ultimate goal is to accomplish a certain objective while defeating monsters encountered on a game map. Note that, although the following description will be given assuming that one user 2 plays using one user character 3, the game may also be one in which one user 2 uses multiple user characters 3.

[0074] At least one support character 4 is associated with one user 2 (one user character 3).

[0075] The support character 4 is an NPC (Non Player Character). A number of different types of support characters 4 with different exterior designs are prepared in advance, and when the game starts, a selection list is presented to the user 2, and the type of support character 4 selected by the user 2 from among them is initially set. In the example of FIG. 2, the support character 4a is initially set for the user 2. The user 2 is treated as owning the support character 4a. Note that support characters 4 of types that were not initially set can be additionally owned as prizes in lotteries related to the game (such as so-called pay-to-play gacha) or as event rewards within the game.

[0076] During the game, the support character 4 of the user 2 accompanies the user character 3 and fights together with him when a battle occurs. It also handles conversations during the journey and responds to orders from the user 2. In other words, the support character 4 associated with the user 2 is an answerer object 5 that can answer a given order that requires an answer, and a reception character 6 that acts as a reception desk that accepts orders from the associated user 2.

[0077] A character AI 8 is associated with the support character 4. When an order is accepted, a response is generated using the character AI 8 associated with the support character 4 and is presented to the user 2.

[0078] FIG. 3 is a diagram for explaining the correspondence between the support character 4 and the character AI 8. As shown in FIG. At least one character AI 8 (8a, 8b, ...) is associated with one support character 4. The character AI 8 (8a, 8b, ...) is a machine learning model constructed by natural language-based machine learning. The character AI 8 learns from natural language text and outputs (answers) in natural language in response to inputs (orders) in natural language. In other words, the character AI is an automatic answer generation model that automatically generates answers to orders that require an answer.

[0079] A plurality of types of character AI 8 (8a, 8b, ...) are prepared with different learning fields, learning methods, learning amounts, etc. A plurality of learning fields are prepared. The learning fields are matters related to the game world, such as the geography of the game world, knowledge of monsters, knowledge of leveling up, knowledge of magic, knowledge of items, knowledge of lore in the game world that is a hint for game strategy, etc. The classification and contents of the learning fields can be set appropriately according to the game content and the setting of the game world.

[0080] A single character AI 8 is constructed by learning one or more fields. A character AI 8 constructed by learning multiple fields may be realized by associating a single learning field with multiple sub-AIs.

[0081] Each support character 4 has an "attribute" set, and a character AI 8 corresponding to the type of attribute is associated with it. The type of "attribute" can be set appropriately based on the character's category (such as magician, warrior, healer, etc.), occupation, level, sex, age, race, and the impression given by the design (such as physical type / intellectual type).

[0082] Character AI8 associated with an attribute is set with a learning field and learning amount that makes the user think, "That's only natural for a character with that attribute." For example, character AI8a associated with support character 4a with the "half-baked level" and "magician" attributes learns more learning fields that the user would think would be natural for a magician of that level to know, and does not learn learning fields that the user would think would be natural for a magician of that level not to know. For example, character AI8 associated with the "male" attribute is trained to output answers in a manner of speech unique to men, but is not trained to output answers in a manner of speech unique to women. Paradoxically, character AI8 is set based on the attributes of support character 4.

[0083] In addition, the answer accuracy of each character AI8 is diagnosed after learning, and the answer accuracy diagnosis results for each learning field are stored in association with each other. For example, the answer accuracy diagnosis may be performed by preparing multiple sets of diagnostic orders and model answers that take into account expected order contents, outputting answers for each diagnostic order, and calculating the similarity with the model answer.

[0084] FIG. 4 is a diagram for explaining orders, responses, and evaluations. When the user 2 inputs an order, the support character 4 accepts the order as a receptionist (reception character 6), and presents the user 2 with a response to the accepted order (received order).

[0085] Orders are accepted and responses are given in a conversational format. For example, when the user 2 inputs an operation to start an order, the order input field 13 is displayed as a speech bubble of the user character 3. The user 2 inputs the order as text in the order input field 13 using a software keyboard or the like. The contents of the order are displayed as text in the order input field 13. The order may be input by voice, which is converted into text using a voice recognition function.

[0086] The first response to a received order is called an "initial response." The initial response is generated based on the output obtained by inputting the received order information (text data) to the character AI8 of the support character 4, which is the receiving character 6. When the initial response is generated, the response presenting unit 14 is displayed as a speech bubble for the support character 4. Then, conversational text including the content of the initial response is displayed in the response presenting unit 14. For example, it is preferable that the text is text that the receiving character 6 speaks to the user 2 as if he or she had devised the initial response. The response may be presented not only as text, but also as audio output using a voice reading function.

[0087] Of course, the design of the order input field 13 and answer presentation section 14 is not limited to this. Text may be displayed in a subtitle format instead of a speech bubble format.

[0088] When the initial answer is presented, the evaluation input unit 15 is displayed, and the user 2 inputs an evaluation of the initial answer in the evaluation input unit 15. The evaluation is used to decide whether to be satisfied with the initial answer presented this time, or to request a re-answer. For example, the satisfaction level may be input as a score out of 5. Alternatively, a two-level evaluation of NG / OK may be used. If the input evaluation result does not satisfy a given pass condition, a re-answer is generated. If the first answer (initial answer) upon receiving an order is considered to be a primary answer, a re-answer is a secondary answer (Xth answer; X is a natural number greater than or equal to 2) that is a secondary answer after the secondary answer.

[0089] FIG. 5 is a diagram for explaining the transition of the effect display related to the re-answer (FIG. 5(1)→FIG. 5(2)→FIG. 5(3)). As shown in FIG. 5(1), if the evaluation result does not satisfy the passing condition, first, a search execution effect 20 is displayed. In the search execution effect 20, the support character 4 tells the user 2, "Since I cannot give a sufficient answer with the extent of my knowledge, I will consult with my companions (other support characters 4) from now on." Specifically, the support character 4 (4a), who is the reception character 6, is made to perform a predetermined action while a balloon section 22 is attached and displayed. Then, in the balloon section 22, a comment text (search execution comment) is displayed to the effect that the reception character 6 will consult with his companions and ask for a second answer. Of course, the search execution comment may be read out loud.

[0090] As shown in Fig. 5(2), after the search execution presentation 20 is displayed, a consultation presentation 24 is displayed. The consultation presentation 24 shows the support character 4 consulting with a companion. Specifically, the reception character 6 is made to perform an action as if consulting with a companion, while a secondary answering character display 25 of another support character 4 (secondary answering character 7; support character 4b in the example of Fig. 5) who has become the companion with whom the consultation is being made is displayed.

[0091] In the internal process, the secondary answer character 7 used to generate the second answer is selected, and the secondary answer character 7 is set as another support character 4b who is a friend to consult with. Then, the received order (received order) is input to the character AI8 (8b) corresponding to the secondary answer character 7 to obtain an output, and the answer (second answer) is determined based on the output.

[0092] Fig. 6 is a diagram for explaining in detail the selection of the character (secondary answer character 7) used to generate a second answer. Incidentally, Fig. 6 is based on the situation in Fig. 5, and therefore shows a case in which the support character 4a, who is the reception character 6, is excluded from the selection of the character used to generate a second answer ([Excluded] in Fig. 6).

[0093] The character (secondary answer character 7) used to generate the second answer is selected from among candidates that satisfy the following (1) to (3). (1) The support character 4 is associated with a character AI 8 (8a) different from the character AI 8 (8a) associated with the support character 4 (4a) who accepted the order. (2) The support character 4 is a character other than the reception character 6 among the support characters 4 owned by the user 2 who placed the order (the ordering user). (3) A support character 4 that is set to a user other than the ordering user who has had a relationship with the ordering user in the past. "Other users with a relationship" refers to, for example, other users with whom the ordering user has played cooperatively in the game, other users with whom the ordering user has chatted, other users who are registered as friends if the game has a friend registration function, other users with whom the ordering user has played against each other, etc.

[0094] A suitable selection point ΣP is calculated for each of the support characters 4 which are candidate characters, and the candidate character with the highest point is set as the secondary answer character 7.

[0095] The suitable-person selection point ΣP is calculated by a predetermined function with variables being the attribute point Pz, the corresponding field precision Ps, the CC relationship point Pcc, and the value of a given function f that is classified according to the magnitude of the UC relationship point Puc. The function for calculating the suitable-person selection point ΣP can be set appropriately, but may be, for example, a linear expression multiplied by weighting coefficients kz, ks, kc, and ku, respectively.

[0096] The attribute points Pz are points calculated based on the attributes of the candidate character. Specifically, the order text entered by the user 2 is subjected to natural language analysis to determine which of the multiple learning fields related to the character AI8 the order is for. This is called the "corresponding field." The attribute points Pz are then calculated using a predetermined function g that uses as a variable the learning field ratio (see FIG. 3) according to the attributes known in advance as the specifications of the character AI8 of the candidate character. The suitable selection points ΣP are set so that the higher the attribute points Pz, the higher the points.

[0097] The corresponding field accuracy Ps is the accuracy value of the determined corresponding field among the learning field-specific accuracy (see FIG. 3) known in advance as the specifications of the character AI8 of each candidate character.

[0098] The CC relationship point Pcc is a point that indicates the degree of the relationship between two support characters 4. The content of the relationship can be set appropriately depending on the game content, etc. For example, the relationship may be set to "comrades," and the relationship point may represent the level of the friendship between the two characters.

[0099] The CC relationship point Pcc is changed when a predetermined relationship change condition is satisfied. An example of a relationship change condition is when multiple users 2 form a party and play cooperatively. In this case, the CC relationship points of the support characters 4 of the users who formed the party are changed to increase. In other words, it is considered that they have become closer. Conversely, when the parties play a competitive match, the CC relationship points of the support characters 4 of the other parties may be changed to decrease.

[0100] The calculation of the suitable selection point ΣP is made with reference to the CC relationship point Pcc in which the accepting character 6 is one of the supporting characters 4 and the candidate character is the other supporting character 4.

[0101] In addition, when the support character 4 of another user other than the ordering user is the candidate character, the CC relationship point Pcc can be replaced with the UU relationship (User / User relationship) point Puu. The UU relationship point Puu is a value indicating the quality of the relationship between the first user and the second user, and the higher the point, the more friendly the relationship between the two users is.

[0102] The UU relationship point Puu is automatically changed when a certain UU relationship change condition is satisfied, for example. When the game system 1000 has a chat function between users or a friend registration function, the UU relationship change condition can be a frequency condition related to the number of chats. When the UU relationship change condition is satisfied, the UU relationship point Puu is changed to increase. When the game system 1000 has a friend registration function, the UU relationship change condition is considered to be satisfied when a friend is registered, or each time a predetermined time has passed while the friend is registered, and the UU relationship point Puu may be changed to increase. Conversely, when the friend registration is cancelled, the UU relationship point Puu may be changed to decrease. The UU relationship point Puu may also be changed to increase according to the number of times the party is formed and the user plays together.

[0103] To calculate the suitable selection point ΣP, the UU relationship point Puu is used, where the user 2 who placed the order (the ordering user) is the first user, and the other user of the candidate character is the second user.

[0104] Also, the UU relationship point Puu can be replaced with the UF relationship point Puf. The UF relationship point Puf indicates the quality of the relationship between the user 2 who placed the order (the ordering user) and the support character 4 of another user. The higher the point, the more friendly the relationship between the two is.

[0105] The UF relationship point Puu is set to have a positive correlation with the UU relationship point Puu between the ordering user 2 (ordering user) and other users other than the ordering user. The UF relationship point Puu between the ordering user 2 (ordering user) and the candidate characters of other users other than the ordering user is used to calculate the suitable selection point ΣP.

[0106] The UC relationship (User / Character relationship) point Puc is a point that indicates the quality of the relationship between the user 2 (user character 3) and the support character 4, which is the reception character 6. The higher the point, the better the relationship, and the lower the point, the worse the relationship.

[0107] The UC relationship points Puc are automatically changed when, for example, a predetermined UC relationship change condition is satisfied. For example, as a result of natural language analysis of the language used when the user 2 places an order, the UC relationship points Puc are changed to be increased if a predetermined friendly language is used, and changed to be decreased if an unfriendly language is used. In addition, if the game content includes an element of training the support character 4, the UC relationship points Puc may be changed to be increased or decreased depending on the frequency and content of the care of the support character 4.

[0108] The function f calculates a correction value of the attribute point Pz according to the UC relationship point Puc. For example, when the UC relationship point Puc reaches a predetermined reference value indicating a good relationship, regardless of the value of the attribute point Pz, the function f is set to f{Puz}=0. When the UC relationship point Puc does not reach the reference value (when the relationship is not good), the function f{Puz} takes a larger negative value as the attribute point Pz increases.

[0109] In other words, the function f acts to lower the suitable selection point ΣP (intentionally lower the evaluation) of a character with higher attribute points Pz only when the relationship between the user 2 who placed the order and the support character 4, who is the reception character 6, is not good.

[0110] Returning to FIG. 5, after the consultation scene 24 is displayed, a second answer (sub-answer) is presented to the user 2 together with a second answer presentation scene 26, as shown in FIG. 5(3). Specifically, while the support character 4 is being made to perform a predetermined action, a speech bubble-like re-answer presentation section 27 is displayed on the support character 4. Then, in the re-answer presentation section 27, a comment text including the reason for the selection of the person consulted and an explanation thereof (comment text as if the reception character 6 itself had selected the secondary answer character 7) and the text of the re-answer are displayed as a statement from the support character 4. Also, together with the re-answer presentation section 27, a secondary answer character display 28 showing another support character 4 (support character 4b; secondary answer character 7 in the example of FIG. 5) who has become a friend of the person consulted is displayed.

[0111] In this way, in the game system 1000, in relation to the control of the reception character 6's response to an order, if the user 2 is not satisfied with the initial response, the support character 4 consults with someone (another support character 4) and obtains a second response (secondary response), which can be expressed by a series of performances including a search execution performance 20, a consultation performance 24, and a second response presentation performance 26. The transition of these successive performance displays makes the user 2 see the support character 4 responding to the order as if it were a human. In entertainment services such as games, a technology can be realized that uses AI to provide a service that is full of interest.

[0112] The game system 1000 not only successively displays the search execution performance 20, the consultation performance 24, and the re-answer presentation performance 26, but also intentionally deteriorates the initial response and the re-answer (secondary response) to further anthropomorphize the response behavior of the support character 4 to be more "human-like."

[0113] FIG. 7 is a diagram for explaining the deterioration of the initial response. If the UC relationship point Puc at the time when the user 2 places an order is less than a predetermined deterioration triggering threshold, it is determined that the predetermined deterioration triggering condition is satisfied, and the deterioration response generation process 30 is performed.

[0114] The degraded answer generation process 30 is executed by combining an order processing process 32, an original answer generation process 34 by the character AI 8, and an original answer processing process 36.

[0115] The order modification process 32 is a process for intentionally distorting the contents of an order by deleting or changing parts of the original order information, for example deleting parts of the order text, changing the subject or object, deleting or modifying specified conditions, etc.

[0116] In the order processing process 32 for the first response (initial response; primary response), the degree of deterioration is determined by the UC relationship point Puc. Specifically, the lower the UC relationship point Puc, the higher the degree of deterioration is set by increasing the number of changes to the original order information (received order information). It can be said that the order processing process 32 for the initial response generates deteriorated order information by processing the original order information based on the UC relationship.

[0117] The original answer generation process 34 is a process for inputting received order information or degraded order information degraded in the order processing process 32 to the character AI 8 and obtaining an original answer output.

[0118] The original answer processing process 36 is a process for generating an initial answer that intentionally distorts the content of the answer by deleting or changing part of the information of the original answer generated in the original answer generation process 34. For example, parts of the text of the original answer are deleted, or the subject or object is changed. Also, a template phrase is inserted. It is preferable that the phrase be content that gives the user 2 the impression that the support character 4 "finds answering troublesome," "is answering reluctantly," or "is actually able to answer correctly but is intentionally obscuring or deceiving."

[0119] The original answer processing process 36 for the initial answer determines the deterioration degree based on the UC relationship point Puc. Specifically, the lower the UC relationship point Puc, the more parts of the original answer are changed to increase the deterioration degree. In other words, it can be said that the original answer processing process 36 for the initial answer generates an initial answer by processing the original answer based on the UC relationship.

[0120] Furthermore, if the UC relationship point Puc is below a predetermined minimum standard value (the relationship between the user 2 and the support character 4 is at the worst level), the original answer obtained by the original answer generation process 34 may be replaced with a predetermined refusal to answer (for example, "... (three dots, continuous middle dots)" which means ignoring or a silent reply).

[0121] The order processing process 32 and the raw response processing process 36 may be performed in either or both ways. For example, it may be determined randomly for each order whether to perform either or both, and if only one is performed, it may be determined again randomly which one to perform.

[0122] When the order processing process 32 or the original response processing process 36 is executed, the response to the original order (received order) will lack information and will be irrelevant and inappropriate. From the user 2's perspective, it will appear to be a mistaken response by the support character 4 who does not correctly understand the intention of the order, a shortcut, or an answer that seems to have been given despite a lack of knowledge.

[0123] If support character 4 were to simply answer questions using AI trained by machine learning as in the past, it would only give answers that it believes to be correct. Therefore, careless answers would not occur, and user 2 would not have any complicated feelings. However, this is precisely what makes support character 4 appear as if it were a person, and is an important element in making user 2 feel interested in it.

[0124] FIG. 8 is a diagram for explaining degradation of re-replies. If a) the UU relationship points Puu between the ordering user 2 and other users of the support character 4 selected as the character (secondary answer character 7) to be used to generate a re-answer, or b) the CC relationship points Pcc between the receiving character 6 and the secondary answer character 7 (support character 4b) are less than a specified deterioration activation threshold, it is determined that the specified deterioration activation condition is met, and the deteriorated answer generation process 30f is performed.

[0125] Furthermore, in the order processing process 32f for the re-reply (secondary reply), the degree of deterioration is determined by the UU relationship point Puu or the CC relationship point Pcc. Specifically, the lower the UU relationship point Puu or the CC relationship point Pcc, the more the parts of the received order information are changed to increase the degree of deterioration. In other words, the order processing process 32f for the re-reply (secondary reply) can be said to generate deteriorated order information by processing the received order information based on the UU relationship.

[0126] The original answer processing process 36f for the re-reply (secondary answer) determines the degree of deterioration based on the UU relationship point Puu or the CC relationship point Pcc. Specifically, the lower the UU relationship point Puu or the CC relationship point Pcc, the more parts of the original answer are changed to increase the degree of deterioration. In other words, it can be said that the original answer processing process 36f for the re-reply (secondary answer) generates a secondary answer by processing the original answer based on the UU relationship.

[0127] As a result, if the secondary answering character 7 is a support character 4 of another user who does not have a good relationship with the ordering user, the second answer (secondary answer) will be insufficient in information, off the mark, or irrelevant.

[0128] If the second reply given by the support character 4 to whom the user 2 has gone to consult a companion is insufficient, off-point, or misguided, the user 2 will think, "Oh, I consulted the wrong person," or "I wonder if they were being mean to me." If the support character 4 is simply made to reply using AI as in the past, the second reply will not be insufficient or off-point or misguided. However, in the game provided by the game system 1000 of this embodiment, this does occur, and the support character 4 can be made to express "humanness," which is one of the elements that makes the user 2 feel the interest.

[0129] The degradation activation for the re-answer may be performed based on the CC relationship point Pcc. That is, the degradation response generation process 30f may be executed when both the CC relationship point Pcc and / or the UU relationship point Puu are less than the degradation activation threshold.

[0130] As a further measure to express the "human-likeness" of the support character 4, the game system 1000 is constructed so that the support character 4 may cancel the generation of a second reply.

[0131] The game system 1000 also performs cancellation control so that a second answer (secondary answer) may not be generated even if a second answer is requested, thereby making the support character 4's answering behavior more "human-like."

[0132] FIG. 9 is a diagram for explaining the cancellation of the generation of a re-answer. If the evaluation of the answer by user 2 does not satisfy the pass condition, it is considered that a re-answer is requested. At this time, the re-answer cancellation decision process 40 is executed to decide whether or not to cancel the generation of the re-answer.

[0133] Specifically, for example, it is determined to "cancel" when a) the UC relationship points Puc between the ordering user 2 (ordering user) and the support character 4, and / or b) the UU relationship points Puu between the ordering user and another user of the support character 4 used for re-replying, satisfy a predetermined cancellation execution condition (for example, when they are below a predetermined reference value). With reference to the reply evaluation history data 622 (see FIG. 14), it may be determined to "cancel" when the frequency at which a re-reply was requested for a predetermined number of past orders, that is, the frequency at which the passing condition was not satisfied, is equal to or exceeds a predetermined reference value. When the ordering user repeatedly gives a low evaluation to the reply, an expression is realized as if the reception character 6 has lost motivation to re-reply.

[0134] If the user "does not cancel," a new answer is generated, and the new answer is presented to the ordering user after going through a search execution process, a consultation process, and a process of presenting a new answer (see FIG. 5). If the user selects "Cancel," the generation of the second reply is canceled, and the search execution presentation 20, the consultation presentation 24, and the second reply presentation presentation 26 are not displayed. Instead, the server system 1100 displays a given "throwing up" presentation 42 to the user 2, indicating that the support character 4 who received the order cannot comply with the request for a second reply. In this presentation, the reception character 6 is made to perform a given action, while the reception character 6 is made to display a comment display section 44 in the form of a speech bubble. Then, comment text to the effect that a second reply is not possible is displayed in the comment display section 44. Audio may be played to read out the comment.

[0135] Next, the functional configuration will be described. FIG. 10 is a block diagram showing an example of the functional configuration of the server system 1100. As shown in FIG. The server system 1100 includes an operation input unit 100s, a server processing unit 200s, a sound output unit 390s, an image display unit 392s, a communication unit 394s, and a server storage unit 500s.

[0136] The operation input unit 100s is a means for inputting various operations for managing the server system 1100. For example, a keyboard, a touch panel, a mouse, etc., correspond to this.

[0137] The server processing unit 200s is realized by electronic components such as processors serving as arithmetic circuits, such as a CPU, GPU, ASIC, and FPGA, as well as IC memories, and controls input and output of data between each functional unit including the operation input unit 100s and the server storage unit 500s.The server processing unit 200s performs various types of arithmetic processing based on predetermined programs and data, data received from operation input signals from the operation input unit 100s, and the like, and comprehensively controls the behavior of the server system 1100.

[0138] The server processing unit 200s has a user management unit 202, a game management unit 210, a timing unit 280s, a sound generation unit 290s, an image generation unit 292s, and a communication control unit 294s.

[0139] The user management unit 202 performs processing related to user registration procedures for the game system 1000 and stores and manages various types of information linked to user accounts.

[0140] The game management unit 210 performs various controls related to game execution. Specifically, the game management unit 210 sets up a game space by arranging background objects in a virtual three-dimensional space, arranges characters such as a user character 3, a support character 4, and various NPCs in the game space, controls the behavior of the user character 3 in response to an operation input by the user 2, and automatically controls the behavior of the support character 4 and the NPCs.

[0141] FIG. 11 is a diagram showing the functional units of the game management unit 210. As shown in FIG. The game management unit 210 has an answerer object setting unit 212, an order reception control unit 214, an order corresponding field determination unit 216, an initial answer generation control unit 218, and a secondary answer generation control unit 220. The game management unit 210 also has a first relationship determination unit 222, a second relationship determination unit 224, a third relationship determination unit 226, a fourth relationship determination unit 228, a pass condition setting change unit 230, a reception character selection unit 232, and a model change control unit 234. The game management unit 210 also has an initial answer presentation control unit 240, a selection comment output control unit 242, a search execution comment output control unit 244, a first evaluation acquisition control unit 250, a second evaluation acquisition control unit 252, and a pass / fail determination unit 254. The game management unit 210 may also include a party play control unit that organizes a party of multiple users to challenge events and missions, a chat control unit that allows users to chat with each other, and the like.

[0142] The answerer object setting unit 212 sets a plurality of answerer objects capable of answering a given order for an answer in a given virtual world in association with an automatic answer generation model that automatically generates an answer for the answerer object. The answerer object setting unit 212 associates a support character 4, which is associated with at least one character AI 8, with a user character 3 (see FIG. 2).

[0143] The order reception control unit 214 controls the reception of a reception order, which is an order to a reception object that is one of a plurality of answerer objects, from the user 2. Specifically, the order reception control unit 214 receives the reception order in a conversational format between the character of the reception object and the user (see FIG. 4).

[0144] The order corresponding field determination unit 216 determines which field of study the content of the received order corresponds to among the fields of study of the character AI 8. For example, the order corresponding field determination unit 216 is realized by a natural language analysis AI.

[0145] The initial answer generation control unit 218 generates an initial answer for a reception order by using an automatic answer generation model of the reception character 6 (see FIG. 4). The initial answer generation control unit 218 also generates an original answer for a reception order by using the automatic answer generation model of the reception object, and processes the original answer based on the relationship between the user 2 and the reception character 6 to generate an initial answer (see FIG. 7; original answer processing process 36).

[0146] When the initial answer does not satisfy the given passing conditions, the secondary answer generation control unit 220 selects a secondary answer object which is an answerer object other than the reception object, and generates a secondary answer for the reception order using the automatic answer generation model of the secondary answer object (see Figure 5 (2)).

[0147] Specifically, the secondary answer generation control unit 220 selects a secondary answer object from among the characters owned by the user 2 based on the attributes of the characters (see FIG. 8; attribute points Pz).

[0148] In addition, the secondary answer generation control unit 220 selects a character of a secondary answer object based on the relationship (UC relationship point Puc) between the user 2 and the receiving character 6. Specifically, in the calculation of the suitable selection point ΣP, the function f{Pz} by case classification of the UC relationship point Puc affects the selection of the support character 4 (secondary answer character 7) to be used for the re-answer (see FIG. 6).

[0149] In addition, the secondary answer generation control unit 220 generates an original answer to a received order using an automatic answer generation model of the secondary answer object (secondary answer character 7), and generates a secondary answer by processing the original answer based on 1) the relationship (UF relationship points Puf) between the user 2 who placed the order and the secondary answer character 7, or 2) the relationship (CC relationship points Pcc) between the reception character 6 and the secondary answer character 7 (see Figure 8; original answer processing process 36f).

[0150] Also, considering that CC relationship points Pcc can be substituted for UU relationship points Puu, the secondary answer generation control unit 220 can be said to generate an original answer to a reception order using an automatic answer generation model of another user's secondary answer object (original answer generation process 34f in Figure 8), and to generate a secondary answer by processing the original answer based on the relationship between the user of the reception object and the other user (original answer processing process 36f in Figure 8).

[0151] In addition, the secondary answer generation control unit 220 processes the received order based on 1) the relationship between the user 2 who placed the order and the secondary answer character 7 (UF relationship points Puf), or 2) the relationship between the reception character 6 and the secondary answer character 7 (CC relationship points Pcc), and generates a secondary answer for the processed received order using the automatic answer generation model of the secondary answer object (secondary answer character 7) (see Figure 8; order processing process 32f).

[0152] Also, taking into consideration that the CC relationship point Pcc can be replaced by the UU relationship point Puu, the secondary answer generation control unit 220 can be said to process the received order based on the relationship between the user of the reception object and the other user, and to generate a primitive answer to the received order after processing using an automatic answer generation model of the secondary answer object of the other user (for example, applying a degree of deterioration according to the UU relationship point Puu to the order processing process 32f in Figure 8).

[0153] Also, the secondary answer generation control unit 220 selects a user other than the user 2 of the reception object (reception character 6) based on the relationship between the user 2 of the reception object and the other user, and selects the answerer object of the other user as a secondary answer object (secondary answer character 7). Specifically, this corresponds to the case where the UU relationship point Puu is used instead of the CC relationship point Pcc in the calculation of the suitable-person selection point ΣP (see FIG. 6).

[0154] In addition, if the Xth secondary answer does not satisfy the pass condition, the secondary answer generation control unit 220 selects the (X+1)th secondary answer object, and repeatedly performs the process of generating the (X+1)th secondary answer using the automatic answer generation model of the (X+1)th secondary answer object.

[0155] The first relationship determination unit 222 determines the relationship between the user 2 and the reception character 6, which is a character of a reception object. This corresponds to setting and updating of the UC relationship point Puc (see FIG. 7).

[0156] The second relationship determination unit 224 determines 1) the relationship between the user 2 who placed the order and the secondary answer character 7 of the secondary answer object (see Figure 6; UF relationship points Puf), or 2) the relationship between the reception character 6 and the secondary answer character 7 (see Figure 6; CC relationship points Pcc).

[0157] In other words, focusing on the degree X of the secondary answer, the second relationship determination unit 224 determines 1) the relationship between the user 2 who placed the order and the Xth secondary answer character 7 related to the Xth secondary answer, or 2) the relationship between the reception character 6 and the secondary answer character 7 related to the Xth secondary answer.

[0158] The third relationship determination unit 226 determines 1) the relationship between the user 2 who placed the order and each of the characters (for example, UF relationship points Puf in Figure 6), or 2) the relationship between the reception character 6 and each of the characters other than the reception character (for example, CC relationship points Pcc in Figure 6).

[0159] The fourth relationship determination unit 228 determines the relationship between the users 2 (for example, the UU relationship point Puu in FIG. 8).

[0160] The pass condition setting change unit 230 sets the pass condition by changing the strictness of the pass condition for the answer based on the relationship between the user 2 and the reception character 6 (see FIG. 6; UC relationship point Puc).

[0161] In addition, the pass condition setting change unit 230 sets the pass conditions by changing the strictness of the pass conditions for the answer based on the automatic answer generation model of the secondary answer character 7 based on 1) the relationship between the user 2 who placed the order and the secondary answer character 7 of the secondary answer object (see Figure 6; UF relationship points Puf), or 2) the relationship between the reception character 6 and the secondary answer character 7 (see Figure 6; CC relationship points Pcc).

[0162] In addition, the pass condition setting change unit 230 sets the pass conditions by changing the strictness of the pass conditions for the answer based on the automatic answer generation model of the secondary answer character 7 based on the relationship between the user 2 (order user) of the reception object (reception character 6) and other users of the secondary answer object (secondary answer character 7) (see Figure 6; UU relationship points Puu).

[0163] As described above, when the Xth secondary answer does not satisfy the pass condition, the secondary answer generation control unit 220 selects the (X+1)th secondary answer object, and repeats the process of generating the (X+1)th secondary answer using the automatic answer generation model of the (X+1)th secondary answer object. Focusing on this repeat loop, it can be said that the pass condition setting change unit 230 changes the strictness of the pass condition based on the relationship between the user 2 (order user) of the reception object (reception character 6) and the other users of the Xth secondary answer object (secondary answer character 7), and sets the pass condition for the Xth secondary answer.

[0164] The reception character selection unit 232 selects a reception character 6 from among the characters (support characters 4) owned by the user 2 based on an operation by the user 2.

[0165] The model change control unit 234 changes the answer automatic generation model (character AI 8) of the reception object (reception character 6) based on the user information.

[0166] The initial answer presentation control unit 240 presents the initial answer to the user as if the receiving object itself had devised the initial answer. This corresponds to the control of displaying the initial answer as text in the answer presentation unit 14 (see FIG. 4).

[0167] The selection comment output control unit 242 performs control to output a comment to the user as if the receiving object itself had selected the sub-answer object.

[0168] When the initial answer does not satisfy the passing condition, the search execution comment output control unit 244 controls the reception object (reception character 6) to output a search execution comment to the user, requesting an answer from an answerer object other than the reception object. This corresponds to the display control of the comment text in the speech bubble section 22 in the search execution performance 20 (see FIG. 5).

[0169] The first evaluation acquisition control unit 250 performs control to acquire the user's evaluation on the initial answer, which corresponds to the control related to the evaluation input unit 15 (see FIG. 4).

[0170] The second evaluation acquisition control unit 252 performs control to present the received order and the initial response to a given survey target user and acquire an evaluation of the initial response.

[0171] The pass / fail determination unit 254 determines whether or not the pass condition is met based on the user's evaluation of the initial answer and the secondary answer.

[0172] The timekeeping unit 280s uses a system clock to measure various times such as the current date and time and time limits.

[0173] The sound generation unit 290s is realized by executing an IC or software that generates and decodes audio data. The sound generation unit 290s outputs the generated audio signal to the sound output unit 390s. The sound output unit 390s is realized by a speaker or the like, and emits sound based on the audio signal.

[0174] The image generation unit 292s generates images of various management screens for system management of the server system 1100, and outputs image data to the image display unit 392s. The image generation unit 292s also generates a part or all of the images to be displayed on the user terminal 1500. The image display unit 392s is realized by a device that displays images, such as a flat panel display, a head mounted display, or a projector.

[0175] The communication control unit 294s executes data processing related to data communication, and realizes data exchange with an external device via the communication unit 394s. The communication unit 394s realizes communication by connecting to the network 9. For example, it is realized by a wireless communication device, a modem, a TA (terminal adapter), a jack of a wired communication cable, a control circuit, etc. In the example of FIG. 1, the communication device 1153 corresponds to this.

[0176] The server storage unit 500s stores programs and various data for implementing various functions for the server processing unit 200s to comprehensively control the server system 1100. The server storage unit 500s is also used as a working area for the server processing unit 200s, and temporarily stores the results of calculations executed by the server processing unit 200s according to various programs. These functions are implemented by, for example, IC memories such as RAM and ROM, magnetic disks such as hard disks, optical disks such as CD-ROMs and DVDs, online storage, etc. In the example of FIG. 1, these correspond to storage media such as the IC memory 1152 and hard disks mounted on the server system 1100.

[0177] FIG. 12 is a diagram showing an example of programs and data stored in the server storage unit 500s. The server storage unit 500s stores a server program 501, a distribution client program 503, game initial setting data 510, character AI definition data 550, user management data 600, play data 700, and a current date and time 900. The server storage unit 500s also stores other programs and data (for example, timers, counters, various flags, etc.) as appropriate.

[0178] The server program 501 is a program for realizing the functions of the user management unit 202 and the game management unit 210 .

[0179] The delivery client program 503 is the original client program provided to the user terminal 1500 and executed therein.

[0180] The game initial setting data 510 stores various initial setting data relating to the game. The game initial setting data 510 includes, for example, user character candidate initial setting data 512, support character initial setting data 514, search execution performance definition data 516, consultation performance definition data 518, re-answer presentation performance definition data 520, and hand up performance definition data 522. Of course, data other than these may also be included as appropriate.

[0181] The user character candidate initial setting data 512 is prepared for each option from which the user 2 selects the user character 3 .

[0182] Support character initial setting data 514 is prepared for each type of support character 4, and stores various initial setting data related to the character. For example, one support character initial setting data 514 includes a character type, character model data, and attributes.

[0183] The search execution performance definition data 516, the consultation performance definition data 518, the re-answer presentation performance definition data 520, and the give up performance definition data 522 each include application requirements and performance execution data. The application requirements of these definition data are conditions that must be satisfied by the character that is the subject of each performance. For example, the application requirements include 1) an ownership condition indicating whether the character is the support character 4 of the ordering user or a support character 4 of another user other than the ordering user, 2) an attribute condition regarding the attributes of the support character 4, and 3) a type condition of the support character 4.

[0184] Character AI definition data 550 is prepared for each type of character AI 8. One character AI definition data 550 includes, for example, an applied attribute condition 551, AI model data 553, a ratio list by learning field 555, and an accuracy list by learning field 557, as shown in FIG.

[0185] The applicable attribute conditions 551 describe the conditions for the attributes of the support character 4 suitable for using the character AI 8. These are the contents that the attributes of the support character 4 must satisfy in order to associate the character AI 8. One or more types of attributes are specified. In a configuration in which multiple categories of attributes are used (for example, occupation attributes and race attributes), a combination of attributes by category is specified.

[0186] Returning to FIG. 12, the user management data 600 is prepared for each user who has completed the registration procedure, and stores the registration information of the user and various data associated with the user. 14, one piece of user management data 600 includes a user account 601, user information 602, and save data 603. Of course, data other than these may also be included as appropriate.

[0187] The user information 602 includes personal information of the user, such as age, sex, place of residence, and information that estimates the user's hobbies and interests (for example, online shopping history, internet search history, etc.).

[0188] The save data 603 includes user character setting data 605, support character setting data 610 for each support character 4 set for the user 2, and a reception character ID 618 indicating the reception character 6. The save data 603 also includes play performance data 620 corresponding to second user information, answer evaluation history data 622, UC relationship data 630, CC relationship data 632, and UU relationship data 634.

[0189] The support character setting data 610 includes a character ID 611, a character type 612, a setting date and time 613, a type of character AI used 614, a setting destination user account 615, and a setting destination user character ID 616. Of course, other data than these may also be included as appropriate.

[0190] The setting destination user account 615 is used when setting the support character 4 for the user 2. The setting destination user character ID 616 is used when the support character 4 is set as the user character 3. For example, it is used in a game in which one user 2 plays a plurality of user characters 3 in a switchable manner.

[0191] The answer evaluation history data 622 stores the evaluation results by the user 2 on answers to orders. The answer evaluation history data 622 is created every time an evaluation on an answer is acquired, and stores the evaluation date and time, the evaluator, and the evaluation result in association with each other. The evaluator is either the ordering user / automatic (when a re-answer is canceled; see FIG. 9) / another user (when a survey is conducted for other users; details will be described later).

[0192] The UC relationship data 630 is prepared for each combination of a user in the user management data 600 and a support character 4 set for that user, and stores the UC relationship points Puc.

[0193] The CC relationship data 632 is prepared for each combination of one support character 4 and the other support character 4, and stores the CC relationship points Pcc. The "one support character 4" here is the support character 4 set for the user in the user management data 600. The "other support character 4" is the support character 4 set for another user with whom the user in the user management data 600 has a relationship.

[0194] The "other users with whom the user has a relationship" refers to, for example, other users with whom the user has cooperated in a game, other users with whom the user has chatted, other users who have registered as friends if a friend registration function is available, other users with whom the user has played against, etc. When a user of the user management data 600 has a relationship with another user, the CC relationship data 632 is created by automatically referring to the support character setting data 610 of the other user.

[0195] The UU relationship data 634 is prepared for each combination of a first user and a second user, and stores the UU relationship point Puu between the two users. The first user is a user of the user management data 600, and the second user is another user who has a relationship with the first user. The UU relationship data 634 is automatically created when a user of the user management data 600 has a relationship with another user.

[0196] FIG. 15 is a diagram showing an example of the data structure of the play data 700. As shown in FIG. The play data 700 is created for each game play, and includes various data related to the game progress control of that game play.

[0197] One piece of play data 700 includes, for example, a player account 701 indicating the user 2 who is the player, user character control data 703 for each user character 3, and support character control data 705 for each support character 4. The play data 700 also includes received order data 710, answer data 720 for each answer and re-answer, and degraded answer generation process data 730. Of course, data other than these may also be included as appropriate.

[0198] Received order data 710 is prepared for each order entered by user 2. One received order data 710 includes an order ID 711, a reception character ID 713 which is the character ID of the support character 4 acting as a receptionist, order text data 714, an order handling field 715, and pass condition data 716. Of course, other data than these may also be included as appropriate.

[0199] The order ID 711 is uniquely set and includes, for example, the date and time of receipt and an ordering user account that indicates the user who placed the order.

[0200] The pass condition data 716 indicates the standard for determining whether a response to the received order is acceptable. In this embodiment, an evaluation score out of 5 is input in the evaluation input unit 15 (see FIG. 4), so that the pass condition data 716 is set to a pass standard score.

[0201] The pass condition indicated by the pass condition data 716 is changed by comparing the UC relationship point Puc with a predetermined strictness criterion. For example, if the UC relationship point Puc meets the strictness criterion indicating a good relationship, the pass standard score is lowered according to the goodness, and if it meets the strictness criterion indicating a bad relationship, the pass standard score is raised according to the badness.

[0202] Answer data 720 is created for each answer. One answer data 720 includes a corresponding order ID 721, an answer order 722, an answerer support character ID 723, an answerer character AI type 724 indicating the character AI 8 used to generate the answer, and answer data 725. Of course, other data may also be included as appropriate. If the answer order 722 is order "1", this indicates that the answer is the first answer, i.e., an initial answer. If the order is "2" or higher, this indicates that the answer is a secondary answer.

[0203] The degradation response generation process data 730 stores various data related to generation of a degradation response. One degradation response generation processing data 730 includes a corresponding order ID 731, processed order data 732, and processed original response data 734. When generating a degradation response, the response data 725 is the data of the original response.

[0204] FIG. 16 is a functional block diagram showing an example of the functional configuration of the user terminal 1500. The user terminal 1500 includes an operation input unit 100 , a device processing unit 200 , a sound output unit 390 , an image display unit 392 , a communication unit 394 , and a terminal storage unit 500 .

[0205] The operation input unit 100 outputs an operation input signal corresponding to various operation inputs made by the player to the device processing unit 200. For example, this can be realized by a push switch, a touch panel, a joystick, a touch pad, a trackball, an acceleration sensor, a gyro, or the like.

[0206] The device processing unit 200 is realized by electronic components such as a microprocessor such as a CPU or GPU, or an IC memory, and controls data input / output between each functional unit including the operation input unit 100 and the terminal storage unit 500. Then, it executes various types of arithmetic processing based on predetermined programs and data, operation input signals from the operation input unit 100, and various types of data received from the server system 1100, and controls the operation of the user terminal 1500.

[0207] The device processing unit 200 has a client control unit 260 , a timing unit 280 , a sound generation unit 290 , an image generation unit 292 , and a communication control unit 294 .

[0208] The client control unit 260 performs various controls to make the user terminal 1500 function as an MMIF (man-machine interface) as a control unit serving as a client in the game system 1000. Specifically, the client control unit 260 has an operation input information providing unit 261 and a display control unit 262.

[0209] The operation input information providing unit 261 performs control to transmit operation input information to the server system 1100 in response to an input from the operation input unit 100 .

[0210] The display control unit 262 performs control for displaying various images based on the data received from the server system 1100 .

[0211] The timekeeping unit 280 uses a system clock to keep track of the current date and time, time limit, and the like.

[0212] The sound generation unit 290 is realized by, for example, a digital signal processor (DSP), a processor such as a voice synthesis IC, an audio codec capable of reproducing voice files, or the like, and generates sound signals of music, sound effects, and various operation sounds, and outputs them to the sound output unit 390. The sound output unit 390 is realized by a device that outputs (emits sound) based on the sound signal input from the sound generation unit 290, such as a speaker.

[0213] The image generation unit 292 generates and outputs an image signal for displaying an image on the image display unit 392 based on the control of the client control unit 260. In the example of Fig. 1, this corresponds to a GPU (Graphics Processing Unit), a graphic controller, a graphic board, etc. mounted on the control board 1550. The image display unit 392 is realized by a device for displaying an image, such as a flat panel display, a head mounted display, or a projector.

[0214] The communication control unit 294 executes data processing related to data communication and realizes data exchange with an external device via the communication unit 394. The communication unit 394 realizes communication by connecting to the network 9. For example, this is realized by a wireless communication device, a modem, a TA (terminal adapter), a jack for a wired communication cable, a control circuit, etc. In the example of FIG. 1, this corresponds to the communication module 1553.

[0215] The terminal storage unit 500 stores programs for causing the terminal processing unit 200 to realize given functions, various data, and the like. It is also used as a working area for the terminal processing unit 200, and temporarily stores the results of calculations executed by the terminal processing unit 200 according to various programs, input data input from the operation input unit 100, and the like. These functions are realized by, for example, IC memories such as RAM and ROM, magnetic disks such as hard disks, optical disks such as CD-ROMs and DVDs, and the like. In the example of FIG. 1, this corresponds to the IC memory 1552 mounted on the control board 1550.

[0216] Specifically, the terminal storage unit 500 stores a client program 800 (application program) for causing the user terminal 1500 to function as a client control unit 260 for the virtual space service, client data 810, and a current date and time 900. Of course, data other than these can also be stored as appropriate.

[0217] The client data 810 is various data for executing control by the client control unit 260. The client data 810 includes, for example, a user account, a password, and the like.

[0218] 17 to 19 are flow charts for explaining the flow of processing executed by the server system 1100. For convenience, the flow will be explained from the start to the end of play for one user, but since this is an MMORPG, in actual operation, processing for multiple users will be executed simultaneously in parallel.

[0219] As shown in FIG. 17, if the user 2 is a new user (YES in step S10), the server system 1100 first performs a process of setting a user character 3 (step S12).

[0220] Next, the server system 1100 initializes the support character 4 associated with the user (step S14). Specifically, the server system 1100 presents characters selectable as the support character 4 to the user 2 and accepts the selection of one of the preferred characters. Then, the server system 1100 creates new support character setting data 610 for the selected character and sets it as the support character 4 owned by the user 2. At this time, the server system 1100 sets the character AI8 corresponding to the attribute of the support character 4 as the used character AI 614 by referring to the applied attribute condition 551 of the character AI definition data 550 (see FIG. 13).

[0221] On the other hand, if User 2 is already a registered user, save data 603 is read (step S16).

[0222] When the game progress control is started (step S20), the server system 1100 starts update control of the UC relationship point Puc, the CC relationship point Pcc, and the UU relationship point Puu (step S22). That is, the points are increased or decreased when a predetermined update condition for each relationship point is satisfied according to the game progress status, playing results, the occurrence of interactions between users, and the like.

[0223] When a predetermined condition for addition is satisfied during game play (YES in step S24), the server system 1100 adds a new support character 4 to the user 2 (step S26). For example, the user 2 can add a new support character 4 through a bonus based on playing results, a lottery, or the like. It may also be possible to add a new support character 4 by purchasing it.

[0224] If the support character 4 owned by the user 2 is single as initially set (single in step S30), the initially set support character 4 is set as the reception character 6 (step S32).

[0225] When the user 2 owns multiple support characters 4 (multiple in step S30), the server system 1100 executes a received character selection process (step S34). Specifically, the server system 1100 presents the owned support characters 4 as options, accepts the user's designation operation, and sets the designated character as the received character 6. It may be possible to automatically select a character from among the added characters with priority.

[0226] When a predetermined order start operation is detected (YES in step S40), the server system 1100 displays the order input field 13 (see FIG. 4) and accepts the order in a conversational format (step S42). Then, when the order acceptance is completed, the server system 1100 determines the corresponding field of the accepted order (accepted order) (step S44).

[0227] Here, the server system 1100 judges whether the deterioration triggering condition is satisfied (step S50; see FIG. 7).

[0228] If the deterioration triggering condition is not met (NO in step S50), the server system 1100 generates an initial answer by inputting the order text data 714 indicated by the latest received order data 710 directly to the character AI8 of the reception character 6, and generates answer data 720 (step S52; see FIG. 15). If the deterioration triggering condition is met (YES in step S50), the server system 1100 executes a deteriorated answer generating process (step S60).

[0229] FIG. 20 is a flowchart for explaining the flow of the degradation answer generating process. In the degraded answer generation process, the server system 1100 first determines a method for generating a degraded answer (step S62). Specifically, one of three methods is determined: either the order processing process 32 or the original answer processing process 36, or both (see FIG. 4). For example, the method may be determined by random drawing. If the UC relationship point Puc is lower than the degradation trigger condition but higher than a predetermined maximum degradation reference value, either the order processing process 32 or the original answer processing process 36 selected by drawing may be executed, and if it is equal to or lower than the maximum degradation reference value, both the order processing process 32 and the original answer processing process 36 may be executed.

[0230] If the order processing process 32 is not adopted (NO in step S64), the server system 1100 executes the original answer generation process 34 (see FIG. 4) with the original received order indicated by the received order data 710 (step S66). That is, the received order is input to the character AI8 of the receiving character 6, and an answer is output, which is used as the original answer.

[0231] If the order processing process 32 is adopted (YES in step S64), the server system 1100 executes the order processing process 32 (see FIG. 4) (step S68), and executes the primitive answer generation process 34 with the processed order (step S70).

[0232] When the original response processing 36 is not adopted (NO in step S80), the server system 1100 uses the original response as the response as is (step S82).

[0233] When the original answer processing process 36 is adopted (YES in step S80), the server system 1100 executes the original answer processing process 36 (step S84), and uses the processed original answer as an answer (step S86).

[0234] Turning now to FIG. 18, the server system 1100 displays the initial answer as text on the answer presenting unit 14 (see FIG. 4) to present the initial answer to the ordering user (step S100).

[0235] Next, the server system 1100 determines whether a given survey implementation condition is met (step S102). The survey implementation condition can be set appropriately. For example, the survey implementation condition may be set when a predetermined number of consecutive unsuccessful evaluations are made in the answer evaluation history data 622 (see FIG. 14), when the corresponding field of the received order is outside the learning field of the character AI8, when the character AI8 is selected by random drawing, etc.

[0236] If the survey implementation conditions are not met (NO in step S102), the server system 1100 displays the evaluation input unit 15 as a first evaluation acquisition process, prompting the ordering user to input an evaluation and acquiring an evaluation (step S104).

[0237] Next, the server system 1100 changes the strictness of the pass condition for the evaluation of the initial answer based on the UC relationship point Puc (see Figure 6) between the user 2 who placed the order (ordering user) and the reception character 6, and sets the pass condition for the evaluation of the initial answer (step S106).

[0238] If the conditions for conducting the survey are met (YES in step S102), the server system 1100 executes a second evaluation acquisition process (step S108).

[0239] FIG. 21 is a flowchart illustrating the flow of the second evaluation acquisition process. In the second evaluation acquisition process, the server system 1100 selects a predetermined number of survey target users from among other users who are playing (step S120), and displays the contents of the received order and the contents of the responses on the user terminals 1500 of the survey target users (step S122).

[0240] Then, the evaluation input unit 15 (see FIG. 4) is displayed on the user terminal 1500 of the survey target user to receive and collect the results of the evaluations (external evaluations) of the answers (step S124). The server system 1100 determines the evaluation of the answers based on the collected external evaluations (step S126).

[0241] Returning to FIG. 18, the server system 1100 judges whether the initial response is a pass or fail (step S110), and if the initial response is not a pass (NO in step S140), executes the re-response cancellation judgment process 40 (see FIG. 9) (step S142).

[0242] When the determination result of the re-answer cancellation determination process 40 is "cancel" (YES in step S144), the server system 1100 controls the display of the "hands up" effect 42 (step S146).

[0243] If the determination result of the second reply cancellation determination process 40 is "not to cancel" (NO in step S144), the server system 1100 controls the display of the search execution presentation 20 (see FIG. 5) (step S148).

[0244] Next, when the ordering user 2 (ordering user) owns only one support character 4 (single in step S150), the server system 1100 selects a support character 4 of another user other than the ordering user as a character (secondary answer character 7) to be used to generate a re-answer (step S152). That is, the support characters 4 of other users who have had a relationship with the ordering user are set as candidate characters, the suitable selection points ΣP are calculated for each candidate character, and a character to be the secondary answer character 7 is selected based on the suitable selection points ΣP (see FIG. 6).

[0245] When the ordering user has multiple support characters 4 (multiple in step S150), the server system 1100 selects a character (secondary answer character 7) to be used for generating a re-answer from among the ordering user's support characters 4 other than the reception character 6 (step S154). That is, the ordering user's support characters 4 other than the reception character 6 are set as candidate characters, the suitable selection point ΣP is calculated for each candidate character, and a character to be the secondary answer character 7 is selected based on the suitable selection point ΣP (see FIG. 6).

[0246] Here, the server system 1100 judges whether the deterioration triggering condition is satisfied for the secondary answer character 7 (step S160; see FIG. 8).

[0247] If the deterioration trigger condition is not met (NO in step S160), the server system 1100 inputs the order text data 714 indicated by the latest received order data 710 directly into the character AI8 of the secondary answer character 7 to generate a re-answer (secondary answer), and generates answer data 720 (step S162; see Figure 15).

[0248] If the deterioration triggering condition is met (YES in step S160), the server system 1100 executes a deterioration response generating process (step S164; see FIG. 9).

[0249] 19, the server system 1100 next controls the display of the consultation performance 24 (step S166). At this time, definition data that satisfies the application requirements among the consultation performance definition data 518 (see FIG. 12) is referenced. For example, if the secondary reply character 7 is a support character 4 of another user other than the ordering user, the consultation performance 24 as exemplified in FIG. 5 is displayed. If the secondary reply character 7 is a support character 4 owned by the ordering user, for example, as shown in FIG. 22(3), a consultation performance 24b is displayed in which the reception character 6 approaches the secondary reply character 7 to consult with the secondary reply character 7.

[0250] Next, the server system 1100 controls the display of the second answer presentation performance 26 to present the second answer (secondary answer) to the ordering user (step S168). At this time, definition data that satisfies the application requirements is referenced from the second answer presentation performance definition data 520 (see FIG. 12). For example, if the secondary answer character 7 is a support character 4 of another user other than the ordering user, the second answer presentation performance 26 as exemplified in FIG. 5 is displayed. If the secondary answer character 7 is a support character 4 owned by the ordering user, for example, as shown in FIG. 22(4), a second answer presentation performance 26b is displayed in which the secondary answer character 7 itself responds to the ordering user with a second answer, as shown in FIG. In the second answer presentation performance 26b, the second answer presentation part 27 is displayed together with the secondary answer character 7.

[0251] Next, the server system 1100 determines whether or not a given survey implementation condition is met (step S180).

[0252] If the survey implementation conditions are not met (NO in step S180), the server system 1100 displays the evaluation input unit 15 to prompt the ordering user to input an evaluation for the re-answer and acquires an evaluation as a first evaluation acquisition process (step S182). Next, the server system 1100 changes the strictness of the pass condition for the evaluation of the initial answer based on the UC relationship point Puc (see FIG. 6) between the ordering user 2 (ordering user) and the reception character 6, and sets the pass condition for the evaluation of the initial answer (step S106).

[0253] If the survey implementation conditions are met (YES in step S180), the server system 1100 sets the pass conditions by changing the strictness of the pass conditions based on the CC relationship points Pcc between the reception character 6 and the secondary answer character 7 (step S186). Note that the CC relationship points Pcc can be replaced with the UU relationship points Puu or the UF relationship points Puf.

[0254] Next, the server system 1100 judges whether the second answer (secondary answer) is acceptable or not (step S188).

[0255] If the re-answer (secondary answer) is not acceptable (NO in step S190), the server system 1100 increases the answer level by "1" (step S192), and if the answer level does not reach the predetermined upper limit, it is determined that the condition for ending the repetition has not been met (NO in step S194), and the process returns to step S142.

[0256] If the re-answer is successful (YES in step S190), the server system 1100 judges whether at least one of the user information 602 and the play performance data 620 (second user information) satisfies a given model change condition (step S200). For example, it is judged that the model change condition is satisfied when the performance indicated by the play performance data 620 satisfies an excellent condition, when a new skill is imparted to the support character 4 as a play result in the play performance data 620, when the attribute of the support character 4 is changed, etc.

[0257] If the model change condition is met (YES in step S200), the server system 1100 adds, changes, or deletes the character AI8 associated with the support character 4 to change it (step S202). For example, if a new skill or new attribute is acquired, the character AI8 corresponding to it is added to the used character AI type 614. If the attribute is changed, the used character AI type 614 is changed to the character AI8 corresponding to the new attribute.

[0258] The server system 1100 repeats steps S24 to S212 until the game end condition is satisfied (NO in step S210). When the game end condition is satisfied (YES in step S210), the save data 603 (see FIG. 14) is saved (step S212), and the series of processes is terminated.

[0259] As described above, according to this embodiment, a technology that utilizes AI in the entertainment field and can improve interest can be provided.

[0260] For example, if the initial answer is not acceptable, a second answer (secondary answer) is generated and presented to the user. If this second answer is also not acceptable, yet another second answer (secondary answer) is generated and presented. The process of generating a second answer and presenting it to the user involves a search execution performance 20, a consultation performance 24, and a second answer presentation performance 26. These successive performances make it appear to the user as if the reception character 6 was "nasty" with the initial answer, so is consulting with a colleague, obtaining a second answer, and reporting it. From the user's perspective, the reception character 6 has an interest that makes it seem very "human-like," making it suitable as an AI technology in the entertainment field.

[0261] Second Embodiment Next, a second embodiment to which the present invention is applied will be described. The second embodiment is basically realized in the same way as the first embodiment, but differs in that the game management unit 210 is realized by a user terminal 1500. In the description of the second embodiment, the differences from the first embodiment will be mainly described, and the same components as the first embodiment will be given the same reference numerals as in the first embodiment, and duplicated descriptions will be omitted.

[0262] FIG. 23 is a diagram illustrating an example of the functional configuration of a user terminal 1500B in the second embodiment. The user terminal 1500 B includes a game management unit 210 . Accordingly, the user terminal 1500B stores in the terminal memory unit 500 a game program 802 for realizing the function of the game management unit 210, game initial setting data 510, character AI definition data 550, user management data 600, and play data 700.

[0263] The process flow in this embodiment is basically the same as that in the first embodiment, but the user terminal 1500B is the entity that executes the process related to the game management unit 210. Therefore, according to the second embodiment, the same effects as those in the first embodiment can be obtained.

[0264] [Modifications] 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-described forms, and components can be added, omitted, or modified as appropriate.

[0265] (Variation 1) For example, the first embodiment shows an example of implementing an online game in a client-server configuration, and the second embodiment shows an example of implementing a game in a stand-alone configuration, but the present invention is not limited to this. For example, a game may be implemented in a system in which multiple user terminals 1500 are connected in a P2P (peer-to-peer) manner. In this case, any one of the user terminals 1500 that constitute the P2P functions as the server system 1100 of the first embodiment.

[0266] (Variation 2) As an intermediate configuration between the first and second embodiments, a configuration in which the functions of the game management unit 210 are shared between the server system 1100 and the user terminal 1500 is also possible. For example, based on the second embodiment, the functions of the initial answer generation control unit 218 and the secondary answer generation control unit 220 are left in the server system 1100. When an initial answer / secondary answer is required in the user terminal 1500, a configuration may be adopted in which a request for generating an initial answer / secondary answer is sent to the server system 1100 together with the reception order each time.

[0267] (Variation 3) In the above embodiment, a machine-learned AI is exemplified as an automatic answer generation model for the character AI 8, but a rule-based AI may also be used depending on the expected content of the order.

[0268] (Variation 4) In the above embodiment, the technology of online games is exemplified as an example of a technology utilizing AI in the entertainment field, but the present embodiment is not limited thereto. For example, the present embodiment may be applied to response control of a support character that guides a user in a three-dimensional virtual space such as the metaverse.

[0269] (Variation 5) Although the support character 4 is given as an example of an answerer object, the answerer object is not limited to a character-like design. For example, when the present embodiment is applied to a support function in a conversational format in a three-dimensional virtual space such as the metaverse, the answerer object may be displayed as a symbol indicating that the support function is operating, or as a speech bubble displaying text of the support function speaking to the user. That is, the answerer object may be designed as a fictional object that handles only input and output such as text and voice, without having a specific display form such as a character. For example, the answerer object may be an object like a window of a chatbot. [Explanation of symbols]

[0270] 2. User 3...User character 4...Support characters 5…Respondent object 6…Receptionist character 7...Secondary answer character 8…Character AI 13...Order entry field 14…Answer presentation section 15...Evaluation input section 20...Search execution performance 24... Consultation and direction 26…Re-answer presentation performance 27…Re-answer presentation section 30…Deterioration response generation process 32…Order processing 34…Primitive answer generation process 36…Original response processing 200s…Server processing section 210…Game Management Department 212…Respondent object setting section 214...Order reception control unit 216…Order Field Judgment Department 218... Initial response generation control unit 220...Sub-answer generation control unit 222...First relationship determination unit 224...Second relationship determination unit 226…Third relationship determination unit 228…Fourth Relationship Determination Section 230... Passing condition setting change section 232…Reception character selection section 234...Model change control unit 240… Initial answer presentation control unit 242...Selected comment output control unit 244...Search execution comment output control unit 250...First evaluation acquisition control unit 252... second evaluation acquisition control unit 500s…Server storage 501...Server program 510... Game initial setting data 514...Supported character initial setting data 516...Search execution performance definition data 518... Consultation performance definition data 520...Reply presentation performance definition data 550…Character AI definition data 551…Applicable attribute conditions 553…AI model data 600...User management data 601...User account 602...User information 610...Supported character setting data 614…Character AI type used 618…Receptionist character ID 630…UC related data 632…CC Relationship Data 634...UU relationship data 700...Play data 710... Received order data 714...Order text data 715…Order-compatible fields 716... Passing condition data 720…Response data 722…Answer order 725…Response data 730…Deterioration response generation processing data 1000…Game System 1100...Server system 1500: User terminal

Claims

1. an answerer object setting means for setting, in a given virtual world, a plurality of answerer objects capable of answering a given order for which an answer is required, in association with an answer automatic generation model for automatically generating an answer for the answerer object; a reception means for receiving, from a user, a reception order, which is the order to a reception object that is one of the plurality of answerer objects; an initial response generation control means for generating an initial response to the received order by using the automatic response generation model of the reception object; a secondary answer generation control means for selecting a secondary answer object, which is an answerer object other than the reception object, when the initial answer does not satisfy a given acceptance condition, and generating a secondary answer for the reception order using the automatic answer generation model of the secondary answer object; A computer system comprising:

2. the respondent object is a character appearing in the virtual world, the reception means receives the reception order in a conversational format between the character of the reception object and the user; 2. The computer system of claim 1.

3. Attributes are set to the characters, the answer automatic generation model of the respondent object is set based on attributes of an answerer character which is the character of the respondent object; 3. The computer system of claim 2.

4. the secondary answer generation control means selects the secondary answer object based on an attribute of the character of the secondary answer object.

4. The computer system of claim 3.

5. a first relationship determination means for determining a relationship between the user and a reception character that is the character of the reception object; Further comprising: The initial answer generation control means generating a primitive reply to the received order using the reply auto-generation model of the receiving object; generating the initial answer by processing the original answer based on the relationship; To carry out 3. The computer system of claim 2.

6. a first relationship determination means for determining a relationship between the user and a reception character that is the character of the reception object; Further comprising: The initial answer generation control means processing the received order based on the relationship; and generating the initial response to the processed reception order using the automatic response generation model of the reception object; To carry out 3. The computer system of claim 2.

7. a first relationship determination means for determining a relationship between the user and a reception character that is the character of the reception object; A pass condition setting change means for changing the strictness of the pass condition based on the relationship to set the pass condition; The computer system of claim 2 further comprising:

8. a first relationship determination means for determining a relationship between the user and a reception character that is the character of the reception object; Further comprising: the secondary answer generation control means selects the character of the secondary answer object based on the relationship.

3. The computer system of claim 2.

9. a second relationship determination means for determining 1) a relationship between the user and a secondary answer character that is the character of the secondary answer object, or 2) a relationship between a reception character that is the character of the reception object and the secondary answer character; Further comprising: The secondary answer generation control means generating a primitive response to the received order using the response automatic generation model of the secondary response object; generating the secondary answer by processing the original answer based on the relationship determined by the second relationship determination means; To carry out 3. The computer system of claim 2.

10. a second relationship determination means for determining 1) a relationship between the user and a secondary answer character that is the character of the secondary answer object, or 2) a relationship between a reception character that is the character of the reception object and the secondary answer character; Further comprising: The secondary answer generation control means processing the received order based on the relationship determined by the second relationship determination means; generating the secondary response to the processed received order using the automatic response generation model of the secondary response object; To carry out 3. The computer system of claim 2.

11. the secondary answer generation control means, when the secondary answer for the Xth time does not satisfy the pass condition, selects the secondary answer object for the (X+1)th time, and repeatedly performs a process of generating the secondary answer for the (X+1)th time using the answer automatic generation model of the secondary answer object for the (X+1)th time; a second relationship determination means for determining 1) a relationship between the user and an X-th secondary answer character which is the character of the secondary answer object for the Xth time, or 2) a relationship between a reception character which is the character of the reception object and the X-th secondary answer character; a pass condition setting change means for changing a strictness of the pass condition based on the relationship determined by the second relationship determination means to set the pass condition; The computer system of claim 2 further comprising:

12. a third relationship determination means for determining 1) a relationship between the user and each of the characters, or 2) a relationship between a reception character that is the character of the reception object and each of the characters other than the reception character; Further comprising: the secondary answer generation control means selects a secondary answer character, which is the character to be set as the secondary answer object, based on the relationship determined by the third relationship determination means.

3. The computer system of claim 2.

13. The user owns a plurality of the characters, a reception character selection means for selecting a reception character, which is the character to be the reception object, from among the characters owned by the user based on an operation by the user; Further comprising: the secondary answer generation control means selects a secondary answer character, which is the character to be set as the secondary answer object, from among the characters owned by the user.

3. The computer system of claim 2.

14. the reception object is associated with the user, a model change control means for changing the automatic answer generation model of the reception object based on information of the user; The computer system of claim 1 further comprising:

15. The virtual world is accessible to multiple users; The respondent object is associated with each of the users, A fourth relationship determination means for determining a relationship between the users; Further comprising: the secondary answer generation control means selects a user other than the user of the reception object based on the relationship between the user of the reception object and the other user, and selects the answerer object of the other user as the secondary answer object.

2. The computer system of claim 1.

16. the secondary answer generation control means, when the secondary answer for the Xth time does not satisfy the pass condition, selects the secondary answer object for the (X+1)th time, and repeatedly performs a process of generating the secondary answer for the (X+1)th time using the answer automatic generation model of the secondary answer object for the (X+1)th time; a pass condition setting change means for changing the strictness of the pass condition to set the pass condition based on the relationship between a user of the reception object and a user of the secondary answer object for the Xth time; 20. The computer system of claim 15, further comprising:

17. The secondary answer generation control means generating a primitive answer to the received order by using the answer automatic generation model of the secondary answer object of the other user; generating the secondary answer by processing the original answer based on the relationship between the user of the reception object and the other user; To carry out 16. The computer system of claim 15.

18. The secondary answer generation control means processing the received order based on the relationship between the user of the reception object and the other user; generating an original answer to the received order from the processed secondary answer to the received order by using the answer automatic generation model of the secondary answer object of the other user; To carry out 16. The computer system of claim 15.

19. an initial answer presentation control means for presenting the initial answer to the user as if the receiving object itself had devised the initial answer; 19. The computer system of claim 1, further comprising:

20. a selection comment output control means for outputting a comment to the user as if the reception object itself had selected the secondary answer object; 19. The computer system of claim 1, further comprising:

21. a search execution comment output control means for, when the initial answer does not satisfy the passing condition, outputting to the user a search execution comment by the receiving object itself requesting an answer from an answerer object other than the receiving object; 21. The computer system of claim 20, further comprising:

22. an initial answer presenting means for presenting the initial answer to the user; a first evaluation acquisition means for acquiring an evaluation of the user for the initial answer; a pass / fail determination means for determining whether or not the pass condition is satisfied based on the evaluation; 19. The computer system of claim 1, further comprising:

23. a second evaluation acquisition means for presenting the received order and the initial response to a given survey target user and acquiring an evaluation of the initial response; a pass / fail determination means for determining whether or not the pass condition is satisfied based on the evaluation; 19. The computer system of claim 1, further comprising:

24. A computer system an answerer object setting means for setting, in a given virtual world, a plurality of answerer objects capable of answering a given order for which an answer is required, in association with an automatic answer generation model for automatically generating an answer for the answerer object; a reception means for receiving, from a user, a reception order, which is the order to a reception object that is one of the plurality of answerer objects; an initial response generation control means for generating an initial response to the received order by using the automatic response generation model of the reception object; a secondary answer generation control means for selecting a secondary answer object which is an answerer object other than the reception object when the initial answer does not satisfy a given acceptance condition, and generating a secondary answer for the reception order using the automatic answer generation model of the secondary answer object; A program to function as a

Images