Program, and System

Abstract

To reduce the burden of a setting operation related to in-game elements such as items used in a game.SOLUTION: A program causes a computer to function as: generation means for generating output data relating to in-game elements when log data regarding player operations in a game is input, to a learned model learned using learning data including story data and log data; and progression means for making the in-game elements appear in the game on the basis of the output data and enabling the game to progress using the in-game elements.SELECTED DRAWING: Figure 11

Description

Technical Field

[0001] The present invention relates to a program and a system.

Background Art

[0002] Conventionally, in a game using a computer (hereinafter simply referred to as a "game"), a technique for granting items used in the game according to a user's play is known.

[0003] For example, in a game, that is, in a casino in a virtual space, when a player wins a casino game, the player obtains virtual currency. Then, when the amount of virtual currency exceeds a predetermined amount, the player is notified. And the virtual currency can be exchanged for items that can be used in the game at an exchange in the game. Thus, a technique for motivating a player to acquire virtual currency is known (for example, Patent Document 1, etc.).

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, in the conventional technique, in a situation where game elements such as items appear, are exchanged, used, granted, bought and sold, improved, or acquired, etc., it is necessary to stipulate in advance settings corresponding to each of these situations at the game development stage. That is, for one game element, it is necessary to set one by one the behaviors corresponding to each situation, and the workload of setting the game elements in advance increases as the number of situations increases.

[0006] The present invention aims to reduce the workload of setting operations for in-game elements such as items handled in a game.

Means for Solving the Problem

[0007] To solve the above problems according to the present invention, when log data regarding a player's operation in the game is input to a learned model obtained by training a computer using learning data including story data and log data, the program functions as generation means for generating output data regarding in-game elements, and progress means for causing the in-game elements to appear in the game and enabling the game to progress using the in-game elements based on the output data.

Effects of the Invention

[0008] According to the present invention, the workload of setting operations for in-game elements such as items handled in a game can be reduced.

Brief Description of the Drawings

[0009]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Mode for Carrying Out the Invention

[0010] Hereinafter, embodiments will be described with reference to the drawings.

[0011] [System Configuration Example] FIG. 1 is a diagram showing an example of a system configuration according to the present embodiment. For example, as shown in FIG. 1, the system 1 mainly includes user terminals 20A, 20B, 20C (hereinafter, these may be collectively referred to as "user terminal 20".) and a server 11.

[0012] Hereinafter, the person who manages the server 11 is referred to as "administrator 5". Also, each person who operates the user terminals 20A, 20B, and 20C is referred to as "user 4A", "user 4B", "user 4C" (hereinafter, these may be collectively referred to as "user 4".).

[0013] The administrator 5 is a person who plays a role in operating the information processing service by the system 1. On the other hand, the user 4 is a person who uses the information processing service by the system 1. Also, the administrator 5 and the user 4 differ in which information processing device of the server 11, which is an example of a management device, or the user terminal 20 they operate. Hereinafter, the user 4 becomes a game player, and the administrator 5 manages the game (for example, generates, changes, and distributes the game, etc.) and the server 11.

[0014] Note that the example shown in FIG. 1 is an example where there are 3 user terminals 20 and 1 server 11, but the number of servers 11, the number of user terminals 20, the number of administrators 5, and the number of users 4 are not limited.

[0015] The server 11 and the user terminal 20 are communicably connected via the communication network 2. For example, the communication network 2 is a public line such as the Internet, a mobile communication system (e.g., a public line based on 4G (4th Generation) or 5G (5th Generation) mobile communication standards), a wireless network such as Wi-Fi (registered trademark), or a combination thereof.

[0016] The user terminal 20 downloads a program for playing a game (hereinafter referred to as the "game program") from the server 11, or when accessing the server 11, the server 11 provides a game service. Note that communication with the server 11 may not be necessary for playing the game. That is, the user terminal 20 may build an environment for playing the game by downloading a program or installing it from a medium. Also, the game may be played on an information processing device other than the user terminal 20.

[0017] [Examples of AI (Artificial Intelligence) learning and execution] Hereinafter, AI learns by "preprocessing" (also referred to as "learning processing"). The AI in the learning stage, that is, in the "preprocessing", is referred to as the "learning model A1". Hereinafter, there are two learning models A1, namely, the "first learning model A11" and the "second learning model A12". Also, the "first learning model A11" and the "second learning model A12" may be collectively referred to as the "learning model A1".

[0018] And when the learning of the learning model A1 progresses to a certain extent, it becomes the "learned model A2". Hereinafter, the execution stage of executing output processing using the learned model A2 is referred to as the "execution processing". Also, when the "first learning model A11" learns, it becomes the "first learned model A21". Similarly, when the "second learning model A12" learns, it becomes the "second learned model A22". Hereinafter, the "first learned model A21" and the "second learned model A22" may be collectively referred to as the "learned model A2".

[0019] The "preprocessing" is performed before the "execution processing". However, the "preprocessing" may continue during the "execution processing", that is, the learned model A2 may perform learning.

[0020] [Preprocessing example] FIG. 2 is a diagram showing a preprocessing example. For example, the preprocessing is performed by the server 11.

[0021] The first learning model A11 inputs the first learning data DL1 and performs learning. Similarly, the second learning model A12 inputs the second learning data DL2 and performs learning. That is, the learning model A1 performs so-called "supervised" learning.

[0022] The first learning data DL1 includes log data DL12, story data DL11, and first correct answer data DL13.

[0023] And the first learning data DL1 is data in which the first correct answer data DL13 that is the "correct answer" is associated with the log data DL12 and the story data DL11. In this way, the log data DL12 and the story data DL11 are data with known correct answers.

[0024] Details of the first learning data DL1, the story data DL11, the log data DL12, and the first correct answer data DL13 will be described later.

[0025] Through the learning process as described above, when the first learning data DL1 is input, the first learning model A11 learns the correlation with the "correct answer" indicated by the first correct answer data DL13 for the combination of the log data DL12 and the story data DL11.

[0026] The second learning data DL2 includes the log data DL12, similar to the first learning data DL1. Therefore, the log data DL12 is used in both the first learning data DL1 and the second learning data DL2 by duplicating the data or reusing it.

[0027] In addition, the second learning data DL2 includes a keyword DL21 and second correct answer data DL22.

[0028] Details of the second learning data DL2, the keyword DL21, and the second correct answer data DL22 will be described later.

[0029] The second learning data DL2 is data in which the log data DL12 and the keyword DL21 are associated with the second correct answer data DL22 that is the "correct answer". In this way, the log data DL12 and the keyword DL21 are data with known correct answers.

[0030] The second correct answer data DL22 is data showing the same content as the story data DL11. Therefore, the second correct answer data DL22 is used in both the first learning data DL1 and the second learning data DL2 by duplication or reuse. However, the second correct answer data DL22 is correct answer data indicating "correct answer" in the second learning data DL2, while in the first learning data DL1, the story data DL11 is input data. In this way, even for data with the same content, the second correct answer data DL22 and the story data DL11 have different roles in the learning process.

[0031] Furthermore, it is desirable for the learning model A1 to be learned with the big data D4. For example, the big data D4 is data on the Internet or the like. However, the big data D4 may also be data input by an administrator 5 or the like. In this way, when learning with the big data D4, an AI that generates natural sentences and data like those written by humans can be generated.

[0032] [Execution Processing Example] FIG. 3 is a diagram showing an execution processing example. For example, the execution processing is performed by a plurality of information processing devices such as the user terminal 20, the server 11, or the user terminal 20 and the server 11 cooperating with each other.

[0033] The learned model A2 is in the state where the learning model A1 has learned through preprocessing. That is, when the preprocessing shown in FIG. 2 is executed, the learned model A2 is generated.

[0034] When the first unknown data DU1 is input to the first learned model A21, the output data D3 is generated.

[0035] The first unknown data DU1 is a combination of data of log data (hereinafter referred to as "unknown log data DU11") and story data (hereinafter referred to as "unknown story data DU12") whose "correct answer" is unknown.

[0036] Details of the first unknown data DU1, the unknown log data DU11, the unknown story data DU12, and the output data D3 will be described later.

[0037] When the second unknown data DU2 is input to the second learned model A22, the unknown story data DU12 is generated.

[0038] The second unknown data DU2 is a combination of data of unknown log data DU11 and a keyword (hereinafter referred to as "unknown keyword DU21") whose "correct answer" is unknown. Note that the unknown log data DU11 is the same data as the log data DL12 in the learning process, and is replicated or reused for the first unknown data DU1 and the second unknown data DU2.

[0039] The unknown story data DU12 generated by the second learned model A22 is input to the first learned model A21 as the first unknown data DU1.

[0040] The above configuration is a configuration in which the story generator 12 performs preprocessing for generating the unknown story data DU12. In the above configuration, the AI is divided into two AIs for preprocessing and postprocessing, but the AI may be configured as one integrated unit or divided into three or more units.

[0041] When the output data D3 is generated, the server 11 distributes a game to the user terminal 20 based on the output data D3. Note that the game may be transmitted to the user terminal 20 with the output data D3. As long as the content of the output data D3 is reflected in the game, the configuration for distributing the game and the data configuration for holding the output data D3 and the like are not limited.

[0042] FIG. 4 is a diagram showing an overall processing example of AI learning and execution. The relationship between the preprocessing shown in FIG. 2 and the execution processing shown in FIG. 3 is as shown in FIG. 4.

[0043] Note that the preprocessing and the execution processing do not have to be executed in a continuous order as illustrated in the figure. Therefore, it is not essential to make continuous the period for preparation by the preprocessing and the subsequent period for execution processing. Thus, the execution processing may be performed after a lapse of time from the preprocessing as long as the learned model A2 has been once created. Also, as long as the learned model A2 has been once generated, the execution processing may be performed by diverting the learned model A2.

[0044] In the learning process and the execution process, the first learning data DL1 and the first unknown data DU1 are different. Further, in the learning process and the execution process, the second learning data DL2 and the second unknown data DU2 are different. Also, the AI is the learning model A1 at the learning stage, but becomes the learned model A2 as learning progresses to a certain extent. In this way, the learned model A2 that has learned big data D4 and the like as learning data is a so-called "generative AI".

[0045] For each learning data, the "correct answer" is known, while for each unknown data, the "correct answer" is unknown. Specifically, since the first learning data DL1 includes the first correct answer data DL13, the first correct answer data DL13 is associated with the combination of the log data DL12 and the story data DL11, whereas the first unknown data DU1 does not include the first correct answer data DL13. Therefore, in the learning process, the relationship between the combination of the log data DL12 and the story data DL11 and the first correct answer data DL13 is known. Thus, when the first learning model A11 inputs the first learning data DL1, it can learn the correlation between the combination of the log data DL12 and the story data DL11 and the "correct answer".

[0046] Similarly, since the second learning data DL2 includes the second correct answer data DL22, the second correct answer data DL22 is associated with the combination of the log data DL12 and the keyword DL21, whereas the second unknown data DU2 does not include the second correct answer data DL22. Therefore, in the learning process, the relationship between the combination of the log data DL12 and the keyword DL21 and the second correct answer data DL22 is known. Thus, when the second learning model A12 inputs the second learning data DL2, it can learn the correlation between the combination of the log data DL12 and the keyword DL21 and the "correct answer".

[0047] On the other hand, the first unknown data DU1 does not include the first correct answer data DL13, and the "correct answer" for the first unknown data DU1 is unknown. Then, based on the correlation between the combination of the log data DL12 and the story data DL11 learned in the preprocessing and the first correct answer data DL13, the first learned model A21 generates the output data D3 for the first unknown data DU1.

[0048] Similarly, the second unknown data DU2 does not include the second correct answer data DL22, and the "correct answer" for the second unknown data DU2 is unknown. Then, based on the correlation between the combination of the log data DL12 and the keyword DL21 learned in the preprocessing and the second correct answer data DL22, the second learned model A22 generates the unknown story data DU12 for the second unknown data DU2.

[0049] The execution process may be a process that uses a table or the like for some executions. In this way, in a configuration that uses a table, namely a so-called rule-based system, the preprocessing is a process of preparing to input a table (also referred to as a look-up table (LUT) or the like), or a mathematical formula or the like.

[0050] [Hardware configuration example of an information processing apparatus] FIG. 5 is a hardware configuration diagram of an information processing apparatus. The information processing apparatus includes a server 11, a user terminal 20, and the like. Hereinafter, it is assumed that the information processing apparatus has the same hardware configuration as the server 11. For example, the information processing apparatus is a general-purpose computer such as a workstation or a personal computer. However, the hardware configurations of the respective information processing apparatuses may be different.

[0051] The server 11 mainly includes a processor 111, a memory 112, a storage 113, an input / output interface 114, and a communication interface 115. Further, each component of the server 11 is connected to a communication bus 116.

[0052] The processor 111 realizes processing and control by executing a series of instructions included in the server program 11P stored in the memory 112 or the storage 113.

[0053] The processor 111 is, for example, an arithmetic unit and a control unit such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an MPU (Micro Processing Unit), an FPGA (Field-Programmable Gate Array), an ASIC (Application Specific Integrated Circuit), or a combination thereof.

[0054] Memory 112 is a main memory device that stores the server program 11P and data, etc. For example, the server program 11P is loaded from the storage 113. Also, the data includes the data input to the server 11 and the data generated by the processor 111. For example, the memory 112 is a RAM (Random Access Memory) or other volatile memory.

[0055] Storage 113 is an auxiliary storage device that stores the server program 11P and data, etc. The storage 113 is, for example, a ROM (Read-Only Memory), a hard disk device, a flash memory, or other non-volatile storage device. Also, the storage 113 may be a detachable storage device such as a memory card. Further, as another example, the storage 113 may be an external storage device. According to such a configuration, for example, in a scenario where a plurality of user terminals 20 such as an amusement facility are used, it becomes possible to collectively update the server program 11P or data.

[0056] The input / output interface 114 is an interface that connects external devices such as a monitor, an input device (for example, a keyboard or a pointing device, etc.), an external storage device, a speaker, a camera, a microphone, and a sensor to the server 11.

[0057] Also, the processor 111 communicates with the external device through the input / output interface 114. The input / output interface 114 is, for example, a USB (Universal Serial Bus), a DVI (Digital Visual Interface), an HDMI (registered trademark) (High-Definition Multimedia Interface), wireless, and other terminals.

[0058] The communication interface 115 communicates with other devices (such as the user terminal 20, etc.) connected to the communication network 2. For example, the communication interface 115 is a wired communication interface such as a LAN (Local Area Network), or a wireless communication interface such as Wi-Fi (registered trademark) (Wireless Fidelity), Bluetooth (registered trademark), or NFC (Near Field Communication).

[0059] However, the information processing device is not limited to the above hardware configuration. For example, the user terminal 20 may further include a sensor such as a camera. Then, various data acquired by the user terminal 20 using the sensor may be transmitted to the server 11.

[0060] [Configuration examples of the learning model and the learned model] FIG. 6 is a network diagram showing a configuration example of an AI. The learning model A1 and the learned model A2 are AIs having a configuration shown by the following network, for example.

[0061] Hereinafter, the learning model A1 and the learned model A2 will be described by taking an example where they are implemented on the server 11, that is, on the cloud. However, part or all of the learning model A1 and the learned model A2 may be implemented on the user terminal 20 or the like.

[0062] The network 300 has a configuration including, for example, an input layer L1, an intermediate layer L2 (also referred to as a "hidden layer", etc.), and an output layer L3, etc.

[0063] The input layer L1 is a layer for inputting data.

[0064] The intermediate layer L2 converts the data input by the input layer L1 based on weights, biases, etc. The result processed by the intermediate layer L2 in this way is transmitted to the output layer L3.

[0065] The output layer L3 is a layer for outputting output contents, etc.

[0066] Then, through learning, the weight coefficients, parameters to be changed during learning, etc. are optimized. Note that the network 300 is not limited to the network structure shown in the figure. That is, the AI may be realized by other machine learning methods.

[0067] For example, the AI may be configured to perform preprocessing such as dimensionality reduction (for example, a process of changing a relationship of three or more dimensions into a relationship that can be obtained by a simple calculation of about three or less dimensions) by machine learning without a teacher. It is desirable that the relationship between the input and the output be processed by a simple calculation such as a linear equation. Such calculations can reduce the computational cost.

[0068] Also, the AI may be subjected to a process of reducing overfitting (also referred to as "overfitting" or "hyperfitting") such as dropout. In addition, preprocessing such as dimensionality reduction and normalization may be performed.

[0069] The AI may have a network structure such as a CNN (Convolution Neural Network). In addition, for example, the network structure may have a configuration such as an LLM (Large Language Model), an RNN (Recurrent Neural Network), or an LSTM (Long Short-Term Memory). That is, the AI may have a network structure other than deep learning.

[0070] Also, the AI may be configured to have hyperparameters. That is, the AI may be configured such that some settings are made by the user or the like. Furthermore, the AI may specify the feature amounts to be learned, or the user may set some or all of the feature amounts to be learned.

[0071] Furthermore, the learning model A1 and the learned model A2 may utilize other machine learning methods. For example, the learning model A1 and the learned model A2 may perform preprocessing such as normalization using an unsupervised model. Additionally, the learning may be reinforcement learning (Reinforcement Learning, which refers to a learning method where AI is made to make selections and an evaluation (reward) is given for the selections to increase the evaluation).

[0072] In learning, data augmentation and the like may be performed. That is, in order to increase the learning data used for the learning of the learning model A1, preprocessing may be performed to expand one piece of experimental data or the like into multiple pieces of learning data. In this way, if the learning data can be increased, the learning of the learning model A1 can be further advanced.

[0073] Also, the learning model A1 and the learned model A2 may be configured to perform transfer learning or fine-tuning. That is, since the user terminal 20 often has different execution environments for each device, the settings may be different for each device according to the execution environment. For example, the basic configuration of the AI is learned on another information processing device. Thereafter, each information processing device may be additionally learned or set, etc. in order to optimize for each execution environment.

[0074] [Example of Learning Process in RPG (Role-Playing Game)] Hereinafter, an example targeting RPG will be described. In RPG, when the game progresses, events and the like occur and various elements appear. Hereinafter, the elements that appear in the game are referred to as "in-game elements".

[0075] In-game elements specifically include items, characters, skills (techniques), magic, and so on that appear in the game's story. In-game elements specifically include various elements that make up the game, such as items, characters, skills (techniques), magic, and so on that appear in the game's story. In-game elements may be objects that are operated on during play, such as items or skills, or they may be objects that are not operated on by user 4.

[0076] For example, items are one of the in-game elements, including event items necessary for events to progress the game, attack items that can attack enemies, recovery items that restore a character's health and abnormal states, and auxiliary items that improve various parameters of a character, such as attack power, defense, and speed (so-called buff effects), or reduce various parameters of enemies (so-called debuff effects).

[0077] Note that items are not limited to the above. In addition to equipment such as weapons, armor, and accessories, they may also be various material items used for composing decks, item synthesis, or synthesis of monsters, etc. Also, items may include items that are not necessarily essential to obtain for the progress of the game and cannot be used in the game (for example, medals, badges, etc. that honor the achievement of predetermined conditions in the game, such as defeating a predetermined enemy (meaning defeating a certain number of times, etc.)).

[0078] Characters are humans, animals, robots, etc. in the game, and are elements that create a trigger for generating events in progressing the game, form a party as the protagonist's companions, or constitute the story. Note that characters include playable characters that can be operated by the player and non-playable characters (so-called NPCs) that cannot be operated by the player. Also, they are not necessarily limited to real entities and may be objects set according to the game's worldview, such as demi-humans, beastmen, androids, etc.

[0079] Skills (techniques) or magic (which has various names depending on the game) are elements that are activated based on the operations of user 4 and include commands that can advantageously advance battles and the like within the game. Note that skills and magic are not necessarily limited to those that are advantageous. For example, those in which some parameters are enhanced while others are weakened, such as fire resistance +50 and ice resistance -50, are also included. Also included are things like so-called "passive skills" that are activated without operations.

[0080] Hereinafter, an example in which elements within the game are referred to as "items" will be described. And in the following example, it is assumed that the output data D3 indicates what kind of item it is (for example, the specifications such as what effects will occur when used), and if there is the output data D3, items can be introduced into the game played on the user terminal 20.

[0081] [Examples of First Learning Data, Story Data, Log Data, and First Correct Answer Data] FIG. 7 is a diagram showing an example of the learning process of the first learning model in an RPG. For example, to train the first learning model A11, the following data is used for training.

[0082] The log data DL12 is generated in accordance with the progress of the game when the game is played, for example. Note that the log data DL12 may also be the settings and plans for advancing the game before the game starts. On the other hand, when the game is advanced halfway, the log data DL12 is a record from the time when the game started to the time when it was advanced.

[0083] The log data DL12 has a data configuration including, for example, image data DL121 and profile data DL122. However, each data does not necessarily need to be separated into files, and a data configuration in which a single data contains a plurality of contents is also possible. Also, the log data DL12 indicates various settings in the game.

[0084] The image data DL121 is data indicating, for example, an output screen (including a screen not disclosed to user 4), the appearance of a character, or the appearance of various items. Note that the image data DL121 may be subjected to image analysis to extract parameters.

[0085] The profile data DL122 indicates a character appearing in the game or the profile of user 4. Which character or the profile of user 4 the profile data DL122 indicates varies depending on game settings, input items, etc.

[0086] For example, the profile data DL122 is input by user 4 at the start of the game. However, the profile data DL122 may be a pre-setting instead of a setting by user 4, or may be acquired from the user terminal 20 or other databases, etc.

[0087] Specifically, the profile data DL122 is setting items such as name, origin, gender, age, occupation, and friendship relationship. What profile items to use are pre-set.

[0088] The name is, for example, the real name, abbreviation, or alias of user 4. Also, the name may be identification information such as a nickname set for a character. Similarly, the origin, gender, age, occupation, and friendship relationship may be settings by user 4 or fictional settings in the game.

[0089] The story data DL11 is, for example, text data indicating a story in text. However, the story data DL11 may include images, etc.

[0090] The first correct answer data DL13 is data indicating items to be introduced in the game. Specifically, the first correct answer data DL13 is specifications such as the appearance, name, and effects when used of the items. The first correct answer data DL13, that is, the specifications for introducing items, vary depending on the game. Therefore, the first correct answer data DL13 is data including items required to introduce items in the game.

[0091] When performing the above learning, the first learning model A11 learns the correlation of what items to generate from the story, profile, etc. As described above, in generating items, in addition to the story, when inputting the profile, image, etc. as auxiliary data, the information of the auxiliary data is considered in the generation of items.

[0092] [Examples of Second Learning Data, Keywords, Log Data, and Second Correct Answer Data] FIG. 8 is a diagram showing an example of the learning process of the second learning model in the RPG. For example, to train the second learning model A12, the following data is used for learning.

[0093] The log data DL12 is, for example, the same as the learning process of the first learning model A11, that is, shown in FIG. 7. However, the log data DL12 may have different data contents or different data amounts from the learning process of the first learning model A11.

[0094] The keyword DL21 is a word or sentence referred to for generating the story data DL11. For example, the keyword DL21 is the meaning of in-game elements, the characteristics of in-game elements, events occurring to characters, elements to be introduced in the story, or the gist of the story, etc.

[0095] The story data DL11 is, for example, text data showing the story in sentences. That is, the story data DL11 is the same as the learning process shown in FIG. 7.

[0096] For example, when keyword DL21 is input, the second learning model A12 learns so that a story containing keyword DL21 is generated. Note that keyword DL21 does not necessarily have to be directly included in story data DL11, and keyword DL21 may be indirectly expressed in the story.

[0097] Also, a story is generated based on log data DL12 as well. For example, when parameters in profiles such as "gender" or "place of origin" in profile data DL122 are different, story data DL11 is generated such that the effects, specifications, types, names, or ways of appearance of items change according to the parameters.

[0098] When performing the above learning, the second learning model A12 learns the correlation of what kind of story to generate from keyword DL21 and profiles, etc. As described above, when generating a story, in addition to profile data DL122, peripheral information, images, etc. are input as auxiliary data, and the information of the auxiliary data is considered in the generation of the story.

[0099] Note that the generation of the story may be realized by a configuration other than AI. For example, when fixing the type of item to be introduced according to "gender" or "age", etc., a configuration (for example, input as auxiliary data) in which the conditions to be fixed (for example, the type of item, etc.) are input from profile data DL122 to the first pre-trained model A21 without going through AI may be used. In this way, when the fixed conditions can be input, the item to be introduced can appear under specific conditions corresponding to this in a certain profile.

[0100] Specifically, when "occupation" is a specific occupation type, it will always be an item of "weapon", while for another occupation type, conditions such as it will always be an item of "armor" can be set.

[0101] [Example of Execution Processing in RPG] FIG. 9 is a diagram showing an example of execution processing in an RPG. After the learned model A2 is generated by the learning processing as shown in FIGS. 7 and 8, the execution processing is executed as follows.

[0102] The second unknown data DU2, that is, the unknown log data DU11 and the unknown keyword DU21, have unknown correct answers at the time when the data is input. In the execution processing, based on the correlation between the learned combination of the keyword DL21 and the log data DL12 and the second correct answer data DL22, the second learned model A22 generates the unknown story data DU12 based on the combination of the unknown log data DU11 and the unknown keyword DU21. After the unknown story data DU12 is generated by the second learned model A22, it is input to the first learned model A21.

[0103] Note that after the unknown story data DU12 is generated and before it is input to the first learned model A21, an administrator 5 or the like may edit the story.

[0104] Next, the unknown log data DU11 and the unknown story data DU12 are input to the first learned model A21 as the first unknown data DU1. The combination of the first unknown data DU1, that is, the unknown log data DU11 and the unknown story data DU12, has an unknown correct answer at the time when the data is input.

[0105] In the execution processing, based on the correlation between the learned combination of the story data DL11 and the log data DL12 and the first correct answer data DL13, the first learned model A21 generates the output data D3 based on the combination of the unknown log data DU11 and the unknown story data DU12.

[0106] When the output data D3 is generated, thereafter, a game in which an item generated based on the output data D3 appears can be played. For example, the game can be played on a user terminal 20 or the like.

[0107] [Overall Processing Example] FIG. 10 is a diagram showing an overall processing example. In the following example, the overall processing performs preprocessing and execution processing continuously. Specifically, steps S11 to S14 are learning processes. Also, steps S21 to S24 are execution processes. However, the overall processing may include other processes in addition to these.

[0108] Hereinafter, it is assumed that steps S11 and S12, and steps S13 and S14, that is, the learning process of the first learning model A11 and the learning process of the second learning model A12 are performed in parallel. However, these learning processes may be performed sequentially instead of in parallel. For example, the second learning model A12 may be learned first to obtain the second learned model A22, and then the learning process of the first learning model A11 may be executed using the second learned model A22.

[0109] In step S11, the server 11 inputs the first learning data DL1.

[0110] In step S12, the server 11 inputs the first learning data DL1 and learns the first learning model A11 to generate the first learned model A21.

[0111] In step S13, the server 11 inputs the second learning data DL2.

[0112] In step S14, the server 11 inputs the second learning data DL2 and learns the second learning model A12 to generate the second learned model A22.

[0113] When steps S11 to S14 are performed, the learning processes shown in FIGS. 7 and 8 are obtained. As described above, when the learning process is performed to a certain extent, the learning model A1 is learned to become the learned model A2. Next, the execution process is performed as follows using the learned model A2 generated by the learning process.

[0114] In step S21, the server 11 inputs the second unknown data DU2 into the second pre-trained model A22.

[0115] In step S22, the server 11 uses the second pre-trained model A22 to generate unknown story data DU12.

[0116] In step S23, the server 11 uses the first pre-trained model A21 to generate output data D3. Subsequently, the server 11 provides the game to the user terminal 20 based on the output data D3.

[0117] In step S24, the server 11 advances the game so that game elements appear based on the output data D3.

[0118] [Functional Configuration Example] FIG. 11 is a diagram showing a functional configuration example. For example, the system 1 is a generation AI system including a learning device 61 and an execution device 62.

[0119] The learning device 61 includes a story data generation means 1F11, a first learning data input means 1F12, and a learning means 1F13.

[0120] When the keyword DL21 and the log data DL12 are input, the story data generation means 1F11 performs a story data generation procedure for generating the story data DL11. For example, the story data generation means 1F11 is realized by a story generator 12 or the like.

[0121] The first learning data input means 1F12 performs a first learning data input procedure for inputting the first learning data DL1 including the story data DL11, the log data DL12, and the first correct answer data DL13. For example, the first learning data input means 1F12 is realized by a communication interface 115 or the like.

[0122] The learning means 1F13 performs a learning procedure of training the first learning model A11 using the first training data DL1 to generate the first trained model A21. For example, the learning means 1F13 is implemented by a processor 111 or the like.

[0123] The execution device 62 includes an unknown story data generation means 1F21, a first unknown data input means 1F22, an execution means 1F23, and a progression means 1F24. The unknown story data generation means 1F21 is an example of the first generation means. Also, the execution means 1F23 is an example of the second generation means and the output data generation means.

[0124] When the unknown keyword DU21 and the unknown log data DU11 are input, the unknown story data generation means 1F21 performs an unknown story data generation procedure for generating the unknown story data DU12. For example, the unknown story data generation means 1F21 is implemented by a story generator 12 or the like.

[0125] The first unknown data input means 1F22 performs a first unknown data input procedure for inputting the first unknown data DU1 including the unknown story data DU12 and the unknown log data DU11. For example, the first unknown data input means 1F22 is implemented by a communication interface 115 or the like.

[0126] The execution means 1F23 performs an execution procedure for generating the output data D3 using the first trained model A21. For example, the execution means 1F23 is implemented by a processor 111 or the like.

[0127] Based on the output data D3, the progression means 1F24 performs a progression procedure for advancing the game on the user terminal 20. For example, the progression means 1F24 is implemented by a processor 111 or the like.

[0128] The learning device 61 and the execution device 62 are, for example, the server 11. Thus, the learning device 61 and the execution device 62 may be implemented by a single information processing device. However, the learning device 61 and the execution device 62 may also be different information processing devices.

[0129] With the above configuration, when keyword data that forms the basis of the story and log data indicating surrounding settings and the like for generating the story are input, a story tailored to characters and the like is generated. Then, in-game elements such as items corresponding to the story and surrounding settings are generated. Specifically, a story is generated, and the items that appear in the story can be made to actually appear in the game. Even when the in-game elements to be presented in the game are made to vary according to various factors such as characters, items are generated according to the story, thus reducing the workload of setting operations regarding the in-game elements handled in the game.

[0130] Also, since items are generated according to the character profile and the like, they are original items. For example, when playing the game repeatedly or when different players play, different stories and items can be generated for each play.

[0131] Also, compared with the method of preparing items in advance, the types of items can be enriched. Since the story or profile changes, the items also change, so even when the game is played repeatedly, the user will not get bored with the game. In this way, an attractive game can be provided to the user.

[0132] [Modification Example] Note that the type of game is not limited to a novel game. For example, the type of game may be a game including RPG, sports, action, puzzle, or fighting, etc. Furthermore, it does not have to be the above-mentioned games.

[0133] For example, the in-game elements may be characters or the like. Therefore, the characters or the like that appear in the game may be changed according to the story or profile or the like.

[0134] In the above description, the configuration in which the second learned model A22 generates the unknown story data DU12 has been described. However, the embodiment is not limited to this. For example, the unknown story data DU12 may be created by an administrator 5 or the like performing an operation of inputting text to create a sentence indicating a story (including those that improve based on the story generated by the AI), and input to the first learned model A21.

[0135] Therefore, the story generator 12 is not limited to a configuration using AI, and may be a device or the like that executes a program that supports an operation of an administrator 5 or the like for editing text.

[0136] Note that the story generator 12 may also serve as the server 11 or the like, or may be an information processing device separate from the server 11 or the like.

[0137] [Other Embodiments] In the above example, the information processing device performs both preprocessing on the learning model and execution processing using the learned model. However, the preprocessing and the execution processing do not necessarily have to be performed by the same information processing device. Also, the preprocessing and the execution processing do not necessarily have to be consistently executed by one information processing device. That is, each process and the storage of data, etc. may be performed by an information system or the like configured by a plurality of information processing devices.

[0138] Note that the learning process may be additionally performed after or before the execution process.

[0139] The above-described processing may be additionally performed by an information processing device other than the server 11 and the user terminal 20.

[0140] FIG. 12 is a diagram showing a configuration example using an auxiliary device. Compared with the example shown in FIG. 1, the configuration shown in FIG. 12 is different in that the auxiliary device 60 is added. Note that the auxiliary device 60 may be a configuration used temporarily.

[0141] The auxiliary device 60 is an information processing device installed near the user terminal 20 (in this example, it is installed near the user terminal 20A, but it may also be near other devices). And the auxiliary device 60 executes part or all of the specific processing on behalf of the user terminal 20 or the server 11.

[0142] For example, the auxiliary device 60 is equipped with a device specialized in graphic processing and performs graphic processing at high speed. In this way, an auxiliary device 60 or the like may be installed to perform so-called Edge Computing. In this way, the above-described processing may be executed by utilizing the hardware resources of various information processing devices. Therefore, the above-described processing may be executed by an information processing device different from those described above.

[0143] The above-described processing and the data used for the processing executed in this embodiment may be executed and stored by the information processing system. For example, the information processing system may execute or store in a plurality of information processing devices in order to realize redundancy, dispersion, parallelism, or a combination thereof. Therefore, the present invention may be realized by a device other than the hardware configuration shown above and a system other than the devices shown above.

[0144] Also, the program according to the present invention is not limited to a single program and may be an aggregate of a plurality of programs. Also, the program according to the present invention is not limited to being executed by a single device and may be executed in cooperation by a plurality of information processing devices. Furthermore, the role sharing of each information processing device is not limited to the above example. That is, part or all of the above-described processing may be executed by an information processing device different from the above-described information processing devices.

[0145] Furthermore, some or all of the means implemented by the program can also be implemented by hardware such as an integrated circuit. Furthermore, the program may be provided by being recorded on a non-transitory computer-readable recording medium. The recording medium refers to, for example, a hard disk, an SD card (registered trademark), an optical disk such as a DVD, or a server on the Internet. Therefore, the program may be distributed via an electric communication line such as the Internet.

[0146] Also, the information processing device and the like constituting the information processing system may be located overseas. That is, among the processes executed by the information processing system, the information processing device that executes some of the processes may be located overseas.

[0147] Note that the present invention is not limited to each of the embodiments illustrated above. Therefore, the present invention can be added to or modified in its components without departing from the technical gist. Thus, all of the technical matters included in the technical idea described in the claims are the subject of the present invention. Note that the embodiments illustrated above are suitable specific examples in implementation. And those skilled in the art can realize various modifications from the disclosed content, and such modifications are included in the technical scope described in the claims.

Explanation of Reference Numerals

[0148] 1: System 1F11: Story Data Generation Means 1F12: First Learning Data Input Means 1F13: Learning Means 1F21: Unknown Story Data Generation Means 1F22: First Unknown Data Input Means 1F23: Execution Means 1F24: Progress Means 4: User 5: Administrator 11: Server 12: Story Generator 20: User Terminal 60: Auxiliary Device 61: Learning device 62: Execution device A1: Learning model A11: First learning model A12: Second learning model A2: Learned model A21: First learned model A22: Second learned model D1: First unknown data D2: Second unknown data D3: Output data D4: Big data DL1: First learning data DL11: Story data DL12: Log data DL121: Image data DL122: Profile data DL13: First correct data DL2: Second learning data DL21: Keyword DL22: Second correct data DU1: First unknown data DU11: Unknown log data DU12: Unknown story data DU2: Second unknown data DU21: Unknown keyword

Claims

1. A computer, When log data regarding a player's operation in the game is input to a learned model learned using learning data including story data and log data, a generation means for generating output data regarding in-game elements; Based on the output data, a progression means for causing the in-game elements to appear in the game and enabling the game to progress using the in-game elements; Functioning as such, Both the story data and the log data are input data, The data regarding the in-game elements is Output data, The story data is A program that shows a story regarding the in-game elements in text or images .

2. A generation means for generating output data regarding in-game elements when log data regarding a player's operation in the game is input to a learned model learned using learning data including story data and log data; Based on the output data, a progression means for causing the in-game elements to appear in the game and enabling the game to progress using the in-game elements; Comprising, Both the story data and the log data are input data, The data regarding the in-game elements is Output data, The story data is A system that shows a story regarding the in-game elements in text or images .

Images