Program, information processing method, and information processing apparatus

Abstract

To provide a program and the like, configured to maximize virtual currency which can be obtained by lending an NFT (Non-Fungible Token) to a game player.SOLUTION: A program causes a computer to execute processing for acquiring operation history information of players in a game in which virtual currency can be obtained, and NFT information related to NFTs which can be lent to the players and are used in the game, and outputting allocation information by inputting the acquired operation history information and the NFT information to a model which is trained so as to output allocation information indicating NFTs to be lent to players and estimated operation time of the players when the operation history information and the NFT information are input.SELECTED DRAWING: Figure 5

Description

[Technical field]

[0001] The present invention relates to a program, an information processing method, and an information processing device. [Background technology]

[0002] Electronic game systems using blockchain technology are known (for example, see Patent Literature 1). Some of these games use items, characters, etc. used in the game as NFTs (Non-Fungible Tokens), and allow players to earn cryptocurrency according to their playing results. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2021-152815 Summary of the Invention [Problem to be solved by the invention]

[0004] On one aspect, the objective is to provide a program etc. that can maximize the virtual currency that can be obtained by lending NFTs to game players. [Means for solving the problem]

[0005] In one aspect, the program acquires operation history information of each player in a game in which virtual currency can be acquired, and NFT information regarding NFTs (Non-Fungible Tokens) for use in the game that can be lent to the player, and when the operation history information and NFT information are input, causes a computer to execute a process of outputting the allocation information by inputting the acquired operation history information and NFT information into a model that has been trained to output allocation information indicating NFTs to be lent to each player and the expected operation time of each player. Effect of the Invention

[0006] On the one hand, it allows gamers to maximize the cryptocurrency they can earn by lending NFTs to them. [Brief description of the drawings]

[0007] [Figure 1] An explanatory diagram showing an example configuration of an NFT lending system. [Diagram 2] FIG. 2 is a block diagram showing an example of the configuration of a server. [Diagram 3] FIG. 1 is an explanatory diagram showing an example of a record layout of a player DB and an NFTDB. [Figure 4] FIG. 1 is an explanatory diagram regarding the scholarship system. [Diagram 5] FIG. 1 is an explanatory diagram showing an overview of a first embodiment. [Figure 6] 13 is a flowchart showing the steps of a learning model generation process. [Figure 7] 13 is a flowchart showing a procedure for estimating allocation information. [Figure 8] 13 is a flowchart showing the steps of a learning model update process. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008] The present invention will now be described in detail with reference to the drawings showing embodiments thereof. (Embodiment 1) FIG. 1 is an explanatory diagram showing an example of the configuration of an NFT lending system. In this embodiment, an NFT lending system that lends NFTs to players of games in which virtual currency can be acquired is described. The NFT lending system includes an information processing device 1, terminals 2, 2, 2..., and a game server 3. Each device is communicatively connected to a network N such as the Internet.

[0009] Terminal 2 is a terminal device of a player who plays a game, and is, for example, a personal computer, a smartphone, a tablet terminal, a game console, etc. As will be described later, a player can borrow an NFT through this system and use the NFT to play a game.

[0010] The game server 3 is a server computer that provides a game for a player to play. In this embodiment, the game server 3 provides a game in which virtual currency can be acquired according to the result of playing (a game based on the concept of "Play to Earn"). As will be described later, when a game player borrows an NFT such as an in-game item through this system to play the game and acquire virtual currency, the acquired virtual currency is distributed among the player, the administrator of this system, and the investor who provided the NFT.

[0011] The information processing device 1 is an information processing device capable of various information processing and transmission and reception of information, such as a server computer or a personal computer. In this embodiment, the information processing device 1 is a server computer, and will be referred to as the server 1 below for simplicity. As described below, the server 1 estimates the optimal allocation of the type and quantity of NFTs to be lent to each player in order to maximize the virtual currency obtained by lending NFTs to game players. Specifically, when the server 1 inputs operation history information indicating the operation history (play history) of each player, NFT information regarding NFTs that can be lent to players, and the like, the server 1 constructs a machine learning model (learning model 50 described below) that outputs allocation information indicating the NFTs to be lent to each player and the expected operation time of each player, and estimates the optimal allocation information using the model.

[0012] 2 is a block diagram showing an example of the configuration of the server 1. The server 1 includes a control unit 11, a main storage unit 12, a communication unit 13, and an auxiliary storage unit . The control unit 11 has one or more arithmetic processing devices such as a CPU (Central Processing Unit), an MPU (Micro-Processing Unit), a GPU (Graphics Processing Unit), etc., and performs various information processing, control processing, etc. by reading and executing a program P stored in the auxiliary storage unit 14. The main storage unit 12 is a temporary storage area such as an SRAM (Static Random Access Memory) or a DRAM (Dynamic Random Access Memory), and temporarily stores data required for the control unit 11 to execute arithmetic processing. The communication unit 13 is a communication module for performing processing related to communication, and transmits and receives information to and from the outside.

[0013] The auxiliary storage unit 14 is a non-volatile storage area such as a large-capacity memory or a hard disk, and stores a program P (program product) and other data necessary for the control unit 11 to execute processing. The auxiliary storage unit 14 also stores a learning model 50, a player DB 141, and an NFT DB 142. The learning model 50 is a machine learning model that has learned predetermined training data, and is a model that has been trained to output allocation information indicating NFTs to be lent to each player and the expected operating time of the player when operation history information of each player in the game, NFT information regarding NFTs that can be lent to the player, and the available operating time of each player are input. The player DB 141 is a database that stores information on each player. The NFT DB 142 is a database that stores information on NFTs that can be lent to the player.

[0014] The auxiliary storage unit 14 may be an external storage device connected to the server 1. The server 1 may be a multi-computer consisting of a plurality of computers, or may be a virtual machine virtually constructed by software.

[0015] In the present embodiment, the server 1 is not limited to the above configuration, and may include, for example, an input unit for accepting operation input, a display unit for displaying images, etc. The server 1 may also include a reading unit for reading a portable storage medium 1a such as a CD (Compact Disk)-ROM or a DVD (Digital Versatile Disc)-ROM, and may read and execute the program P from the portable storage medium 1a.

[0016] FIG. 3 is an explanatory diagram showing an example of the record layout of the player DB 141 and the NFT DB 142. The player DB141 includes a player ID column, a player name column, a play history column, an operating rate column, an acquisition efficiency column, and an available operating time column. The player ID stores a player ID for identifying each player. The player name column, the play history column, the operating rate column, the acquisition efficiency column, and the available operating time column store, in association with the player ID, the player name, the game play history, the player's operating rate (e.g., playing time per day), the virtual currency acquisition efficiency (e.g., the amount of virtual currency acquired per unit time), and the available operating time (e.g., play time per month).

[0017] The NFTDB142 includes an NFTID column, an NFT name column, and a unit price column. The NFTID column stores an NFTID for identifying each NFT. The NFT name column and the unit price column store the name of the NFT and the unit price of the NFT (for example, the price per token when purchased with virtual currency) in association with the NFTID, respectively.

[0018] 4 is an explanatory diagram regarding the scholarship system. Before describing the outline of the present embodiment, the scholarship system for lending NFTs to game players (scholars) will be described with reference to FIG. 4.

[0019] The scholarship system is a mechanism in which NFTs are lent to game players in exchange for receiving a portion of the rewards (cryptocurrency) earned in the game from the players. In the scholarship system, the party that lends the NFT is also called a "guild," and the player who borrows the NFT is also called a "scholar."

[0020] In this system, the guild (the administrator of this system) receives management commission from investors who own NFTs and lends the NFTs to scholars. Scholars use the NFTs to play games and earn virtual currency. The earned virtual currency is distributed among the investors, guilds, and scholars in a predetermined ratio.

[0021] Under such a scholarship system, the guild lends NFTs to a large number of scalars, and each scalar earns rewards by playing the game. However, there is currently no established method for determining how much of each type of NFT should be lent to each scalar in order to maximize profits, and the decision is made manually. Therefore, in this embodiment, a machine learning model is used to automate the decision and maximize profits.

[0022] Fig. 5 is an explanatory diagram showing an overview of the first embodiment. The overview of the present embodiment will be described with reference to Fig. 5.

[0023] In this embodiment, the server 1 maximizes profits by using a learning model 50 that outputs optimal allocation of NFTs and operating time (play time). The learning model 50 is a machine learning model that has learned predetermined training data, such as a neural network.

[0024] The learning model 50 may be a machine learning model other than a neural network, such as a decision tree or a support vector machine (SVM).

[0025] When the learning model 50 receives operation history information indicating the operation history of each player (scalar), NFT information regarding NFTs that can be lent to the player, and the available operation time of each player, the learning model 50 outputs allocation information indicating the type and quantity of NFTs to be lent to each player and the expected operation time of each player. The operation history information includes, for example, the play time of the past game, the amount of virtual currency acquired in the past, the operation rate (play time per unit period), the efficiency of acquiring virtual currency (amount acquired per unit time), and the like. The NFT information includes the type and quantity of NFTs that can be lent to the player. The available operation time is, for example, the time that can be played in a certain period of time in the future (for example, the next month).

[0026] In this embodiment, the input to the learning model 50 includes the available operating time of each player, but it is not essential to use the available operating time as an input, and the learning model 50 only needs to be able to output allocation information using at least the operating history information and NFT information as input.

[0027] The server 1 may directly use the operation history information of each player as input to the learning model 50, but in this embodiment, the server 1 determines the rank of each player from the operation history information and uses the rank as input. For example, the server 1 determines the rank according to the length of time the game is played, the amount of virtual currency acquired, the operation rate, and the efficiency of acquiring virtual currency.

[0028] The server 1 generates a learning model 50 using training data in which correct allocation information is associated with operation history information, NFT information, and available operation time for training. First, the server 1 determines the rank of each player from the operation history information of each player given for training. The server 1 then inputs the determined rank, NFT information for training, and available operation time to the learning model 50, thereby outputting allocation information. The server 1 compares the output allocation information with a correct answer value, and updates parameters such as weights between neurons so that the two are close to each other. The server 1 learns each operation history information, NFT information, and available operation time given for training, optimizes the parameters, and finally generates a learning model 50.

[0029] When actually estimating (outputting) allocation information using the learning model 50, the server 1 acquires operation history information of each player, NFT information that can be lent to the player, and the available operation time of each player. As in the case of learning, the server 1 determines the rank of each player from the operation history information. The server 1 then inputs the determined rank, the acquired NFT information, and the available operation time into the learning model 50, and outputs allocation information indicating the NFTs to be lent to each player and the expected operation time of each player, as shown in the table in the upper right of FIG. 5.

[0030] In this embodiment, the server 1 performs a simulation to determine whether the target amount of virtual currency earned can be achieved if NFTs are lent in accordance with the output allocation information, and if it is unlikely that the target can be achieved, it performs a process of recalculating (outputting) the allocation information using the learning model 50.

[0031] First, the server 1 estimates the expected amount of virtual currency to be acquired when lending NFTs to players according to the allocation information output from the learning model 50. For example, the server 1 estimates the expected amount of acquisition by multiplying each player's past acquisition efficiency by the expected operating time included in the allocation information. Alternatively, the server 1 may train the learning model 50 to simultaneously output the expected amount of virtual currency to be acquired in addition to the allocation information, and use the learning model 50 to estimate the expected amount of acquisition.

[0032] The server 1 calculates an evaluation value for each player by evaluating the expected play result based on the estimated expected acquisition amount and the expected operation time estimated (output) as allocation information. For example, the server 1 calculates a high evaluation value when the expected operation time is less than a specified time (recommended time) and the expected acquisition amount is equal to or greater than a specified value. The server 1 calculates a medium evaluation value when the expected operation time is equal to or greater than a specified time and the expected acquisition amount is equal to or greater than a specified value. The server 1 calculates a low evaluation value when the expected operation time is equal to or greater than a specified time and the expected acquisition amount is less than a specified value. The server 1 calculates a minimum evaluation value when the expected operation time is less than a specified time and the expected acquisition amount is less than a specified value.

[0033] In addition, the server 1 calculates the evaluation value by referring to the efficiency of acquiring virtual currency, attendance (whether or not the player plays every day), etc. For example, the server 1 calculates an evaluation value for each player, and calculates an overall evaluation value by summing up or averaging the evaluation values ​​of each player.

[0034] The server 1 determines whether the calculated evaluation value is equal to or greater than a predetermined target value. If it is determined that the evaluation value is equal to or greater than the target value, the server 1 grants (lends) NFTs to each player according to the allocation information output from the learning model 50. This allows each player to play the game using NFTs and obtain virtual currency as a reward.

[0035] On the other hand, if the server 1 determines that the evaluation value is not equal to or greater than the target value as a result of the simulation, the server 1 performs a re-simulation by recalculating (outputting) the allocation information. For example, the server 1 changes parameters such as NFT information (type and quantity of NFTs that can be lent to each player) and available operating time to be input to the learning model 50, inputs the changed NFT information, etc. to the learning model 50, and re-outputs the allocation information. The server 1 recalculates the evaluation value based on the re-output allocation information, and determines whether the evaluation value is equal to or greater than the target value.

[0036] The server 1 repeats the estimation of the allocation information by the learning model 50 and the simulation until it determines that the evaluation value is equal to or greater than the target value. This makes it possible to identify allocation information that is expected to result in an amount of virtual currency earned being equal to or greater than the target.

[0037] As described above, according to this embodiment, when lending NFTs to each player of a game in which virtual currency can be earned, it is possible to automatically determine what type and how many NFTs to lend to each player, thereby maximizing the amount of virtual currency earned.

[0038] Fig. 6 is a flowchart showing the procedure of the generation process of the learning model 50. The process contents when generating the learning model 50 by machine learning will be described with reference to Fig. 6. The control unit 11 of the server 1 acquires training data for generating the learning model 50 (step S11). The training data is data in which operation history information indicating the operation history of each player in a game in which virtual currency can be acquired, NFT information regarding NFTs that can be lent to a player and used in the game, and correct answers of allocation information indicating the NFTs to be lent to each player and the expected operation time of each player are associated with each other with respect to the available operation time of each player.

[0039] The control unit 11 determines the rank of each player based on the training operation history information (step S12). The control unit 11 generates a trained learning model 50 based on the determined rank and other training data so as to output allocation information when the player's rank, NFT information, and available operation time are input (step S13), and ends the series of processes.

[0040] 7 is a flowchart showing the procedure of the allocation information estimation process. Based on FIG. 7, the process of estimating optimal allocation information using the learning model 50 and lending NFTs to a player will be described. The control unit 11 of the server 1 acquires operation history information of each player in the game (step S31). The control unit 11 determines the rank of each player based on the acquired operation history information (step S32).

[0041] The control unit 11 acquires NFT information regarding NFTs that can be lent to players and the available operation time of each player (step S33). The control unit 11 inputs the rank determined in step S32 and the NFT information and available operation time acquired in step S33 into the learning model 50, and outputs allocation information indicating the NFTs to be lent to each player and the expected operation time of each player (step S34).

[0042] The control unit 11 estimates the expected amount of virtual currency to be acquired when lending NFTs to the player according to the output allocation information (step S35). The control unit 11 calculates an evaluation value that evaluates the expected play results of the player based on the estimated expected amount of acquisition and the expected operating time (step S36).

[0043] The control unit 11 determines whether the calculated evaluation value is equal to or greater than the target value (step S37). If it is determined that the evaluation value is not equal to or greater than the target value (S37: NO), the control unit 11 changes the NFT information that can be lent to the player and the available operation time of each player (step S38), and returns the process to step S34. In this case, the control unit 11 outputs (recalculates) the allocation information again by inputting the changed NFT information, etc., into the learning model 50, recalculates the evaluation value, and redetermines whether the evaluation value is equal to or greater than the target value.

[0044] When it is determined that the evaluation value is equal to or greater than the target value (S37: YES), the control unit 11 grants (lends) NFTs to each player according to the allocation information output in step S34 (step S39), and ends the series of processes.

[0045] In the above, the allocation information is estimated for each individual player, but the present embodiment is not limited to this, and the allocation information may be estimated for each team consisting of multiple players. A team is, for example, a unit of players participating in a competitive game played by a group of multiple people. When estimating allocation information for such a game, the server 1 acquires operation history information for each team consisting of multiple players, and inputs it together with NFT information into the learning model 50, thereby outputting allocation information indicating NFTs to be lent to each team and the expected operation time of each team. This makes it possible to predict the optimal allocation of NFTs to be lent to each team, even for games played by multiple people.

[0046] As described above, according to this first embodiment, it is possible to maximize the virtual currency that can be obtained by lending NFTs to game players.

[0047] (Embodiment 2) In this embodiment, a form will be described in which the learning model 50 is updated (re-learned) based on the operation history information of each player after lending NFTs according to the allocation information. Note that the same reference numerals will be used to designate the same contents as in the first embodiment, and the description will be omitted.

[0048] Fig. 8 is a flowchart showing the procedure of the update process of the learning model 50. The process of updating the learning model 50 by performing re-learning will be described with reference to Fig. 8. Note that the description will be given on the assumption that the server 1 has already output (estimated) the allocation information according to the flowchart of Fig. 7 and granted (lent) NFTs to each player according to the allocation information. The control unit 11 of the server 1 acquires operation history information indicating the operation history of each player after lending NFTs according to the allocation information (step S201). The control unit 11 determines the rank of each player based on the acquired operation history information (step S202).

[0049] The control unit 11 acquires the NFT information and operable time input to the learning model 50 in step S34, and the allocation information output from the learning model 50 (step S203). The control unit 11 updates the learning model 50 based on the rank determined in step S202, and the NFT information, operable time, and allocation information acquired in step S203 (step S204). That is, the control unit 11 uses the rank determined in step S202 and the NFT information and operable time acquired in step S203 as input data for training, and uses the allocation information acquired in step S203 as the correct value of the output data, and updates parameters such as the weights between neurons. The control unit 11 ends the series of processes.

[0050] As described above, according to the second embodiment, the estimation accuracy of the allocation information by the learning model 50 can be improved.

[0051] The embodiments disclosed herein are illustrative in all respects and should not be considered as limiting. The scope of the present invention is defined by the claims, not by the above meaning, and is intended to include all modifications within the scope and meaning equivalent to the claims.

[0052] The matters described in each embodiment can be combined with each other. In addition, the independent claims and dependent claims described in the claims can be combined with each other in any and all combinations regardless of the citation format. Furthermore, the claims use a format in which a claim cites two or more other claims (multi-claim format), but this is not limited to this. They may also be written using a format in which a multiple claim cites at least one other multiple claim (multi-multi claim). [Explanation of symbols]

[0053] 1. Server (information processing device) 11 Control section 12 Main memory 13. Communications Department 14 Auxiliary storage P Program 2. Terminal 3 Game Server

Claims

1. Acquire operation history information of each player in a game in which virtual currency can be acquired, and NFT information regarding NFTs (Non-Fungible Tokens) that can be lent to the player and used in the game; When operation history information and NFT information are input, a model that has been trained to output allocation information indicating NFTs to be lent to each player and the expected operation time of each player is inputted, and the allocation information is outputted by inputting the acquired operation history information and NFT information. A program that causes a computer to carry out processing.

2. Further acquiring the available time of each of the players; The operation history information, NFT information, and available time are input to the model, and the allocation information is output. The program according to claim 1.

3. Estimate an expected amount of the virtual currency to be acquired when the NFT is lent to the player in accordance with the allocation information; calculating an evaluation value for evaluating the expected play result of the player based on the expected operating time and the expected winning amount; determining whether the evaluation value is equal to or greater than a target value; If it is determined that the evaluation value is not equal to or greater than a target value, the NFT information is changed; The allocation information is output again by re-inputting the changed NFT information and the operation history information into the model. The program according to claim 1.

4. determining a rank for each of the players based on the operation history information; The rank and the NFT information are input to the model to output the allocation information. The program according to claim 1.

5. The operation history information is acquired on a team basis consisting of a plurality of the players; The operation history information and the NFT information of each team are input into the model, and the allocation information indicating the NFT to be lent to each team and the expected operation time of each team is output. The program according to claim 1.

6. acquiring the operation history information indicating an operation history after lending the NFT to the player in accordance with the allocation information; The model is updated based on the acquired operation history information, the NFT information, and the allocation information. The program according to claim 1.

7. Acquire operation history information of each player in a game in which virtual currency can be acquired, and NFT information regarding NFTs that can be lent to the player and can be used in the game; When operation history information and NFT information are input, a model that has been trained to output allocation information indicating NFTs to be lent to each player and the expected operation time of each player is input, and the allocation information is output by inputting the acquired operation history information and NFT information. An information processing method in which processing is performed by a computer.

8. An information processing device including a control unit, The control unit: Acquire operation history information of each player in a game in which virtual currency can be acquired, and NFT information regarding NFTs that can be lent to the player and can be used in the game; When operation history information and NFT information are input, a model that has been trained to output allocation information indicating NFTs to be lent to each player and the expected operation time of each player is input, and the allocation information is output by inputting the acquired operation history information and NFT information. Information processing device.

Images