Gaming information provision device and gaming information provision method

Abstract

To efficiently and appropriately determine whether or not a game machine as an introduction candidate is to be introduced to a game parlor.SOLUTION: Model data on a new game machine is input to a nation-wide out average prediction model and a parlor out average prediction model, and a first week out average prediction value for the new game machine is predicted (S1). For example, first week out average prediction values are predicted as "26,000" balls for the nation-wide average, and as "36,000" balls for a user's game parlor average. Using the first week out average prediction values and a first week result of a model type average, a first week result ratio and a first week contribution degree are calculated (S2). If the first week out average prediction value of the nation-wide average is "26,000" balls, and the model type average is "15,000" balls, for example, the first week result ratio of the nation-wide average is calculated as "1.7." If the first week out average prediction value of the user's game parlor is "36,000" balls, and the model type average is "18,000" balls, the first week result ratio of the user's game parlor is calculated as "2.0." The first week contribution degree is calculated as "1.2."SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a game information providing apparatus and a game information providing method capable of efficiently and appropriately determining whether or not a candidate gaming machine to be introduced should be introduced into a game hall. [Background Art]

[0002] Conventionally, game halls such as pachinko parlors have installed a large number of gaming machines of various models. They regularly consider introducing new models of gaming machines and also replace existing gaming machines. Here, even if a new model of gaming machine is introduced, the new model of gaming machine may fail to attract the interest of players and end up being unsuccessful. Since the price per gaming machine is high, the impact when the introduction of a new model fails is significant.

[0003] For this reason, a technology is known that predicts the operation after a certain period of time has elapsed since introduction, in order to determine the future operation status as soon as possible and early update models whose operation status will deteriorate (see, for example, Patent Document 1). [Prior Art Documents] [Patent Documents]

[0004] [Patent Document 1] Japanese Unexamined Patent Publication No. 2017-176586 [Summary of the Invention] [Problems to be Solved by the Invention]

[0005] However, the technology of Patent Document 1 described above is premised on the introduction of a new model of gaming machine, and does not predict the operation of the gaming machine before introduction. Therefore, when introducing a new model of gaming machine, it is inevitable to rely on the experience of the store manager of the game hall, a consultant, or the like. Even if this point is carefully considered when introducing a new model of gaming machine, it is not uncommon for the introduction to end in failure. The same problem arises not only when introducing a new model of gaming machine, but also when newly introducing an existing model of gaming machine.

[0006] The present invention has been made to solve the problems of the prior art described above, and aims to provide a gaming information providing device and a gaming information providing method that can efficiently and appropriately determine whether or not a candidate gaming machine should be introduced to a gaming parlor. [Means for solving the problem]

[0007] To solve the above problems, the present invention provides a gaming information providing device that outputs gaming information including the introduction prediction results of candidate gaming machines to be newly introduced to a gaming parlor, comprising: a first trained model generation means that generates a first trained model targeting the gaming parlor by performing machine learning based on performance information at the gaming parlor, including the average number of gaming media played by players over a predetermined period, and the model information of the specific type of gaming machine, with respect to a specific type of gaming machine installed in the gaming parlor; a first prediction result generation means that inputs input information including the model information of the candidate gaming machine belonging to the specific type of machine into the first trained model to generate a first prediction result; and a device located in a predetermined region The invention is characterized by comprising: a second trained model generation means that generates a second trained model targeting a predetermined region by performing machine learning based on performance information in a predetermined region, including an average number of game media played by players over a predetermined period, and model information of the specific type of game machine, with respect to the game machine of a specific model installed in multiple game parlors; a second prediction result generation means that inputs input information, including model information of the candidate game machine belonging to the specific type, into the second trained model to generate a second prediction result; and a calculation means that calculates the introduction prediction result of the candidate game machine based on at least the first prediction result and the second prediction result.

[0008] Furthermore, in the above invention, the calculation means is The aforementioned amusement store The method is characterized by calculating the predicted introduction result of the candidate gaming machine based on the performance information in the specified region, the first prediction result, the performance information in the specified region, and the second prediction result.

[0009] Furthermore, in the above invention, the calculation means includes the average output included in the first prediction result. The aforementioned amusement store The system is characterized by comprising: a first historical level ratio calculation means for calculating a first historical level ratio by dividing the average output included in the performance information in the area; and a second historical level ratio calculation means for calculating a second historical level ratio by dividing the average output included in the second prediction result by the average output included in the performance information in the predetermined area.

[0010] Furthermore, the present invention is characterized in that, in the above invention, the calculation means further comprises a means for calculating the store's contribution, which calculates the store's contribution to the gaming store by dividing the first past level ratio by the second past level ratio.

[0011] Furthermore, in the above invention, the first trained model generation means uses a machine learning framework for multiple regression models of a linear regression algorithm. The aforementioned amusement store The second pre-trained model generation means generates a pre-trained linear regression model for the target region, and the second pre-trained model generation means generates a pre-trained linear regression model for the target region using a machine learning framework for multiple regression models of the linear regression algorithm.

[0012] Furthermore, the present invention is characterized in that, in the above invention, the first trained model generation means receives the number of balls dispensed at multiple percenttile values, sales model information, and model characteristics of the specific model type of gaming machine installed in the gaming parlor as model information for the specific model type of gaming machine installed in the gaming parlor, and the second trained model generation means receives the number of balls dispensed at multiple percenttile values, sales model information, and model characteristics of the specific model type of gaming machine installed in gaming parlors located in the predetermined region as model information for the specific model type of gaming machine installed in multiple gaming parlors located in the predetermined region.

[0013] Furthermore, in the above invention, the first trained model generation means provides information including the average number of payouts per week from the first week to the tenth week after the start of play for multiple gaming machines installed in the gaming parlor. The aforementioned amusement store The second trained model generation means accepts performance information in the predetermined region, and is characterized by accepting information including the average number of payouts per week from the first week to the tenth week after the start of play for multiple gaming machines installed in multiple gaming parlors located in the predetermined region, as performance information in the predetermined region.

[0014] Furthermore, the present invention relates to a method for providing game information in a game information providing device that outputs game information including the prediction result of the introduction of candidate game machines to be newly introduced to a game parlor, comprising: a first trained model generation step of generating a first trained model targeting the game parlor by performing machine learning based on performance information at the game parlor, including the average number of game media played by players over a predetermined period with respect to a specific type of game machine installed in the game parlor, and the model information of the specific type of game machine; a first prediction result generation step of inputting input information including the model information of candidate game machines belonging to the specific type of game machine into the first trained model to generate a first prediction result; and located in a predetermined region. The method is characterized by including, with respect to the gaming machines of a specific type installed in multiple amusement parlors, a second trained model generation step of generating a second trained model targeting the predetermined region by performing machine learning based on performance information in the predetermined region, including an average number of game media played by players over a predetermined period, and model information of the gaming machines of the specific type; a second prediction result generation step of inputting input information, including model information of the candidate gaming machines belonging to the specific type, into the second trained model to generate a second prediction result; and a calculation step of calculating the introduction prediction result of the candidate gaming machines based on at least the first prediction result and the second prediction result. [Effects of the Invention]

[0015] According to the present invention, it is possible to efficiently and appropriately determine whether a gaming machine that is a candidate for introduction should be introduced into a game arcade. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] [Figure 1] FIG. 1 is an explanatory diagram outlining a game information providing system according to Embodiment 1. [Figure 2] FIG. 2 is a diagram showing the system configuration of the game information providing system according to Embodiment 1. [Figure 3] FIG. 3 is a diagram showing the external configuration of the inter-stand card processing machine and the gaming machine shown in FIG. 2. [Figure 4] FIG. 4 is a functional block diagram showing the configuration of the inter-stand card processing machine shown in FIG. 2. [Figure 5] FIG. 5 is a diagram showing an example of own device status data and card data shown in FIG. 4. [Figure 6] FIG. 6 is a functional block diagram showing the configuration of the gaming machine shown in FIG. 2. [Figure 7] FIG. 7 is a functional block diagram showing the configuration of the management device shown in FIG. 2. [Figure 8] FIG. 8 is a diagram showing an example of card management data, device management data, and member management data shown in FIG. 7. [Figure 9] FIG. 9 is a diagram showing an example of operation data, game-related card processing history data, and non-game card processing history data shown in FIG. 7. [Figure 10] FIG. 10 is a diagram showing an example of count settlement processing history data, game history data, and movement data shown in FIG. 7. [Figure 11] FIG. 11 is a diagram showing an example of operation performance data, first-week performance ratio data, and first-week contribution data shown in FIG. 7. [Figure 12] FIG. 12 is a functional block diagram showing the configuration of the information management device shown in FIG. 2. [Figure 13] FIG. 13 is a diagram showing an example of model data and model-specific performance data shown in FIG. 12. [Figure 14]Figure 14 shows an example of the performance data by type shown in Figure 12. [Figure 15] Figure 15 shows an example of the percentile data and forecast data shown in Figure 12. [Figure 16] Figure 16 shows an example of the first-week performance comparison data and first-week contribution data shown in Figure 12. [Figure 17] Figure 17 shows an example of the features used in model prediction according to Embodiment 1. [Figure 18] Figure 18 is a flowchart showing the processing procedure for generating an out-of-bounds average prediction model according to Embodiment 1. [Figure 19] Figure 19 is a flowchart showing the processing procedure for calculating the first-week performance ratio and first-week contribution according to Embodiment 1. [Figure 20] Figure 20 is an explanatory diagram illustrating the overview of a modified version of the game information provision system. [Figure 21] Figure 21 is a flowchart showing the processing procedure for calculating the performance ratio and contribution related to the modified example. [Figure 22] Figure 22 is an explanatory diagram illustrating the overview of the game information provision system according to Embodiment 2. [Figure 23] Figure 23 is an explanatory diagram (part 1) of the overview of the game information provision system according to Embodiment 3. [Figure 24] Figure 24 is an explanatory diagram (part 2) of the overview of the game information provision system according to Embodiment 3. [Figure 25] Figure 25 is an explanatory diagram (part 1) of the overview of the game information provision system according to Embodiment 4. [Figure 26] Figure 26 is an explanatory diagram (part 2) of the overview of the game information provision system according to Embodiment 4. [Modes for carrying out the invention]

[0017] [Embodiment 1] The embodiments of the game information provision device and game information provision method according to this first embodiment will be described in detail below with reference to the drawings.

[0018] In this specification, "balls held" refers to data indicating the gaming media acquired by a player during gameplay, which can only be used again on the same day (before the arcade's closing time). "Number of balls held" refers to the number of gaming media acquired by a player through gameplay. "Stored balls" refers to data indicating gaming media deposited by a player with the arcade, which can be used again on subsequent days (after the arcade's closing time). "Number of stored balls" refers to the number of gaming media deposited by a player with the arcade. Balls held can be used by both general and member players, and are used when moving to a different gaming machine, etc. Stored balls can, in principle, only be used by member players, and are used when acquiring gaming media to play on a gaming machine on subsequent days. "Number of game balls" refers to the number of game balls available for a player to play.

[0019] <Overview of the game information provision system according to Embodiment 1> First, an overview of the game information provision system according to this first embodiment will be described. Figure 1 is an explanatory diagram illustrating the overview of the game information provision system according to this first embodiment.

[0020] A gaming parlor has multiple gaming machines installed, and each gaming machine is equipped with an inter-machine card processing device. The gaming information provision system collects historical data from these inter-machine card processing devices and operational data from the gaming machines. From this historical data and operational data, it generates operational performance data, which is an aggregate of the average weekly output for each gaming machine since its introduction.

[0021] The gaming information provision system stores this operational performance data for all stores nationwide and for each individual store, and uses this data, along with machine model data related to gaming machines, to perform machine learning to predict the average payout. By performing this machine learning training for all stores nationwide and for each individual store, a nationwide average payout prediction model and a store-specific average payout prediction model are generated.

[0022] By inputting data on the new slot machine being considered for introduction into these pre-trained machine learning models—the nationwide average payout prediction model and the store average payout prediction model—we can predict the average payout for the first week of the new slot machine (S1).

[0023] For example, the predicted average payout for the first week nationwide is 26,000 balls, while the predicted average payout for the first week at your own store is 36,000 balls.

[0024] Using the first-week average output forecast and the average for each machine type in the first week's actual results, the first-week performance ratio and the first-week contribution are calculated (S2). The first-week performance ratio is calculated by dividing the first-week average output forecast by the average for each machine type, and the first-week contribution is calculated by dividing the store's first-week performance ratio by the national average first-week performance ratio.

[0025] For example, if the national average first-week average payout forecast is "26,000" balls, and the national average for each machine type is "15,000" balls, then the national average first-week performance ratio is calculated to be "1.7". If your store's first-week average payout forecast is "36,000" balls, and the store's average for each machine type is "18,000" balls, then your store's first-week performance ratio is calculated to be "2.0". In addition, the first-week contribution is calculated to be "1.2".

[0026] The first-week performance ratio is a value that indicates the expected value of a new machine relative to past performance, based on the average output during the first week. The national average first-week performance ratio is the expected value relative to the average performance of stores nationwide, while the first-week performance ratio for a specific store is the expected value relative to the performance of that specific store. Therefore, it is possible to perform both national and individual store analyses.

[0027] Furthermore, the first-week contribution is a value that shows the expected value at the store compared to the national average expected value, allowing for comparative analysis of the expected value at the store with that of other stores nationwide.

[0028] As described above, the gaming information provision system according to this embodiment 1 uses a national average out-of-pocket prediction model and a store average out-of-pocket prediction model, which are pre-trained models using machine learning, to predict the average out-of-pocket value for the first week. The system is configured to calculate the first-week performance ratio and the first-week contribution using this predicted average out-of-pocket value for the first week and the average for each machine type in the first week's performance. This makes it possible to efficiently and appropriately determine whether or not a candidate gaming machine should be introduced to a gaming store.

[0029] <System Configuration> Next, the system configuration of the game information provision system according to this embodiment 1 will be described. Figure 2 is a diagram showing the system configuration of the game information provision system according to this embodiment 1. As shown in Figure 2, a game parlor is equipped with multiple game machines 20 and inter-machine card processing machines 10, each of which is provided in conjunction with a game machine 20.

[0030] The inter-machine card processing machine 10 is connected to the store's network, which is a communication line, via the island controller 30. The island controller 30, management device 50, prize management device 60, and settlement machine 70 are connected to the communication line.

[0031] Furthermore, the management device 50 is connected to the internet, which is an external network. The information management device 80 is connected to the internet.

[0032] The gaming machine 20 is a device that allows players to play by launching game balls, which are sealed inside the device, onto the game board. These game balls are physical "balls" used in the game and are different from "game balls" that the gaming machine 20 handles as data. The game board of the gaming machine 20 is provided with multiple prize areas (prize slots), and a predetermined number of game balls are awarded as prizes based on when a game ball passes through these prize areas. The game board is also provided with a predetermined number of starting areas (start slots), and a predetermined lottery is conducted based on when a game ball passes through these starting areas. If the lottery is a jackpot, predetermined movable parts are activated to improve the probability of the game ball passing through the aforementioned prize areas or other starting areas, thereby providing an advantage in the game. There may also be areas that serve as both starting areas and prize areas.

[0033] The prize-winning areas are equipped with prize-winning sensors to detect when a game ball passes into them. The prize-winning sensors detect when a game ball is played into a prize-winning area (i.e., wins a prize). The control unit of the game machine 20 also has a prize-winning ball memory that stores how many game balls are awarded as prizes for each prize-winning area, and a playable ball count memory that shows the number of game balls available. The control unit of the game machine 20 also periodically transmits the number of game balls to the inter-machine card processing machine 10.

[0034] When the inter-machine card processing machine 10 receives a deposit from a player, it associates the prepaid value corresponding to the deposit amount with the identification information (card ID) of the card stored inside the machine. This prepaid value can be used to dispense game balls. The inter-machine card processing machine 10 can also accept the insertion of cards associated with prepaid value, held balls, stored balls, etc. At the end of the game, it ejects the card associated with the prepaid value, held balls, stored balls, etc., and returns it to the player. The inter-machine card processing machine 10 also transmits the history of operations such as deposit, card insertion, and card ejection to the management device 50.

[0035] The gaming machine 20 transmits out data to the inter-machine card processing unit 10 when a predetermined number of game balls have been launched onto the game board. The gaming machine 20 also transmits safe data to the inter-machine card processing unit 10 when a game ball passes through the winning area (so-called winning) and a predetermined number of prize balls have been added to the game balls. Furthermore, if a special state such as a jackpot occurs as a result of the game, the gaming machine 20 transmits special prize data indicating the occurrence of such a state to the inter-machine card processing unit 10.

[0036] Out data corresponds to the number of balls played by the player onto the game board. Safe data corresponds to the number of balls awarded for winning. Special prize data indicates the status of the game machine 20, such as a jackpot. Each piece of data is transmitted to the management device 50 via the inter-machine card processing machine 10, and the management device 50 can obtain a history of the game machine 20's operation by accumulating the out data, safe data, and special prize data.

[0037] The inter-machine card processing machine 10 accepts deposits, dispenses game balls, communicates with the management device 50, and communicates with the game machines 20. When the inter-machine card processing machine 10 receives banknotes inserted by a player, it sends a deposit notification including the deposit amount to the management device 50, adding the equivalent prepaid value to the prepaid value managed by the management device 50. Then, when a predetermined ball dispensing operation is performed, it sends a ball dispensing request to the management device 50, deducting the prepaid value managed by the management device 50, notifying the game machine 20 of the number corresponding to the deducted prepaid value, and adding it to the number of game balls.

[0038] When the inter-machine card processing machine 10 receives a card insertion notification, it sends a card insertion notification to the management device 50. Furthermore, when the inter-machine card processing machine 10 receives the balance of prepaid value, held balls, or stored balls from the management device 50, it stores the balance. If it receives and stores the balance of held balls, it sends a ball deduction request to the management device 50, thereby clearing the balance (number of balls) of held balls managed by the management device 50 to zero.

[0039] When the inter-machine card processing unit 10 receives a request to replay held balls, it subtracts a predetermined number of balls from the number of balls it manages, notifies the gaming machine 20 of the number of balls subtracted, and adds it to the number of game balls. Also, when the inter-machine card processing unit 10 receives a request to replay stored balls, it sends a request to the management device 50 to replay stored balls, causing the management device 50 to subtract a predetermined number of stored balls, notifies the gaming machine 20 of the number of balls subtracted, and adds it to the number of game balls.

[0040] When the inter-machine card processing machine 10 receives a card return operation, it sends a request to the management device 50 to add the number of balls held, and after the management device 50 adds the number of balls held, it sends a card ejection notification to the management device 50 and controls the ejection of the card. If a counting operation is received from the gaming machine 20 before the card return operation, the number of game balls is added to the number of balls held.

[0041] The inter-machine card processing unit 10 has the function of obtaining a game machine ID from the game machine 20 and authenticating the game machine 20 using the obtained game machine ID and the authentication key received from the management device 50. This authentication is performed during opening procedures, etc., and the game machine 20 becomes playable only if the authentication is successful.

[0042] The island controller 30 is a device that bundles a group of gaming machines 20 and inter-machine card processing machines 10 installed on a gaming island, and relays various information to the management device 50.

[0043] When the management device 50 receives a card insertion notification from the inter-machine card processing machine 10, it manages the identification information of the inserted card (hereinafter referred to as "card ID") in association with the inter-machine card processing machine 10, and transmits the prepaid value and balance of tokens associated with the card ID to the inter-machine card processing machine 10. Furthermore, if the card ID indicated in the card insertion notification is the card ID of a member card, it transmits the token replay data to the inter-machine card processing machine 10.

[0044] Furthermore, when the management device 50 receives a request to deduct balls from the inter-machine card processing machine 10, it clears the balance of balls to zero, and when it receives a request to add balls from the inter-machine card processing machine 10, it adds the number of balls indicated in the ball addition request to the balance of balls.

[0045] Furthermore, when the management device 50 receives a ball dispensing request from the inter-machine card processing machine 10, it subtracts a predetermined value from the prepaid value associated with the card ID and transmits permission to dispense balls to the inter-machine card processing machine 10. When it receives a request to replay stored balls, it transmits stored ball replay data to the inter-machine card processing machine 10.

[0046] Furthermore, if the management device 50 receives a card ID from the prize management device 60, it notifies the prize management device 60 of the number of balls associated with this card ID. In addition, if it receives a card ID from the payment machine 70, it notifies the payment machine 70 of the prepaid value associated with this card ID.

[0047] Furthermore, the management device 50 obtains the gaming machine ID from the inter-machine card processing machine 10 and manages the gaming machines 20 installed in its own store. It then obtains an authentication key for authenticating the gaming machines 20 set up in its own store from an authentication key management center (not shown) outside the gaming store and distributes it to the inter-machine card processing machine 10.

[0048] Furthermore, the management device 50 manages member management data of members registered with the amusement parlor. Specifically, it manages the number of tokens stored, points, PIN, and name, etc., in association with the member card ID issued to the member.

[0049] When the management device 50 receives a card insertion notification from the inter-machine card processing machine 10, it transmits to the inter-machine card processing machine 10 stored ball replay data, including the balance of stored balls and the PIN corresponding to the card ID indicated in the card insertion notification. Also, when the management device 50 receives a stored ball replay request from the inter-machine card processing machine 10, it subtracts a predetermined number of stored balls associated with the card ID indicated in the stored ball replay request, and transmits to the inter-machine card processing machine 10 stored ball replay data, including the remaining stored ball balance after the subtraction.

[0050] Furthermore, if the management device 50 receives an inquiry from the prize management device 60 regarding the number of stored tokens, it notifies the prize management device 60 of the balance of stored tokens corresponding to the specified card ID.

[0051] Furthermore, when the management device 50 receives out data, safe data, and special prize data from the gaming machine 20 from the inter-machine card processing machine 10, it stores them as operational data associated with the time and gaming machine ID, and uses this operational data to generate a chronological history of the gaming machine 20's operation and stores it as game history data.

[0052] Furthermore, the management device 50 stores in the game-related card processing history data the operation history of the inter-machine card processing machines 10, such as deposits, card insertions, and card ejections, as well as the operation history of the prize management device 60, the settlement machine 70, etc.

[0053] Furthermore, when the management device 50 receives wagon prize exchange data and ball distribution notification data from the inter-machine card processing machine 10, it associates the time, game machine ID, card ID, etc. with the content of the processing and stores it in the non-game card processing history data.

[0054] Furthermore, when the management device 50 receives notification of operation from the prize management device 60, the settlement machine 70, etc., it associates the time, device ID, the game machine ID of the last used game machine 20, the card ID, etc. with the details of the processing and stores them in the counting settlement processing history data.

[0055] Furthermore, the management device 50 generates and stores game history data using operational data and game-related card processing history data at predetermined times, such as after closing time.

[0056] Furthermore, the management device 50 generates and stores gameplay data indicating a single continuous game at predetermined timings, such as after closing time, using the non-game card processing history data, the counting and settlement processing history data, and the game history data.

[0057] Furthermore, once the operational data is generated / updated, the management device 50 uses the game machine ID and the number of balls played in this operational data to calculate the average output for each machine model for each week after its introduction, and stores this in the operational performance data. At this time, it calculates and stores the data for the first to tenth weeks after the introduction of the game machine.

[0058] Furthermore, if the management device 50 updates the operational performance data, it notifies the information management device 80 of this operational performance data.

[0059] Furthermore, if the management device 50 receives an analysis instruction regarding the first-week performance ratio and the first-week contribution, it notifies the information management device 80 of the analysis request. If the management device 50 receives the first-week performance ratio data and the first-week contribution data from the information management device 80, it displays the received data.

[0060] The prize management device 60 is a terminal device for prize exchange located next to the prize exchange counter inside the amusement store, and it processes the exchange of acquired balls, stored balls, and held balls for prizes. The prize management device 60 is connected to a card reader that reads the card ID of a card and a prize dispensing device that dispenses prizes. When the prize management device 60 reads a card ID from a general card or a membership card, it transmits the card ID to the management device 50 and requests the balance of balls held for that card ID. Also, when exchanging stored balls for prizes, it requests the balance of stored balls from the management device 50.

[0061] When a card with associated prepaid value is inserted into the payment machine 70, it transmits the card ID of the card to the management device 50, retrieves the prepaid value associated with the card, and dispenses cash equivalent to the retrieved prepaid value.

[0062] The information management device 80 predicts the national average and the average output for each store for the installed models, and calculates the first-week performance ratio and the first-week contribution. When the information management device 80 receives information about the models of gaming machines, it stores this information in the model data. When the information management device 80 receives operational performance data from the management device 50, it stores the data in the model-specific performance data, associating it with the store ID for each model, and also stores it in the type-specific performance data, categorized by model type.

[0063] Furthermore, the information management device 80 uses model data and model-specific performance data to train a national average out-of-pocket prediction model that predicts the average out-of-pocket value for the target model nationwide, and a store average out-of-pocket prediction model that predicts the average out-of-pocket value for each store, thereby generating a national average out-of-pocket prediction model and a store average out-of-pocket prediction model.

[0064] Furthermore, if the machine data is updated, the information management device 80 uses the parameters related to the updated machine to perform simulations regarding the number of balls dispensed nationwide and for each store. From the results of these simulations, the information management device 80 calculates the average number of balls dispensed and predetermined percentile values ​​and stores them in percentile data.

[0065] Furthermore, the information management device 80 inputs data related to the characteristics of the target models extracted from percentile data and model data into the national average out-of-pocket prediction model and the store average out-of-pocket prediction model, thereby outputting predicted values ​​for the national average and the average out-of-pocket for each store, and storing them in the prediction data.

[0066] Furthermore, if the forecast data is updated, the information management device 80 calculates the first-week performance ratio using the first-week average forecast value of the forecast data and the first-week average value of the corresponding machine type in the type-specific performance data, and stores it in the first-week performance ratio data. If the first-week performance ratio data is updated, the information management device 80 calculates the first-week contribution using the performance ratio for each store and the national average performance ratio in the first-week performance ratio data, and stores it in the first-week contribution data.

[0067] Furthermore, if the information management device 80 receives an analysis request from the management device 50, it notifies the management device 50 of the first-week performance ratio and first-week contribution for the store of the management device 50, as well as the first-week performance ratio for the national average.

[0068] Next, the ball dispensing process in the game information provision system according to this embodiment 1 will be described. When a player performs a ball dispensing operation on the inter-machine card processing machine 10, the inter-machine card processing machine 10 sends a message addressed to the local address of the management device 50. This message includes the card ID of the card inserted in the inter-machine card processing machine 10, an address or arbitrary identification information that identifies the inter-machine card processing machine 10 as the sender, and a ball dispensing request.

[0069] When the management device 50 receives a message containing a request to borrow balls, it updates the card management data by subtracting a predetermined value (for example, 500 units, representing 500 yen if 1 unit is 1 yen) from the prepaid value associated with the card ID in the message, and sends a ball lending permission to the inter-machine card processing machine 10 that sent the message. Upon receiving the ball lending permission, the inter-machine card processing machine 10 sends an addition signal to the gaming machine 20 requesting the addition of a number (for example, "125") corresponding to the subtracted prepaid value to the number of gaming balls.

[0070] Upon receiving the addition signal, the gaming machine 20 adds the number of game balls and transmits the added number of game balls to the inter-machine card processing machine 10. The inter-machine card processing machine 10 compares the sum of the number of game balls before addition and the number of balls dispensed with the added number of game balls to determine whether the addition of game balls was performed correctly.

[0071] Next, the process for replaying held tokens in the game information provision system according to this embodiment 1 will be described. When the inter-machine card processing machine 10 receives the insertion of a general card or a member card, it sends a message addressed to the local address of the management device 50. This message includes the card ID of the card inserted into the inter-machine card processing machine 10, an address or arbitrary identification information that identifies the inter-machine card processing machine 10 as the sender, information for identifying the rate of the inter-machine card processing machine 10, and a card insertion notification.

[0072] When the management device 50 receives a card insertion notification from the inter-machine card processing machine 10, it notifies the inter-machine card processing machine 10 of the remaining balance of balls held at the rate specified in the message, from among the balls held that are associated with the card ID in the message.

[0073] The inter-machine card processing machine 10 stores the remaining balance of the player's balls received from the management device 50. It then sends a message to the management device 50 containing information identifying the card ID and rate, as well as a request to deduct the remaining balls. If the management device 50 receives a request to deduct the remaining balls, it clears the remaining balance of the balls identified by the card ID and rate to zero. Furthermore, if the inter-machine card processing machine 10 receives a request to replay the remaining balls, it deducts a certain number of balls from the remaining balance, notifies the gaming machine 20 of the corresponding number, and has it added to the number of game balls.

[0074] When the inter-machine card processing machine 10 receives a management transfer request from the gaming machine 20 to transfer the number of game balls in the gaming machine 20 to the inter-machine card processing machine 10's holdings, the inter-machine card processing machine 10 adds the number of game balls indicated in the management transfer request to its own holdings. Subsequently, upon receiving a card return operation, it sends a message to the management device 50 including a request to add to the holdings. This message includes the card ID of the card to be ejected, an address or arbitrary identification information that identifies the inter-machine card processing machine 10 as the sender, information to identify the rate of the inter-machine card processing machine 10, the remaining balance of the holdings, and the request to add to the holdings.

[0075] If the management device 50 receives a request to add to the number of balls held, it updates the balance (number of balls held) of the balls associated with the card ID in the message for the rate specified in the message with the received value. Subsequently, the inter-machine card processing machine 10 sends a card ejection notification to the management device 50 and controls the ejection of the card.

[0076] Next, the process for replaying stored tokens in the game information provision system according to this embodiment 1 will be described. When the inter-machine card processing machine 10 receives a member card, it sends a message addressed to the local address of the management device 50. This message includes the card ID of the card inserted into the inter-machine card processing machine 10, an address or arbitrary identification information that identifies the inter-machine card processing machine 10 as the sender, information to identify the rate of the inter-machine card processing machine 10, and a card insertion notification.

[0077] The management device 50 transmits to the inter-machine card processing machine 10 stored ball replay data, which includes the PIN associated with the card ID in the message and the balance of stored balls at the rate specified in the message.

[0078] The inter-machine card processing machine 10 stores the received stored ball replay data, and if the balance of stored balls indicated in the stored ball replay data is equal to or greater than the number of game balls dispensed in stored ball replay (stored ball replay unit number; for example, "125 balls"), it accepts the stored ball replay operation.

[0079] After storing the stored ball replay data, the inter-machine card processing machine 10, upon receiving the first stored ball replay operation, prompts the player to enter a PIN and verifies whether the entered PIN matches the PIN shown in the stored ball replay data.

[0080] The inter-machine card processing machine 10 sends a message to the local address of the management device 50 when the PIN matches. This message includes the card ID of the card inserted in the inter-machine card processing machine 10, an address or arbitrary identification information that identifies the sending inter-machine card processing machine 10, information to identify the rate of the inter-machine card processing machine 10, and a request for replay of stored tokens.

[0081] The management device 50 subtracts a predetermined amount from the balance of stored balls associated with the card ID in the message, specifically those at the rate specified in the message, and transmits the stored ball replay data, including the remaining balance, to the inter-machine card processing machine 10.

[0082] The inter-machine card processing unit 10 receives the stored ball replay data, updates the stored ball balance, and notifies the gaming machine 20 of the number corresponding to the stored ball replay unit, adding it to the number of game balls. It also determines whether the updated stored ball balance is less than the number of stored ball replay units. If the updated stored ball data balance is equal to or greater than the number of stored ball replay units, it can accept another stored ball replay operation, and if it accepts the stored ball replay operation, it sends a stored ball replay request. For the second and subsequent stored ball replay operations, verification of the PIN is not required. If the updated stored ball data balance is less than the number of stored ball replay units, it will not accept a stored ball replay operation.

[0083] Next, we will explain the transfer from held balls to stored balls during the closing process. The management device 50 checks the balance of held balls in the card management data during the closing process, and if there are held balls with a balance greater than "0", it adds the balance of those held balls to the stored ball balance and updates it. After that, the management device 50 clears the balance of those held balls to zero.

[0084] <External configuration of the inter-machine card processing unit 10 and the gaming machine 20> Next, the external configuration of the inter-machine card processing machine 10 and gaming machine 20 shown in Figure 2 will be described. Figure 3 is a diagram showing the external configuration of the inter-machine card processing machine 10 and gaming machine 20 shown in Figure 2. The diagram shows an inter-machine card processing machine 10 that accepts only banknotes, but a unit that can accept electronic money can also be provided.

[0085] As shown in Figure 3, the inter-machine card processing machine 10 is equipped with a status display unit 11 that indicates the status of the inter-machine card processing machine 10 by the illumination or flashing of a lamp of a predetermined color, a banknote slot 12a that accepts various banknotes, a display operation unit 13 such as a touch panel display, a card slot 14a that accepts cards associated with a card ID, and a tapping unit 14b that accepts a mobile terminal held by a player in order to associate a unique ID. The tapping unit 14b also accepts a mobile terminal associated with a unique ID and reads the unique ID.

[0086] The gaming machine 20 is equipped with a game board, a handle 21 used to dispense the enclosed game balls onto the game board, and a counting button 22.

[0087] The counting button 22 is an operation button for transferring the management of the number of playable balls from the gaming machine 20 to the inter-machine card processing machine 10 by subtracting the number of playable balls from the gaming machine 20 and adding the corresponding number to the number of playable balls held by the inter-machine card processing machine 10. In the case of an open-type gaming machine 20 where physical playable balls are dispensed upon winning, this management transfer corresponds to the counting process in which the inter-machine card processing machine 10 counts the dispensed playable balls and adds them to the number of playable balls held.

[0088] <Configuration of the inter-unit card processing machine 10> Next, the configuration of the inter-machine card processing machine 10 shown in Figure 2 will be described. Figure 4 is a functional block diagram showing the configuration of the inter-machine card processing machine 10 shown in Figure 2. As shown in Figure 4, the inter-machine card processing machine 10 has a banknote transport unit 12, a display operation unit 13, a reader / writer 14, a communication unit 15, a storage unit 16, a control unit 17, and a game management unit 18.

[0089] The banknote transport unit 12 is a transport unit that transports banknotes inserted through the banknote insertion slot 12a to a banknote storage unit (not shown) while determining the denomination and authenticity of the banknotes. The display and operation unit 13 is an input / output device such as a touch panel display that displays various information such as monetary value and accepts various operations such as ball dispensing operations.

[0090] The reader / writer 14 is a reading unit that reads the card ID from the card inserted into the card slot 14a and the unique ID from the mobile terminal held over the tapping unit 14b. The card inserted into the card slot 14a is stored in a card storage unit (not shown) via this reader / writer 14. The communication unit 15 is an interface unit for data communication with the management device 50 via the island controller 30.

[0091] The storage unit 16 is a storage device consisting of a hard disk drive, non-volatile memory, etc. The storage unit 16 stores its own device status data 16a and card data 16b.

[0092] The self-device status data 16a is data indicating the status of the inter-machine card processing machine 10. This self-device status data 16a includes the inter-machine card processing machine ID, gaming machine ID, game type, etc. The inter-machine card processing machine ID is identification information for uniquely identifying the inter-machine card processing machine 10 within the gaming establishment. The game type is data indicating the rate set in the inter-machine card processing machine 10. When handling gaming balls of multiple rates within the gaming establishment, a game type name is set and managed for each rate, such as "Ball 1" for 4-yen rate gaming balls, "Ball 2" for 2-yen rate gaming balls, and "Ball 3" for 1-yen rate gaming balls. The inter-machine card processing machine 10 selects the rate to be used for the game from these game types and stores it as the game type. The game type may be fixed for each section in which the inter-machine card processing machine 10 is installed, or it may be changeable by the player.

[0093] Card data 16b is data relating to the card being used by the player. Card data 16b includes card ID, PIN, prepaid value, balls held, stored balls, etc. The card ID is the card ID read by the reader / writer 14. When a card is transported from a card storage unit (not shown) to the card slot 14a and then ejected, card data 16b is updated with the card ID read by the reader / writer 14 during this transport. Similarly, when a card is transported from the card slot 14a to a card storage unit (not shown), card data 16b is updated with the card ID read by the reader / writer 14 during this transport. Prepaid value indicates the remaining balance of prepaid value usable for ball dispensing, and balls held and stored balls indicate their respective balances.

[0094] The game management unit 18 is a processing unit that performs authentication and communication with the gaming machine 20. It is desirable that this game management unit 18 be formed on a different circuit board from the control unit 17. The game management unit 18 includes an authentication processing unit 18a, an authentication key management unit 18b, and a gaming machine state management unit 18c. In practice, these programs are loaded into the CPU (Central Processing Unit) and executed, causing the authentication processing unit 18a, the authentication key management unit 18b, and the gaming machine state management unit 18c to execute the processes corresponding to them, respectively.

[0095] The authentication processing unit 18a performs authentication of the gaming machine 20 using the authentication key received from the management device 50. If authentication is successful, the authentication processing unit 18a sends permission to operate to the gaming machine 20. If authentication fails, the authentication processing unit 18a prohibits the use of the gaming machine 20. However, even if authentication fails, the system may be configured to send permission to operate to the gaming machine 20 if it is within a predetermined period since the last successful authentication, and not send permission to operate to the gaming machine 20 after the predetermined period has elapsed since the last successful authentication, thereby prohibiting the use of the gaming machine 20.

[0096] The authentication key management unit 18b is a processing unit that manages authentication keys. Each authentication key has a set usage time, and when the cumulative operating time of the gaming machine 20 reaches the usage time of the authentication key, the authentication key becomes unusable. Specifically, the authentication key management unit 18b uses the power-on time of the circuit board on which the gaming management unit 18 is formed as the operating time of the gaming machine 20 and manages the usage time of the authentication key. If an authentication key becomes unusable, the authentication key management unit 18b obtains a new authentication key from an authentication key management center outside the store via the management device 50 and updates the authentication key.

[0097] The gaming machine status management unit 18c is a processing unit that communicates with the gaming machine 20 and manages the status of the gaming machine 20. Specifically, the gaming machine status management unit 18c obtains the gaming machine ID from the gaming machine 20, sends a request to add to the number of game balls when a ball is dispensed, sends a request to switch management from held balls to the number of game balls when a held ball is replayed, sends a request to add to the number of game balls when a stored ball is replayed, receives game results such as the number of balls played and the number of prize balls dispensed from the gaming machine 20, and receives a request to switch management from the number of game balls to held balls when counting. Communication between the inter-machine card processing unit 10 and the gaming machine 20 uses encrypted communication with a predetermined encryption method.

[0098] Furthermore, the gaming machine state management unit 18c acquires the current status of the gaming machine 20, such as the number of game balls, start status, jackpot status, jackpot type, and number of draws. It is desirable that the time interval at which the gaming machine state management unit 18c acquires the status of the gaming machine 20 be set shorter than the interval at which game balls are launched (used) in the gaming machine 20. Also, once the gaming machine state management unit 18c has acquired the number of game balls from the gaming machine 20, it hands over the number of game balls to the control unit 17.

[0099] Furthermore, the timing for the gaming machine status management unit 18c to acquire a new gaming machine ID from the gaming machine 20 is not limited to before opening. The gaming machine ID can be acquired or the determination of whether to replace the gaming machine 20 can be made when the inter-machine card processing unit 10 is powered on or off, during periodic communication, when processing related to the number of game balls (ball dispensing, replay, counting, etc.) is performed, or when the system returns from an offline state. These timings may also be used in combination.

[0100] The control unit 17 is a control unit that provides overall control for the inter-unit card processing machine 10 and includes a data management unit 17a. In practice, the process is executed by loading a program corresponding to the data management unit 17a into the CPU and executing it.

[0101] When a card is inserted, the data management unit 17a sends a card insertion notification to the management device 50, which includes the card ID and the inter-machine card processing machine ID. Furthermore, when the data management unit 17a receives data from the management device 50 that includes a monetary value (at least one of prepaid value, held balls, and stored balls), it updates the card data 16b with the received monetary value.

[0102] Furthermore, when a banknote is inserted into the banknote slot 12a, the data management unit 17a sends a deposit notification to the management device 50, which includes the amount of the inserted banknote, the card ID, and the inter-machine card processing machine ID.

[0103] Furthermore, when a player performs a ball dispensing operation, the data management unit 17a transmits a ball dispensing request, including the card ID and the inter-machine card processing machine ID, to the management device 50. Upon receiving permission to dispensing balls in response to this request, the data management unit 17a instructs the game management unit 18 to subtract the prepaid value of the card data 16b and add a predetermined number of balls to the number of game balls in the game machine 20.

[0104] Furthermore, if the data management unit 17a receives a request to replay the balls held, it instructs the game management unit 18 to subtract the number of balls held in the card data 16b and update it, and to add the corresponding number of balls to the number of game balls in the game machine 20.

[0105] Furthermore, if the balance of stored balls shown in the card data 16b is equal to or greater than the number of stored ball replay units, the data management unit 17a displays a stored ball replay button on the display operation unit 13, making it possible to accept a stored ball replay operation. When a player performs a stored ball replay operation, the data management unit 17a transmits a stored ball replay request, including the card ID and the inter-machine card processing machine ID, to the management device 50. Upon receiving stored ball replay data in response to this stored ball replay request, the data management unit 17a updates the balance of stored balls in the card data 16b and instructs the game management unit 18 to add the number of balls corresponding to the number of stored ball replay units to the number of game balls in the game machine 20. Note that for the first stored ball replay operation, the player is required to enter a PIN, and the condition for transmitting the stored ball replay request is that the PIN matches the one shown in the card data 16b.

[0106] Furthermore, when the game management unit 18 receives a request from the game machine 20 to transfer game balls, the data management unit 17a adds the number of balls instructed by the game management unit 18 to the number of balls held in the card data 16b and updates it.

[0107] Furthermore, when the data management unit 17a receives a card return operation, it sends a request to the management device 50 to add to the remaining balls, including the card ID, the inter-machine card processing machine ID, and the remaining ball balance. After clearing the card data 16b, it sends a card ejection notification to the management device 50 and ejects the card.

[0108] Next, an example of data stored in the storage unit 16 of the inter-unit card processing machine 10 shown in Figure 4 will be described. Figure 5 shows an example of the self-device status data 16a and card data 16b shown in Figure 4.

[0109] The self-device status data 16a shown in Figure 5(a) indicates that the ID of the inter-machine card processing machine 10 is "A101", the game machine ID of the connected game machine 20 is "B201", and "Ball 1" is set as the game type for the inter-machine card processing machine 10.

[0110] The card data 16b shown in Figure 5(b) indicates that the ID of the card inserted into the inter-machine card processing machine 10 is "2001". Here, the first digit of the card ID indicates the type of card; a card with the first digit "1" is a general card, and a card with the first digit "2" is a member card. The card data 16b also indicates that the PIN for replaying stored tokens is "1234" and the prepaid value is "3000" units.

[0111] Furthermore, card data 16b indicates the following state: as held balls, Ball 1 has a balance of "1500" balls, Ball 2 has a balance of "2500" balls, and Ball 3 has a balance of "0" balls; as stored balls, Ball 1 has a stored ball balance of "500" balls, Ball 2 has a stored ball balance of "0" balls, and Ball 3 has a stored ball balance of "0" balls.

[0112] <Configuration of the gaming machine 20> Next, the configuration of the gaming machine 20 shown in Figure 2 will be described. Figure 6 is a functional block diagram showing the configuration of the gaming machine 20 shown in Figure 2. As shown in Figure 6, the gaming machine 20 has a communication control unit 23, a performance control unit 24, a game control unit 25, and a game ball control unit 26. The communication control unit 23 is a control unit for controlling data communication with the inter-machine card processing machine 10. Communication with the inter-machine card processing machine 10 uses encrypted communication using a predetermined encryption method.

[0113] When the gaming machine 20 is started, the communication control unit 23 reads identification information from the control CPU provided in the gaming control unit 25 and / or the gaming ball control unit 26, checks whether the identification information is appropriate, and if it is appropriate, establishes communication with the inter-machine card processing unit 10 and enters a standby state. In the standby state, if it receives permission to operate from the inter-machine card processing unit 10, it starts the performance control unit 24, the gaming control unit 25 and the gaming ball control unit 26, making the machine ready for play.

[0114] The game control unit 25 is a control unit that controls the game played by the gaming machine 20. In practice, the process is executed by loading a program corresponding to the game control unit 25 into the CPU and executing it.

[0115] Specifically, the game control unit 25 controls the launch of game balls onto the game board based on handle operation detection, detects game balls that have entered the winning slots on the game board, acquires and draws random numbers (within a predetermined range of 0 to 65535, each number is assigned to a big win, small win, or loss) based on game balls that have entered the start slot (special symbols and regular symbols), controls movable members (such as tulips) on the game board, controls the display of special symbol display devices on the game board, and detects abnormalities that may indicate fraud (such as the front frame being open or vibration being detected) and notifies the higher-level device.

[0116] The game board is fitted with numerous obstacle pins, and when the game ball is launched onto the game board by operating the handle, it falls between the obstacle pins and either enters a prize entry hole or a start hole, or is ejected from the game board through an out hole without entering a prize. If the game ball enters the start hole, a lottery is held, and if a jackpot is won, a jackpot game is played in which the designated prize entry holes on the game board are controlled to open multiple times, making it easier for the game ball to enter these entry holes, and thus awarding prize balls to the player.

[0117] Furthermore, there are bonus jackpot games that grant players a predetermined bonus game after the jackpot game ends. Bonus jackpot games include probability variation jackpots and time-saving jackpots. Probability variation jackpots and time-saving jackpots grant probability variation games and time-saving games as bonus games after the jackpot, respectively. Time-saving games are bonus games that increase the number of draws per unit time and increase the probability of winning a jackpot per unit time by shortening the variation time (time from the start of the draw process to the display of the result) of the normal symbol draw (the draw for opening and closing the movable parts as described above) and / or special symbol draw (the draw for winning a jackpot by winning the game ball as described above). These bonus games may include normal symbol probability variation, which increases the probability of winning a normal symbol. Time-saving games due to time-saving jackpots end when the number of draws for special symbol draws has been performed a predetermined number of times after the jackpot. Probability-based games will be discussed later.

[0118] The game control unit 25 has a probability variation function. The probability variation function is a function that controls the random number range used for the next draw (generally by about 10 times) when the draw result falls within a particularly defined random number range for a jackpot. Probability variation game is a special symbol draw that uses this changed probability. Probability variation game continues until the next jackpot is won. In addition, probability variation game and time-saving game may be added at the same time. However, in order to prevent unforeseen losses to the amusement parlor, if the player acquires a certain amount of game media after the start of probability variation game or time-saving game, the game may be stopped (the card is ejected once, and the player is requested to exchange for a prize or move to another machine). In this case, the number of game media after entering the probability variation game state is counted by the game machine 20, and when this counted value reaches a predetermined value, game stop control is performed, such as stopping the launch of game balls.

[0119] The performance control unit 24 is a control unit that controls the performance during gameplay, and includes a performance symbol lottery unit 24a and a performance lottery unit 24b. In practice, by loading these programs into the CPU and executing them, the performance symbol lottery unit 24a and the performance lottery unit 24b are made to execute the processes corresponding to them, respectively.

[0120] The performance symbol lottery unit 24a performs a lottery for performance symbols (symbols to be displayed on the performance symbol display device, such as a display device on the game board) based on the special symbol lottery in the game control unit 25 (specifically, whether to stop the display at 7, 7, 7, etc.).

[0121] The performance lottery unit 24b performs a lottery to determine the type of performance to be performed before displaying the lottery result based on the performance symbols (such as a character appearing and a reach performance). The performances selected are set to differ depending on the main lottery result.

[0122] The performance control unit 24 stores performance data for each performance. Furthermore, the performance control unit 24 controls the display of the performance symbol display device, and when a performance is executed, it overlays the performance data on the background symbol data to display the performance. In addition, during probability fluctuations, it sets background color data different from the normal state.

[0123] The game ball control unit 26 is a control unit that controls the number of game balls. In practice, the process is executed by loading a program corresponding to the game ball control unit 26 into the CPU and executing it. The playable number management unit 26a is a management unit that manages the playable number, i.e., the number of game balls.

[0124] Specifically, the game ball control unit 26 stores the number of game balls in the playable number memory, deducts "1" from the number of game balls each time a game ball is launched, and adds the number of prize balls obtained through winning to the number of game balls. In addition, when the number of balls is notified by the inter-machine card processing machine 10, the notified number of balls is added to the number of game balls and a confirmation signal is sent to the inter-machine card processing machine 10 indicating that the number of balls has been added to the number of game balls. Furthermore, the game ball control unit 26 can also perform cleaning control of the game balls (circulating balls) sealed in the game machine 20.

[0125] Furthermore, the game ball control unit 26 periodically transmits the number of game balls stored in the playable number memory to the inter-machine card processing unit 10. In this case, the game ball control unit 26 transmits the number of game balls at the time of data acquisition to the inter-machine card processing unit 10 while continuing to update the number of game balls according to the game status, without resetting the contents of the playable number memory to zero.

[0126] Furthermore, the gaming machine 20 is equipped with a counting button 22, and a signal indicating that this counting button 22 has been pressed is input to the game ball control unit 26. The game ball control unit 26 detects the pressing operation of the counting button 22, and if the time from the rising edge to the falling edge of the detection signal is less than a certain period of time, it sends a management transfer request to the inter-machine card processing machine 10 along with data on the number of game balls in fixed increments (for example, 100 balls per operation). The number of balls that have been transferred to the management of the inter-machine card processing machine 10 is then subtracted (invalidated) from the game ball count. In addition, if the pressing operation of the counting button 22 is detected and no falling edge is detected for a predetermined time or longer after the detection signal rises, it sends a management transfer request to the inter-machine card processing machine 10 along with data on the number of game balls in fixed increments (for example, 200 balls every 5 seconds) for each period of time that the detection signal remains in a rising state. The number of balls that have been transferred to the management of the inter-machine card processing machine 10 is then subtracted (invalidated) from the game ball count. When subtracting from the number of game balls, possible methods include decrementing the value to be subtracted from a value of 1 stored in memory, or updating the value stored in memory with the value after the subtraction and deleting or logically deleting the value before the subtraction to invalidate it. However, there are no particular limitations as long as the value after the subtraction can be identified.

[0127] The game control unit 25 and the game ball control unit 26 are each configured on separate circuit boards. The game control board on which the game control unit 25 is configured and the game ball control board on which the game ball control unit 26 is configured each have a unique ID. The gaming machine 20 combines the ID of the game control board and the ID of the game ball control board and uses them as the gaming machine ID.

[0128] <Configuration of the control device 50> Next, the configuration of the management device 50 shown in Figure 2 will be described. Figure 7 is a functional block diagram showing the configuration of the management device 50 shown in Figure 2. As shown in Figure 7, the management device 50 is connected to a display unit 51 and an input unit 52, and has an external network communication unit 53, a store network communication unit 54, a storage unit 55, and a control unit 56.

[0129] The display unit 51 is a display device such as an LCD panel or a display device. The input unit 52 is an input device such as a keyboard or mouse. The external network communication unit 53 is an interface unit for data communication with the information management device 80 and an authentication key management center (not shown) via an external network. The store network communication unit 54 is an interface unit for data communication with the island controller 30, prize management device 60 and settlement machine 70 within the gaming store via a communication line.

[0130] The memory unit 55 is a storage device such as a hard disk drive or non-volatile memory, and stores card management data 55a, device management data 55b, member management data 55c, operation data 55d, game-related card processing history data 55e, non-game card processing history data 55f, counting and settlement processing history data 55g, game history data 55h, idle data 55i, operation performance data 55j, first-week performance comparison data 55k, and first-week contribution data 55l.

[0131] Card management data 55a is data that associates the card ID with the balance of prepaid value, the balance of held tokens, etc. Device management data 55b is data related to the devices installed in the amusement parlor. This device management data 55b includes the inter-machine card processing machine ID, installation location, game machine number, game machine ID, and game machine 20 model data, etc. Member management data 55c is data that associates the card ID of the membership card issued to the member with the member's name, stored token information, etc.

[0132] Operational data 55d is data showing the operation history of the gaming machine 20. Game-related card processing history data 55e is data showing the operation history of the inter-machine card processing machine 10, prize management device 60, settlement machine 70, etc. Non-game card processing history data 55f is data showing the history of non-game activities such as wagon prize exchange.

[0133] The counting and settlement processing history data 55g is data that shows the history of counting and settlement processing. The game history data 55h is data that shows the history of gameplay, and the status of the game, such as deposit, card insertion, card ejection, number of balls played, number of balls awarded, and occurrence of special prizes, is shown in chronological order.

[0134] The 55i play data represents a single continuous game session, associating the start time, end time, machine ID, card ID, customer type, deposit amount, number of balls played (out), number of balls won (in), number of jackpots, number of wagon prize exchanges, number of ball splits, and win / loss amounts for the same gaming machine. Here, the customer type indicates whether the player is a regular player or a member player.

[0135] The operational performance data 55j shows the average payout for each machine after its introduction, and includes the average payout for the first 10 weeks after the machine's introduction.

[0136] The first-week performance comparison data (55k) shows the first-week performance comparison of new models and is divided into national average and local store data. The first-week contribution data (55l) shows the first-week contribution of new models.

[0137] The control unit 56 is a control unit that performs overall control of the management device 50 and includes a card management unit 56a, a device management unit 56b, a member management unit 56c, an operation data storage and aggregation unit 56d, a card processing history management unit 56e, a game history data generation unit 56f, a mobile data generation unit 56g, an operation performance management unit 56h, and an analysis unit 56i. In practice, by loading these programs into the CPU and executing them, the processes corresponding to the card management unit 56a, the device management unit 56b, the member management unit 56c, the operation data storage and aggregation unit 56d, the card processing history management unit 56e, the game history data generation unit 56f, the mobile data generation unit 56g, the operation performance management unit 56h, and the analysis unit 56i are executed, respectively.

[0138] The card management unit 56a is a processing unit that manages the card management data 55a. The card management unit 56a communicates with the inter-machine card processing machine 10, the prize management device 60, and the settlement machine 70 to update the prepaid value and balance of tokens associated with the card ID. In addition, when the card management unit 56a receives a card insertion notification from the inter-machine card processing machine 10, it notifies the card management unit 56a of the value and balance of tokens associated with the card.

[0139] The device management unit 56b is a processing unit that manages the device management data 55b. The device management unit 56b generates and updates the device management data 55b based on the inter-machine card processing unit ID, installation location, and gaming machine ID obtained from the inter-machine card processing unit 10.

[0140] The Member Management Unit 56c is a processing unit that manages the Member Management Data 55c. The Member Management Unit 56c communicates with the Prize Management Device 60 to update the balance of stored points associated with the card ID.

[0141] Here, the processing of stored balls for replay by the member management unit 56c will be explained. When the member management unit 56c receives stored ball replay request data from the inter-machine card processing machine 10, it deducts the amount of stored balls associated with the card ID in the stored ball replay request data that matches the payout rate, and recalculates the number of available stored balls for replay. Then, it transmits stored ball replay data, including the balance after the deduction and the number of available stored balls for replay, to the inter-machine card processing machine 10.

[0142] When the operational data storage and aggregation unit 56d receives out data, safe data, and special prize data from the gaming machine 20 from the inter-machine card processing unit 10, it registers the time, gaming machine ID, and data type in the operational data 55d.

[0143] Furthermore, the operational data storage and aggregation unit 56d is also a processing unit that acquires the operation history of the gaming machine 20 by aggregating data based on the operational data 55d. The operation history of the gaming machine 20 includes the number of balls played, the number of balls dispensed as prizes, and the occurrence of special prizes.

[0144] The card processing history management unit 56e is a processing unit that stores the operation history of the inter-unit card processing machine 10, such as deposits, card insertions, and card ejections, as well as the operation history of the prize management device 60, the settlement machine 70, and the like.

[0145] Specifically, when the card processing history management unit 56e receives card insertion notification data, card ejection notification data, deposit notification data, individual machine counting notification data, and general card depletion data from the inter-machine card processing machine 10, it associates the time, game machine ID, card ID, etc. with the processing details and adds them to the game-related card processing history data 55e.

[0146] Furthermore, when the card processing history management unit 56e receives wagon prize exchange data and ball distribution notification data from the inter-machine card processing machine 10, it associates the time, game machine ID, card ID, etc. with the processing details and adds them to the non-game card processing history data 55f.

[0147] Furthermore, when the card processing history management unit 56e receives notification of an operation from the prize management device 60, the settlement machine 70, etc., it associates the time, device ID, the game machine ID of the last used game machine 20, the card ID, etc. with the details of the processing and adds them to the counting settlement processing history data 55g.

[0148] The game history data generation unit 56f generates game history data 55h using the operation data 55d and game-related card processing history data 55e at a predetermined timing, such as after closing time, and stores it in the storage unit 55.

[0149] The mobile data generation unit 56g generates mobile data 55i representing a single continuous game using the non-game card processing history data 55f, the counting and settlement processing history data 55g, and the game history data 55h at a predetermined timing such as after closing time, and stores it in the storage unit 55.

[0150] The operational performance management unit 56h is a processing unit that manages operational performance data 55j. When idle data 55i is generated / updated, the operational performance management unit 56h uses the gaming machine ID and number of balls played in the idle data 55i to calculate the average output for each machine for each week after the introduction of the gaming machine, and stores it in the operational performance data 55j. At this time, it calculates and stores the data for the first to tenth weeks after the introduction of the gaming machine.

[0151] Furthermore, if the operational performance management unit 56h updates the operational performance data 55j, it notifies the information management device 80 of this operational performance data 55j.

[0152] The analysis unit 56i is a processing unit that requests and displays analysis related to the first-week performance ratio and the first-week contribution. When the analysis unit 56i receives an analysis instruction related to the first-week performance ratio and the first-week contribution from the input unit 52, it notifies the information management device 80 of the analysis request.

[0153] Furthermore, when the analysis unit 56i receives first-week performance comparison data and first-week contribution data from the information management device 80, it stores them in the first-week performance comparison data 55k and first-week contribution data 55l, and displays them on the display unit 51.

[0154] Next, an example of data stored in the storage unit 55 of the management device 50 shown in Figure 7 will be explained. Figures 8 to 11 show an example of the card management data 55a, device management data 55b, member management data 55c, operation data 55d, game-related card processing history data 55e, non-game card processing history data 55f, counting settlement processing history data 55g, game history data 55h, idle data 55i, operation performance data 55j, first-week performance comparison data 55k, and first-week contribution data 55l shown in Figure 7.

[0155] In the card management data 55a shown in Figure 8(a), card ID "1001" is associated with a prepaid value of "0" units, a balance of "0" balls for each rate, and a usage ID of "A101". In other words, the card with card ID "1001" is inserted into the inter-machine card processing machine 10 with device ID "A101", and the management of the balls held has been transferred to the inter-machine card processing machine 10. Therefore, the balance of balls held for each rate is zero.

[0156] Furthermore, in the card management data 55a, card ID "2001" is associated with a prepaid value of "3000" units, and the balance of balls held is "1500" for ball 1, "2500" for ball 2, and "0" for ball 3. However, the destination ID is not associated. In other words, the card with card ID "2001" is not inserted into the inter-machine card processing machine 10, and the management of balls is handled by the management device 50.

[0157] The device management data 55b shown in Figure 8(b) indicates that for the inter-machine card processing machine 10 with ID "A101", its installation location is "Island 1-1", its game machine number is "11", the ID of the connected game machine 20 is "B201", the game type is "Ball 1", the model of game machine 20 is "EV01", and its serial number is "a". It also indicates that this game machine 20 is in use.

[0158] Furthermore, the device management data 55b indicates that for the inter-machine card processing machine 10 with ID "A201", its installation location is "Island 2-1", its gaming machine number is "21", the ID of the connected gaming machine 20 is "B506", the game type is "Ball 3", the model of gaming machine 20 is "DX03", and its serial number is "g". It also indicates that this gaming machine 20 is in use.

[0159] The member management data 55c shown in Figure 8(c) indicates the name, stored balls, etc., associated with the card ID. Specifically, the member management data 55c shows that for card ID "2005", the name is "Taro Toka", the stored ball balance for "Ball 1" is "1500", the stored ball balance for "Ball 2" is "200", and the stored ball balance for "Ball 3" is "0".

[0160] The operational data 55d shown in Figure 9(a) associates the following states: the time "11:10:14" corresponds to the state where the game machine ID is "P001" and the data type is "Out"; the time "11:10:16" corresponds to the state where the game machine ID is "P001" and the data type is "Out"; the time "11:11:30" corresponds to the state where the game machine ID is "P001" and the data type is "Safe"; and the time "13:30:00" corresponds to the state where the game machine ID is "P015" and the data type is "Special Prize (Big Win)".

[0161] The game-related card processing history data 55e shown in Figure 9(b) is data that associates time, game machine ID, card ID, and processing. Specifically, the game-related card processing history data 55e indicates that at the time "11:10:00", a "card insertion" process was performed using card ID "1002" at the inter-machine card processing machine 10 attached to game machine 20 with game machine ID "P001".

[0162] Furthermore, the game-related card processing history data 55e indicates that at the time "12:35:30", a "card ejection" process was performed using card ID "2005" at the inter-machine card processing machine 10 attached to game machine 20 with game machine ID "P023".

[0163] Furthermore, the game-related card processing history data 55e indicates that at the time "13:20:20", a "deposit" transaction was performed using card ID "1055" at the inter-machine card processing machine 10 attached to game machine 20 with game machine ID "P015".

[0164] Furthermore, the game-related card processing history data 55e indicates that at the time "13:52:40", the inter-machine card processing machine 10, which is attached to game machine 20 with game machine ID "P007", performed the "individual machine counting" process using card ID "1021".

[0165] Furthermore, the game-related card processing history data 55e indicates that at the time "14:04:00", the inter-machine card processing machine 10, which is attached to game machine 20 with game machine ID "P015", processed a "general card use-up" for card ID "1055".

[0166] The non-game card processing history data 55f shown in Figure 9(c) is data that associates time, game machine ID, card ID, and processing. Specifically, the non-game card processing history data 55f indicates that at the time "13:40:01", a "wagon prize exchange" process was performed using card ID "1003" at the inter-machine card processing machine 10 attached to game machine 20 with game machine ID "P011".

[0167] Furthermore, the non-game card processing history data 55f indicates that at the time "15:00:50", a "ball splitting" process was performed using card ID "1002" at the inter-machine card processing machine 10, which is attached to game machine 20 with game machine ID "P007".

[0168] The counting and settlement processing history data 55g shown in Figure 10(a) is data that associates time, device ID, game machine ID, card ID, and processing. The device ID is identification information that uniquely identifies the prize management device 60, settlement machine 70, etc. The game machine ID is identification information that indicates the game machine 20 last used by the player, and is identified by having the player input it when performing prize exchange, settlement, counting, etc. Alternatively, it may be identified by referring to the card usage history held by the management device 50.

[0169] Specifically, the counting settlement processing history data 55g associates the time "14:22:33" with the state where the prize management device 60's device ID is "6001", the game machine ID is "P001", the card ID is "1002", and the processing is "prize exchange". It also associates the time "14:22:33" with the state where the prize management device 60's device ID is "6001", the game machine ID is "P001", the card ID is "2013", and the processing is "prize exchange". In other words, the general card with card ID "1002" and the membership card with card ID "2013" are being used for prize exchange simultaneously.

[0170] Furthermore, the counting and settlement processing history data 55g associates the time "16:08:08" with the device ID of settlement machine 70 being "7011", the game machine ID being "P019", the card ID being "1033", and the processing status being "settlement".

[0171] The game history data 55h shown in Figure 10(b) is data that associates time, game machine ID, card ID, and processing, and shows the state of the game in chronological order, such as deposit, card insertion, card ejection, number of balls played, number of prize balls dispensed, and occurrence of special prizes.

[0172] Specifically, in the game history data 55h, the time "11:10:00" is associated with a state where the game machine ID is "P001", the card ID is "1002", and the action is "card inserted". The time "11:11:20" is associated with a state where the game machine ID is "P001", the card ID is "1002", and the action is "number of balls inserted 100". The time "11:12:00" is associated with a state where the game machine ID is "P001", the card ID is "1002", and the action is "number of balls dispensed 30".

[0173] Furthermore, in the game history data 55h, the following states are associated: the time "12:35:30" corresponds to a state where the game machine ID is "P023", the card ID is "2005", and the process is "card ejection"; the time "13:20:20" corresponds to a state where the game machine ID is "P015", the card ID is "1055", and the process is "deposit"; and the time "13:30:00" corresponds to a state where the game machine ID is "P015", the card ID is "1055", and the process is "special prize (jackpot)".

[0174] The mobile data 55i shown in Figure 10(c) represents a single continuous game session, and is associated with the start time, end time, game machine ID, card ID, customer type, deposit amount, number of balls played (out), number of prizes won (in), number of jackpots, number of wagon prize exchanges, number of ball divisions, and win / loss amounts for the same game machine. Here, the customer type indicates whether the customer is a regular player or a member player.

[0175] Specifically, the mobile data 55i associates the following conditions: start time "11:10:00", end time "14:15:33", game machine ID "P001", card ID "2005", customer type "member", deposit amount "3000" yen, number of balls played (out) "0", number of prize balls dispensed (safe) "6500", number of jackpots "10", number of wagon prize exchanges "0", number of ball divisions "1", and total win / loss amount "100000" yen.

[0176] Furthermore, the mobile data 55i associates the following conditions: start time "13:20:20", end time "18:00:00", game machine ID "P015", card ID "1002", customer type "general", deposit amount "12000" yen, number of balls played (out) "13000", number of prize balls dispensed (safe) "10150", number of jackpots "13", number of wagon prize exchanges "2", number of ball divisions "0", and total win / loss amount "30000" yen.

[0177] Furthermore, the mobile data 55i associates the following conditions: start time "14:30:00", end time "17:00:00", game machine ID "P020", card ID "2005", customer type "member", deposit amount "15000" yen, number of balls played (out) "10000", number of prize balls dispensed (safe) "8000", number of jackpots "5", number of wagon prize exchanges "1", number of ball divisions "0", and win / loss amount "-10000" yen.

[0178] The operational performance data 55j shown in Figure 11(a) associates the average output of the "EV01" model with states where the average output for the first week was "37,000" balls, the average output for the second week was "35,000" balls, and the average output for the tenth week was "20,000" balls. For the "GW30" model, it associates the average output of the "GW30" model with states where the average output for the first week was "41,000" balls, the average output for the second week was "39,000" balls, and the average output for the tenth week was "10,000" balls.

[0179] The 55k data points for first-week performance ratios shown in Figure 11(b) correspond to the "EV02" model, where the national average performance ratio is "1.7" and the store's ratio is "2.0". For the "DX04" model, the national average performance ratio is "2.5" and the store's ratio is "2.2".

[0180] The first-week contribution data 55l shown in Figure 11(c) associates a contribution of "1.2" with the model "EV02" and a contribution of "0.9" with the model "DX04".

[0181] <Configuration of Information Management Device 80> Next, the configuration of the information management device 80 shown in Figure 2 will be described. Figure 12 is a functional block diagram showing the configuration of the information management device 80 shown in Figure 2. As shown in Figure 12, the information management device 80 is connected to a display unit 81 and an input unit 82, and has a communication unit 84, a storage unit 85, and a control unit 86.

[0182] The display unit 81 is a display device such as an LCD panel or a display device. The input unit 82 is an input device such as a keyboard or mouse. The communication unit 84 is an interface unit for data communication with the management device 50 via an external network.

[0183] The storage unit 85 is a storage device such as a hard disk drive or non-volatile memory, and stores model data 85a, model-specific performance data 85b, type-specific performance data 85c, percentile data 85d, forecast data 85e, first-week performance comparison data 85f, and first-week contribution data 85g.

[0184] Machine data 85a is data that shows information about the type of gaming machine, including the machine type, year of sale, parameters, machine type, and characteristics. Machine-specific performance data 85b is data that shows the average output for each machine after its introduction, including the machine type, machine type, store ID, and average output for the first to tenth weeks after introduction.

[0185] The performance data by type 85c shows the average output for each machine type after the introduction of the gaming machine, and includes the machine type, store ID, and the average output from the first to the tenth week after introduction. The percentile data 85d shows the percentile values ​​extracted from the simulation results performed using the parameters of the machine data.

[0186] Predictive data 85e shows the predicted average payout for each type of gaming machine after its introduction, and includes the machine type, store ID, and the predicted average payout for weeks 1 through 10 after introduction.

[0187] The first-week performance ratio data 85f shows the first-week performance ratio for each model. The first-week performance ratio data 85f shows the first-week performance ratio broken down into national averages and by store. The first-week contribution data 85g shows the first-week contribution for each model. The first-week contribution data 85g shows the first-week contribution broken down by store.

[0188] The control unit 86 is a control unit that performs overall control of the information management device 80, and includes a model data management unit 86a, an operation performance management unit 86b, a nationwide model training unit 86c, a store model training unit 86d, a prediction control unit 86e, a nationwide model prediction unit 86f, a store model prediction unit 86g, and an analysis unit 86h. In practice, by loading these programs into the CPU and executing them, the processes corresponding to the model data management unit 86a, the operation performance management unit 86b, the nationwide model training unit 86c, the store model training unit 86d, the prediction control unit 86e, the nationwide model prediction unit 86f, the store model prediction unit 86g, and the analysis unit 86h are executed, respectively.

[0189] The machine data management unit 86a is a processing unit that manages the machine data 85a. When the machine data management unit 86a receives information about the machine model from the input unit 82, it stores this information in the machine data 85a.

[0190] The operational performance management unit 86b is a processing unit that manages machine-specific performance data 85b and type-specific performance data 85c. When the operational performance management unit 86b receives operational performance data from the management device 50, it associates it with the store ID for each machine and stores it in the machine-specific performance data 85b. When the operational performance management unit 86b updates the machine-specific performance data 85b, it uses this machine-specific performance data 85b to calculate the average output for each machine type and stores it in the type-specific performance data 85c.

[0191] The National Model Training Unit 86c is a processing unit that trains the National Out-Average Prediction Model. Using the model data 85a and the performance data 85b for each model, the National Model Training Unit 86c trains the National Out-Average Prediction Model to predict the average out-of-run value for the target model nationwide. In this training, training is performed for each model, divided into weeks from week 1 to week 10 after introduction. Therefore, a National Out-Average Prediction Model is generated for each week of each model.

[0192] The store model training unit 86d is a processing unit that trains the store average outflow prediction model. The store model training unit 86d uses the machine data 85a and the machine-specific performance data 85b to train the store average outflow prediction model, which predicts the average outflow for each store for the target machine. In this training, training is performed for each machine, divided into weeks from week 1 to week 10 after introduction. Therefore, store average outflow prediction models are generated for each week for each machine at each store.

[0193] The prediction control unit 86e is a processing unit that controls the prediction of the average output for the target machine and manages the percentile data 85d. If the machine data 85a is updated, the prediction control unit 86e extracts data related to the updated machine from the machine data 85a and uses the extracted parameters to perform simulations regarding the number of balls dispensed nationwide and for each store.

[0194] The prediction control unit 86e calculates the average number of balls dispensed and predetermined percentile values ​​(for example, percentile values ​​at 10%, 30%, 50%, 70%, and 90%) from the simulation results and stores them in the percentile data 85d.

[0195] The nationwide model prediction unit 86f is a processing unit that uses a trained nationwide out-of-town average prediction model to predict the out-of-town average related to the national average. The nationwide model prediction unit 86f inputs the percentile values ​​related to the national average from the percentile data 85d and data related to the characteristics of the target machine extracted from the machine data 85a into the nationwide out-of-town average prediction model, outputs a predicted value of the out-of-town average related to the national average, and stores it in the prediction data 85e.

[0196] The store model prediction unit 86g is a processing unit that uses a trained store out-of-pocket average prediction model to predict the out-of-pocket average for each store. The store model prediction unit 86g inputs the percentile values ​​for each store in the percentile data 85d and data related to the characteristics of the target models extracted from the model data 85a into the store out-of-pocket average prediction model, outputs a predicted value for the out-of-pocket average for each store, and stores it in the prediction data 85e.

[0197] The analysis unit 86h is a processing unit that manages the first-week performance ratio data 85f and the first-week contribution data 85g. When the forecast data 85e is updated, the analysis unit 86h calculates the first-week performance ratio using the forecast data 85e's first-week average output forecast value and the corresponding first-week output average value of the type-specific performance data 85c. Specifically, the first-week performance ratio is calculated by dividing the forecast data 85e's average output forecast value by the type-specific performance data 85c. The analysis unit 86h stores the first-week performance ratio calculated for the national average and for each store in the first-week performance ratio data 85f.

[0198] Furthermore, if the first-week performance ratio data 85f is updated, the analysis unit 86h calculates the first-week contribution using the performance ratio for each store and the national average performance ratio from the first-week performance ratio data 85f. Specifically, it calculates the first-week contribution by dividing the performance ratio for each store by the national average performance ratio. The analysis unit 86h stores the first-week contribution calculated for each store in the first-week contribution data 85g.

[0199] Furthermore, if the analysis unit 86h receives an analysis request from the control device 50, it extracts the first-week performance ratio and first-week contribution for the stores of the control device 50, as well as the first-week performance ratio for the national average, from the first-week performance ratio data 85f and the first-week contribution data 85g, and notifies the control device 50.

[0200] Next, an example of the data stored in the storage unit 85 of the information management device 80 shown in Figure 12 will be described. Figures 13 to 16 show an example of the model data 85a, model-specific performance data 85b, type-specific performance data 85c, percentile data 85d, forecast data 85e, first-week performance ratio data 85f, and first-week contribution data 85g shown in Figure 12.

[0201] The machine data 85a shown in Figure 13(a) associates the machine "EV01" with a release year of "2019", parameters of "ad7eb67v", machine type of "mid-range machine", and features a "○" for explosive power and a "×" for easy initial wins. For the machine "GW30", the release year of "2020", parameters of "4wh3v9xz", machine type of "amateur pachinko machine", and features a "×" for explosive power and a "○" for easy initial wins.

[0202] Furthermore, model data 85a associates the model "EV02" with a release year of "2023", parameters of "t2hqm6bz", model type of "mid-range machine", and features such as explosive power being "○" and ease of initial wins being "×". Similarly, it associates the model "DX04" with a release year of "2023", parameters of "en83nfeo", model type of "mid-range machine", and features such as explosive power being "○" and ease of initial wins being "○".

[0203] The performance data 85b by machine type shown in Figure 13(b) associates the following conditions: the machine type is "EV01", the machine type is "Middle Machine", and the store ID is "National Average", with the average output being "15,000" balls in the first week, "15,000" balls in the second week, and "15,000" balls in the tenth week.

[0204] Furthermore, the performance data 85b for each machine type associates the following conditions with the machine being "EV01", the machine type being "Middle Machine", and the store ID being "ABC789", where the average output for the first week is "37,000" balls, the second week is "35,000" balls, and the tenth week is "20,000" balls.

[0205] Furthermore, the performance data 85b for each machine type associates the following conditions: the machine type is "GW30", the machine type is "amateur pachinko machine", and the store ID is "national average" with the average output of "13,000" balls in the first week, "13,000" balls in the second week, and "13,000" balls in the tenth week.

[0206] Furthermore, the performance data 85b for each machine type associates the following conditions with the machine being "GW30", the machine type being "amateur pachinko machine", and the store ID being "ABC789", where the average output for the first week is "41,000" balls, the second week is "39,000" balls, and the tenth week is "10,000" balls.

[0207] The performance data 85c by type shown in Figure 14 associates a situation where the machine type is "Middle Machine" and the store ID is "National Average" with a situation where the average output for the first week is "15,000" balls, the second week is "15,000" balls, and the tenth week is "15,000" balls.

[0208] Furthermore, the performance data 85c by type associates the state where the machine type is "Middle Machine" and the store ID is "ABC789" with the state where the average output for the first week is "18,000" balls, the second week is "18,000" balls, and the tenth week is "18,000" balls.

[0209] Furthermore, the performance data 85c by type associates a situation where the machine type is "amateur pachinko machine" and the store ID is "national average" with a situation where the average output for the first week is "16,000" balls, the second week is "16,000" balls, and the tenth week is "16,000" balls.

[0210] Furthermore, the performance data 85c by type associates the state where the machine type is "amateur pachinko machine" and the store ID is "ABC789" with the state where the average output for the first week is "19,000" balls, the second week is "19,000" balls, and the tenth week is "19,000" balls.

[0211] The percentile data 85d shown in Figure 15(a) associates the following states with the machine model being "EV02", the store ID being "National Average", and the time period being "Week 1", where the 10th percentile value is "2000", the 30th percentile value is "9000", and the 90th percentile value is "30000".

[0212] Furthermore, percentile data 85d associates the following states with the machine model being "EV02", the store ID being "ABC789", and the time period being "Week 1": the 10th percentile value is "4000" balls, the 30th percentile value is "18000" balls, and the 90th percentile value is "70000" balls.

[0213] Furthermore, percentile data 85d associates the following states with the machine model being "EV02", the store ID being "ABC789", and the time period being "Week 2": the 10th percentile value is "3000" balls, the 30th percentile value is "15000" balls, and the 90th percentile value is "65000" balls.

[0214] Furthermore, percentile data 85d associates the following states with the machine model being "EV02", the store ID being "ABC789", and the time period being "Week 10": the 10th percentile value is "2000" balls, the 30th percentile value is "8000" balls, and the 90th percentile value is "38000" balls.

[0215] The prediction data 85e shown in Figure 15(b) associates the following conditions: the machine model is "EV02", the machine type is "Middle Machine", and the store ID is "National Average", with the predicted average output values ​​being "26,000" balls in the first week, "24,000" balls in the second week, and "15,000" balls in the tenth week.

[0216] Furthermore, prediction data 85e associates the following conditions with the machine model being "EV02", the machine type being "Mid-range", and the store ID being "ABC789": the average output for the first week is "36,000" balls, for the second week it is "34,000" balls, and for the tenth week it is "21,000" balls.

[0217] Furthermore, the prediction data 85e associates the following conditions with the machine model being "DX04", the machine type being "Mid-range", and the store ID being "National Average": the average output for the first week is "38,000" balls, for the second week it is "35,000" balls, and for the tenth week it is "11,000" balls.

[0218] Furthermore, the prediction data 85e associates the following conditions with the machine model being "DX04", the machine type being "Mid-range", and the store ID being "ABC789": the average output for the first week is "40,000" balls, for the second week it is "37,000" balls, and for the tenth week it is "12,000" balls.

[0219] The first-week performance ratio data 85f shown in Figure 16(a) associates a performance ratio of "1.7" with a model "EV02" and a store ID "National Average," and associates a performance ratio of "2.0" with a model "EV02" and a store ID "ABC789."

[0220] Furthermore, the first-week performance comparison data 85f associates a performance ratio of "2.5" with a model "DX04" and a store ID "National Average," and associates a performance ratio of "2.2" with a model "DX04" and a store ID "ABC789."

[0221] The first-week performance comparison data 85f shown in Figure 16(b) associates a state where the model is "EV02" and the store ID is "ABC789" with a contribution of "1.2", and a state where the model is "DX04" and the store ID is "ABC789" with a contribution of "0.9".

[0222] <Machine Learning Procedure According to Embodiment 1> Next, the procedure for machine learning according to this embodiment 1 will be described. Machine learning is classified into supervised learning, unsupervised learning, and reinforcement learning, but in this embodiment 1, supervised learning is adopted. In supervised learning, machine learning is trained using training data that includes inputs and correct outputs for these inputs.

[0223] While various algorithms have been developed for machine learning, such as nearest neighbor, support vector machines, decision trees, and linear regression, this embodiment 1 employs an algorithm that utilizes linear regression, which has the characteristics of being simple in calculation formula, easy for humans to understand, and the most widely used.

[0224] Linear regression is a type of regression analysis in statistics that finds the straight line that best fits a given data distribution. There are two types of linear regression: simple regression and multiple regression. A simple regression model is the case where there is only one dependent variable x, and the calculation formula is given by the following equation. y = ax + b (Equation 1)

[0225] Furthermore, the multiple regression model is used when there are multiple dependent variables x. For example, if the number of dependent variables is n, the calculation formula is given by the following equation. y = a1x1 + a2x2 + ... + a n x n +b (formula 2)

[0226] In the machine learning according to this embodiment 1, the calculation formula (Equation 2) is used to obtain the training data as y and the target variable (x1, x2, ..., x nEnter the values ​​into (Equation 2) and perform training, and the coefficients and constants (a1, a2, ..., a n We will perform the task of finding (b).

[0227] <An example of features used in model prediction according to Embodiment 1> Next, an example of features used in model prediction according to this embodiment 1 will be described. Figure 17 is a diagram showing an example of features used in model prediction according to this embodiment 1.

[0228] As shown in Figure 17, there are 13 types of features used in the model prediction according to this embodiment 1, numbered from 1 to 13. Numbers 1 to 6 are numerical values ​​extracted from the results of a simulation regarding the number of balls dispensed using the parameters of the machine to be predicted (hereinafter referred to as "dispensing simulation").

[0229] Specifically, number 1 is the average value of the payout simulation, number 2 is the 10th percentile value of the payout simulation, number 3 is the 30th percentile value of the payout simulation, number 4 is the 50th percentile value of the payout simulation, number 5 is the 70th percentile value of the payout simulation, and number 6 is the 90th percentile value of the payout simulation.

[0230] Additionally, number 7 represents whether the model was sold in 2019, number 8 represents whether the model was sold in 2020, number 9 represents whether it belongs to the U series, and number 10 represents whether it belongs to a model with explosive potential.

[0231] Additionally, number 11 indicates whether the machine has a high initial payout rate, number 12 indicates whether it is a mid-range machine, and number 13 indicates whether it is a fast-paced machine.

[0232] <Processing procedure for generating an average out-of-bounds prediction model according to Embodiment 1> Next, the processing procedure for generating the average out-of-pocket forecast model according to this embodiment 1 will be described. Figure 18 is a flowchart showing the processing procedure for generating the average out-of-pocket forecast model according to this embodiment 1. In this embodiment 1, a nationwide average out-of-pocket forecast model and a store average out-of-pocket forecast model are generated, but the processing procedure for generating both is the same, so here, both will be collectively referred to as the average out-of-pocket forecast model.

[0233] As shown in Figure 18, if the information management device 80 receives operational performance data from the management device 50 (Step S101: Yes), it stores the received data in the machine-specific performance data 85b and updates it (Step S102).

[0234] If the performance data 85b for each model is insufficient as training data and training is not performed (step S103: No), proceed to step S101.

[0235] If the machine-specific performance data 85b is sufficiently prepared as training data and training is to be performed (Step S103: Yes), training is performed using the machine-specific performance data 85b to generate an average out-of-pocket prediction model, i.e., a nationwide average out-of-pocket prediction model and a store average out-of-pocket prediction model (Step S104), and the process is terminated.

[0236] In this training, each model will be trained in separate weeks from week 1 to week 10 after deployment. Therefore, an average output prediction model will be generated for each week of each model.

[0237] <Processing procedure for calculating the first-week performance ratio and first-week contribution related to Embodiment 1> Next, the processing procedure for calculating the first-week performance ratio and first-week contribution according to this embodiment 1 will be described. Figure 19 is a flowchart showing the processing procedure for calculating the first-week performance ratio and first-week contribution according to this embodiment 1.

[0238] As shown in Figure 19, if the information management device 80 updates the model data 85a (step S201: Yes), it calculates percentile values ​​using the data related to the new model updated in the model data 85a. Then, by inputting these percentile values ​​and data related to the characteristics of the new model into the out-of-bounds average prediction model, it predicts the first-week out-of-bounds average for the new model (step S202).

[0239] Using the predicted average output for the first week, the ratio to the actual results for the first week and the contribution for the first week are calculated (step S203) and stored in the first-week actual results ratio data 85f and the first-week contribution data 85g.

[0240] If an analysis request is received from the control device 50 (Step S204: Yes), the first-week performance comparison data 85f and the first-week contribution data 85g are notified to the control device 50 (Step S205), and the process is terminated.

[0241] As described above, the gaming information provision system according to this embodiment 1 uses a national average out-of-pocket prediction model and a store average out-of-pocket prediction model, which are pre-trained models using machine learning, to predict the average out-of-pocket value for the first week. The system is configured to calculate the first-week performance ratio and the first-week contribution using this predicted average out-of-pocket value for the first week and the average for each machine type in the first week's performance. This makes it possible to efficiently and appropriately determine whether or not a candidate gaming machine should be introduced to a gaming store.

[0242] In the above embodiment 1, a configuration was described in which the average output value for the first week of a new model was predicted using the average output value for the first week. However, the present invention is not limited to this, and it is also possible to configure the invention to predict the average output value for a new model using weeks that exceed the average for the model type (weeks that contribute to operation).

[0243] [Differentiation] By the way, in Embodiment 1 described above, a configuration was described in which the first-week performance ratio and first-week contribution of a new model are calculated using the average output value in the first week. However, the present invention is not limited to this, and it is also possible to configure the system to calculate the performance ratio and contribution of a new model using the cumulative average output value up to a predetermined week. In this modified example, a game information provision system that calculates the performance ratio and contribution of a new model using the cumulative average output value up to a predetermined week will be described.

[0244] <Overview of the modified gaming information provision system> First, an overview of the game information provision system related to this modified example will be explained. Figure 20 is an explanatory diagram illustrating the overview of the game information provision system related to this modified example.

[0245] As shown in Figure 20, in the modified game information provision system, the average output from the first week to the tenth week is predicted, and the cumulative total up to a predetermined week is calculated (S11). For example, the cumulative total up to the fifth week is calculated to be 60,000 balls.

[0246] The actual ratio and contribution are calculated using the cumulative average forecast values ​​up to 5 weeks and the cumulative average values ​​for each model type up to 5 weeks (S12).

[0247] For example, if the cumulative total for each machine type over the first five weeks is 55,000 balls nationwide and 50,000 balls at your store, then your store's performance ratio is 1.1 compared to the national average, your store's performance ratio is 1.2, and your contribution is 1.1.

[0248] <Processing procedure for calculating the performance ratio and contribution related to the modified example> Next, the procedure for calculating the performance ratio and contribution related to this modified example will be explained. Figure 21 is a flowchart showing the procedure for calculating the performance ratio and contribution related to this modified example.

[0249] As shown in Figure 21, if the information management device 80 updates the model data 85a (step S301: Yes), it calculates percentile values ​​using the data related to the new model updated in the model data 85a. Then, by inputting these percentile values ​​and data related to the characteristics of the new model into the out-of-bounds average prediction model, it predicts the out-of-bounds average for the new model (step S302).

[0250] If an analysis request is received from the control device 50 (Step S303: Yes), the actual ratio and contribution up to the specified week are calculated using the predicted average output (Step S304), and stored in the actual ratio data and contribution data. Then, the actual ratio data and contribution data are notified to the control device 50 (Step S305), and the process is terminated.

[0251] As described above, the gaming information provision system according to this embodiment 1 is configured to calculate the performance ratio and contribution of new models based on the cumulative average output up to a predetermined week, so that it is possible to efficiently and appropriately determine whether or not a gaming machine that is a candidate for introduction should be introduced to a gaming parlor.

[0252] [Embodiment 2] By the way, in Embodiment 1 described above, a configuration was described in which the average output for the first week of a new model is predicted and the first-week performance ratio and first-week contribution are calculated. However, the present invention is not limited to this, and it is also possible to configure the invention to extract existing models similar to the new model using images of existing models and the new model, and to predict the target audience for the new model from the movement data of these models.

[0253] This second embodiment describes a gaming information provision system that extracts existing models similar to the new model using images of existing and new models, and predicts the target audience for the new model from the movement data of these models.

[0254] <Overview of the game information provision system according to Embodiment 2> First, an overview of the game information provision system according to this second embodiment will be described. Figure 22 is an explanatory diagram illustrating the overview of the game information provision system according to this second embodiment.

[0255] As shown in Figure 22, the game information provision system according to this second embodiment performs clustering analysis using images of existing models and images of new models (S21). Specifically, images similar in color and outline to the images of new models are extracted from the images of existing models.

[0256] Using nationwide mobile data, clustering analysis is performed to extract representative customer segments that played on the selected gaming machines (S22). This customer segment becomes the target audience for the new machine. For example, a target audience for the new machine would be men in their 30s who are company employees.

[0257] As described above, the game information provision system according to this second embodiment is configured to extract existing models similar to the new model using images of existing and new models, and to predict the target audience for the new model from the play data of these models. Therefore, it is possible to efficiently and appropriately determine whether or not a candidate game machine should be introduced to a game parlor.

[0258] In the above-described embodiment 2, an existing model similar to the new model was extracted using images of existing and new models, and the target audience for the new model was predicted from the play data of this model. However, the present invention is not limited to this, and a service can also be configured to recommend a new model that is thought to be to the player's liking based on the player's gender, age, etc.

[0259] Furthermore, while the above-described embodiment 2 explains a configuration in which existing models similar to the new model are extracted using images of existing and new models, and the target audience for the new model is predicted from the movement data of these models, the present invention is not limited to this. The invention can also be configured to extract existing models similar to the new model by targeting only models that have been introduced in the store. This makes it possible to select a model that suits the customer base of the store.

[0260] Furthermore, while Embodiment 2 described above explains a configuration in which existing models similar to the new model are extracted using images of existing and new models, and the target audience for the new model is predicted from the movement data of these models, the present invention is not limited to this. The system can also be configured to perform image analysis on the employee's smartphone app in real time using images taken during a test play of the new model by a store employee. This can accelerate the introduction of the new model.

[0261] [Embodiment 3] By the way, in the above embodiment 1, we described a configuration that predicts the average output for the first week of a new model and calculates the first-week performance ratio and the first-week contribution. However, the present invention is not limited to this, and it is also possible to configure the invention to use various information about the new model, perform cluster analysis for each piece of information, and generate a catchy phrase that intuitively shows what kind of model the new model is.

[0262] This third embodiment describes a gaming information provision system that uses various information about a new model to perform cluster analysis on each piece of information and generates a catchy phrase that intuitively conveys what kind of model the new model is.

[0263] <Overview of the game information provision system according to Embodiment 3> First, an overview of the game information provision system according to this third embodiment will be described. Figures 23 and 24 are explanatory diagrams illustrating the overview of the game information provision system according to this third embodiment.

[0264] As shown in Figure 23, in the game information provision system according to this embodiment 3, existing models are classified based on a classification table (S31). The classification table is divided into five classification categories, and detailed classifications are set for each classification category.

[0265] Specifically, classification category 1 is defined as "probability category", wherein for detailed subcategories, "1" is defined as "100 or less (operation: high)", and predetermined numerical values are also set for "2" and subsequent subcategories. Classification category 2 is "payout distribution classification 1", wherein for detailed subcategories, "1" is defined as "small variation (stable) (operation: high)", and predetermined variation settings are also implemented for "2" and subsequent subcategories. Classification category 3 is "payout distribution classification 2", wherein for detailed subcategories, "1" is defined as "no outlier payout (operation: low)", and "2" is defined as "has outlier payout (one-hit knockout) (operation: high)".

[0266] Classification category 4 is "minimum payout classification", wherein for detailed subcategories, "1" is defined as "1000 or less", and "9" is defined as "10000 or more". Classification category 5 is "maximum payout classification", wherein for detailed subcategories, "1" is defined as "10000 or less", and "9" is defined as "90000 or more".

[0267] For example, when classifying Model A and Model B, for Model A, classification is performed such that the detailed subcategory of classification category 1 is "1", the detailed subcategory of classification category 2 is "1", the detailed subcategory of classification category 3 is "1", the detailed subcategory of classification category 4 is "2", and the detailed subcategory of classification category 5 is "3". Since this classification result is characteristic, the name "Kamen Taro" is registered as the "representative" for this classification (S32).

[0268] For Model B, classification is performed such that the detailed subcategory of classification category 1 is "1", the detailed subcategory of classification category 2 is "2", the detailed subcategory of classification category 3 is "1", the detailed subcategory of classification category 4 is "9", and the detailed subcategory of classification category 5 is "9". Since this classification result is not characteristic, it is not registered as the "representative" for this classification.

[0269] Next, as shown in Figure 24, clustering analysis of information related to the new model, that is, a classification category table is used to determine the degree of matching with similar existing models (S33). Based on this degree of matching with existing models, the "representative" name and glossary of the classification category table are used to generate a catchphrase for the new model (S34).

[0270] In this glossary, the terms to be used are defined as follows: If the degree of agreement is "matches in all classification categories," the term used is "almost" or "exactly the same." If the degree of agreement is "matches except for one classification category," the term used is "rough waves."

[0271] Furthermore, if the degree of similarity is "no similarity in all classification categories and the number of classifications relative to the operating value is above a certain level," the term used will be "promising newcomer," "new type," or "dawn of a new era." If the degree of similarity is "no similarity in all classification categories and the number of classifications relative to the operating value is below a certain level," the term used will be "miss." If the degree of similarity does not fall into any of the above categories, no catchphrase will be generated.

[0272] For example, if the information about a new model matches the information about model A to the degree of "matching except for one classification category," the catchphrase for the new model will be generated as "Kamen Taro of the Rough Waves."

[0273] As described above, the game information provision system according to this third embodiment is configured to use various information about new models to perform cluster analysis on each piece of information and generate catchphrases that intuitively convey what kind of new model it is. Therefore, it is possible to efficiently and appropriately determine whether or not a candidate game machine should be introduced to a game parlor.

[0274] [Embodiment 4] By the way, in the above embodiment 1, a configuration was described in which the average output for the first week of a new model is predicted and the first-week performance ratio and first-week contribution are calculated. However, the present invention is not limited to this, and it is also possible to configure the invention to predict which new models stores in a trading area will introduce, using model information and store information within the trading area.

[0275] This fourth embodiment describes a gaming information provision system that uses model information and store information within the trading area to predict which new gaming machines will be introduced by stores within the trading area.

[0276] <Overview of the game information provision system according to Embodiment 4> First, an overview of the game information provision system according to this fourth embodiment will be described. Figures 25 and 26 are explanatory diagrams illustrating the overview of the game information provision system according to this fourth embodiment.

[0277] As shown in Figure 25, the gaming information provision system according to this embodiment 4 stores machine data and installation data of machines in stores within the trading area in a database. Using the machine data and installation data for existing machines, the regression variable for logistic regression analysis is extracted, with whether or not the target machine was installed in a gaming store as the dependent variable (S41).

[0278] As shown in Figure 26, logistic regression analysis is performed using model data and installation data for the new model to predict the number of installations of the new model in stores within the trading area (S42). Then, the predicted installation values ​​are classified using a classification table (S43) and displayed with classifications such as "○" and "×".

[0279] Specifically, we will make installation predictions for new models D, E, and F, which will be released in October, at stores X, Y, and Z. For example, the model data for new model D is that the manufacturer is "Hachiyama Bussan", the model type is "play type", the category is "sea-themed", and other details include "playtime available", "model price", and "TY". The installation data for store X is that the installation rate of Hachiyama Bussan is "7%", the installation rate of play type is "20%", the installation rate of sea-themed machines is "10%", the number of units installed in the same month last year was "5", and the total number of units introduced by the same manufacturer last year was "30".

[0280] Furthermore, the classification table uses the following criteria: a predicted installation value of "less than 0.5" is classified as "×", a predicted installation value of "0.5 or more but less than 0.7" is classified as "△", a predicted installation value of "0.7 or more but less than 0.9" is classified as "〇", and a predicted installation value of "0.9 or more" is classified as "◎".

[0281] Then, using logistic regression analysis, we predict the installation of new models D, E, and F in stores X, Y, and Z for the October release. When these predictions are then classified using a classification table, the following prediction results are output.

[0282] The installation forecast for model D is "◎" for store X, "◎" for store Y, and "◎" for store Z. The installation forecast for model E is "△" for store X, "×" for store Y, and "△" for store Z. The installation forecast for model F is "◎" for store X, "〇" for store Y, and "×" for store Z.

[0283] As described above, the gaming information provision system according to this embodiment 4 is configured to predict new gaming machines that stores within a trading area are likely to introduce, using machine information and store information within the trading area. Therefore, it is possible to efficiently and appropriately determine whether or not a gaming machine that is a candidate for introduction should be introduced to a gaming store.

[0284] In the above-described embodiment 4, a configuration was explained in which new models are predicted to be introduced by stores within a trading area using model information and store information within the trading area. However, the present invention is not limited to this, and it is also possible to configure the system to analyze the regularity of new machine replacement dates at stores within a trading area from data, perform logistic regression analysis using the analysis results, and derive a strategy to maintain the store's competitive advantage.

[0285] Furthermore, while Embodiment 4 described above explains a configuration that predicts new models that stores within a trading area are likely to introduce using model information and store information within the trading area, the present invention is not limited to this. It is also possible to configure the system to predict new models by obtaining more detailed installation information of stores within the trading area through web scraping or the like.

[0286] Furthermore, each configuration illustrated in each of the above embodiments is schematic in terms of function, and does not necessarily need to be physically configured as illustrated. That is, the form of dispersion and integration of each device is not limited to that illustrated, and all or part of the device can be functionally or physically dispersed and integrated in any unit in accordance with various loads and usage conditions. [Industrial Applicability]

[0287] The gaming information providing apparatus and gaming information providing method according to the present invention are suitable for efficiently and appropriately determining whether or not a candidate gaming machine to be introduced should be introduced into a gaming parlor. [Description of Reference Numerals]

[0288] 10 Inter-stand card processing machine 11 Status display unit 12 Banknote conveying unit 12a Banknote insertion slot 13 Display operation unit 14 Reader / writer 14a Card insertion slot 14b Holding unit 15 Communication unit 16 Storage unit 16a Own device status data 16b Card data 17 Control unit 17a Data management unit 18 Gaming management unit 18a Authentication processing unit 18b Authentication key management unit 18c Gaming machine status management unit 20 Gaming machine 21 Handle 22 Counting button 23 Communication control unit 24 Performance control unit 24a Performance symbol lottery unit 24b Performance lottery unit 25 Gaming control unit 26 Gaming ball control unit 26a Playable number management unit 30 Island controller 50 Management device 51 Display section 52 Input section 53 External Network Communication Unit 54 Store Network Communications Department 55 Storage section 55a Card Management Data 55b Device Management Data 55c Member Management Data 55d Operational Data 55e Gaming-related card processing history data 55f Non-game card processing history data 55g Counting and Settlement Processing History Data 55h Gaming History Data 55i Mobile Data 55j operational performance data 55k data compared to first-week performance 55L First Week Contribution Data 56 Control Unit 56a Card Management Department 56b Equipment management department 56c Membership Management Department 56d Operation Data Storage and Aggregation Unit 56e Card Processing History Management Department 56f Game History Data Generation Unit 56g Mobile data generation unit 56h Operational Performance Management Department 56i Analysis Department 60 Prize Management Device 70 Payment machine 80 Information management device 81 Display section 82 Input section 84 Communications Department 85 Storage section 85a Model Data 85b Performance data by model 85c Type-Specific Performance Data 85d percentile data 85e Prediction Data 85f First week performance comparison data 85g First Week Contribution Data 86 Control Unit 86a Model Data Management Department 86b Performance Management Department 86c National Model Training Department 86d Store Model Training Department 86e Prediction Control Unit 86f National Model Prediction Department 86g Store Model Prediction Department 86h Analysis Department

Claims

1. A gaming information providing device that outputs gaming information including the predicted results for the introduction of candidate gaming machines to be newly introduced to a gaming parlor, A first trained model generation means generates a first trained model targeting the amusement parlor by performing machine learning based on performance information at the amusement parlor, including the average number of game media played by players over a predetermined period, and the model information of the specific type of amusement machine, with respect to a specific type of amusement machine installed at the amusement parlor. A first prediction result generation means that inputs input information including model information of the candidate gaming machines belonging to the specified model type into the first trained model to generate a first prediction result, A second trained model generation means generates a second trained model targeting the predetermined region by performing machine learning based on performance information in the predetermined region, including the average number of game media played by players over a predetermined period, and model information of the specific type of game machine, with respect to the game machines of a particular model installed in multiple game parlors located in a predetermined region. A second prediction result generation means that inputs input information including model information of the candidate gaming machines belonging to the specified model type into the second trained model to generate a second prediction result, A calculation means for calculating the predicted introduction result of the candidate gaming machine based on at least the first prediction result and the second prediction result, A gaming information providing device characterized by having the following features.

2. The aforementioned calculation means is, The gaming information providing device according to claim 1, characterized in that it calculates the predicted introduction result of the candidate gaming machine based on the performance information at the gaming parlor, the first prediction result, the performance information in the predetermined region, and the second prediction result.

3. The aforementioned calculation means is, A first historical level ratio calculation means that calculates a first historical level ratio by dividing the average output included in the first prediction result by the average output included in the performance information of the amusement parlor, A first historical level ratio calculation means calculates a second historical level ratio by dividing the average output included in the second prediction result by the average output included in the actual performance information in the predetermined region. The gaming information providing device according to claim 2, characterized by being equipped with the following:

4. The aforementioned calculation means is, The gaming information providing device according to claim 3, further comprising a means for calculating the store's contribution to the gaming store by dividing the first past level ratio by the second past level ratio.

5. The first trained model generation means described above is: Using a machine learning framework for multiple regression models of linear regression algorithms, a trained linear regression model targeting the aforementioned amusement parlor is generated. The second pre-trained model generation means described above is: A trained linear regression model targeting the predetermined region is generated using the machine learning framework of the multiple regression model of the linear regression algorithm. A gaming information providing device according to any one of features 1 to 4.

6. The first trained model generation means described above is: The number of balls dispensed at multiple percentile values, sales model information, and model characteristics of the specific model type of gaming machine installed in the aforementioned gaming parlor are accepted as model information for the specific model type of gaming machine installed in the aforementioned gaming parlor. The second pre-trained model generation means described above is: The system accepts the number of balls dispensed at multiple percentile values, sales model information, and model characteristics of the specified type of gaming machine installed in the specified gaming parlors located in the specified region, as model information for the specified type of gaming machine installed in multiple gaming parlors located in the specified region. The gaming information providing device according to feature 5.

7. The first trained model generation means described above is: The aforementioned amusement parlor receives information including the average number of payouts per week from the first week to the tenth week after the start of play for multiple gaming machines installed at the aforementioned amusement parlor, as performance information for the aforementioned amusement parlor. The second pre-trained model generation means described above is: Information including the average number of payouts per week from the first to the tenth week after the start of play for multiple gaming machines installed in multiple amusement parlors located in the aforementioned designated area will be accepted as performance information for the aforementioned designated area. The gaming information providing device according to feature 5.

8. A method for providing game information in a game information providing device that outputs game information including the predicted results for the introduction of candidate game machines to be newly introduced to a game parlor, A first trained model generation step involves generating a first trained model targeting the amusement parlor by performing machine learning based on performance information at the amusement parlor, including the average number of game media played by players over a predetermined period, and the model information of the specific type of amusement machine, with respect to a specific type of amusement machine installed at the amusement parlor. A first prediction result generation step involves inputting input information, including model information of the candidate gaming machines belonging to the specified model type, into the first trained model to generate a first prediction result; A second trained model generation step involves generating a second trained model targeting the predetermined region by performing machine learning based on performance information in the predetermined region, including the average number of game media played by players over a predetermined period, and the model information of the specific type of game machine, with respect to the game machines of a particular model installed in multiple game parlors located in a predetermined region. A second prediction result generation step involves inputting input information, including model information of the candidate gaming machines belonging to the specified model type, into the second trained model to generate a second prediction result. A calculation step of calculating the predicted introduction result of the candidate gaming machine based on at least the first prediction result and the second prediction result, A method for providing game information, characterized by including the following.

Images