Media recommendation system, table creation system, media recommendation device and table creation method

Abstract

Provided are media recommendation systems and the like that enable the recommendation of media or media collections suitable for users in online games. The present invention is a media recommendation system for recommending media or media collections in a game that progresses as a user selects media from an owned media group constructed to include D media and places the media on a game field. The recommended media or media collections are media or media collections to be added to a game in order to construct at least a portion of the owned media group from the media, wherein the individual media included in the owned media group are selected from T types of media included in a media set pre-set in the game. The media recommendation system includes: an input acceptance unit for accepting input of a single media from the T types of media; a table storage unit for storing a first table; and a media recommendation unit capable of outputting, as recommendation information, a media or media collection associated in the first table with the single media accepted by the input acceptance unit.

Description

Technical Field

[0001] The present invention relates to a media recommendation system, a table creation system, a media recommendation device and a table creation method. Background Art

[0002] Recently, an increasing number of players are enjoying online games (games that multiple players can participate in via a network). These games are implemented using a gaming system in which a portable terminal device or the like communicates with a server device or the like of a game administrator, allowing the player operating the portable terminal device to battle with other players.

[0003] As an online game, there is a card game called a digital collectible card game (DCCG), in which various actions are performed based on the combination of game media such as cards and characters. As the number of game media provided increases linearly, the strategic complexity of the game increases dramatically, and the number of combinations of game media increases explosively.

[0004] Prior art literature

[0005] Patent Literature

[0006] Patent Document 1: Japanese Patent 4886749 Summary of the Invention

[0007] Problems to be solved by the invention

[0008] The mainstream of existing information recommendation technology is to provide recommendations using a limited combination of factors or similarities in user preferences as a key. Specifically, existing information recommendation technology estimates the purchase likelihood of other items based on user attribute information or historical information such as the user's purchase history. This model can be constructed in the form of a matrix interpolation problem. For example, Patent Document 1 discloses a recommended product selection device related to existing information recommendation technology.

[0009] The above-mentioned technology is used in e-commerce websites and the like. However, unlike purchasing products in e-commerce, in DCCG, users may own all media and may randomly acquire media. Therefore, it is difficult to directly apply the above-mentioned technology to technology for recommending media (such as cards) in DCCG. Therefore, there is a need for technology for recommending media or a collection of media suitable for users in online games such as DCCG.

[0010] The present invention has been made to solve the above-mentioned problems, and its main object is to provide a media recommendation system or the like that makes it possible to recommend a media or a set of media suitable for a user in an online game.

[0011] Solutions for solving problems

[0012] To achieve the above-mentioned object, a media recommendation system according to an aspect of the present invention is a media recommendation system for recommending a medium or a set of media in a game that progresses as a user selects a medium from an owned media group constructed to include D media and places the medium in a game field, wherein the recommended medium or set of media is a medium or set of media to be added to the game in order to construct at least a part of the owned media group from the medium, wherein each media included in the owned media group is selected from T types of media included in a media set preset in the game, and the media recommendation system is characterized by comprising:

[0013] an input accepting unit, configured to accept an input of a single medium from the T types of media;

[0014] a table storage unit for storing a first table created by further extracting, for each category of media, a set of a predetermined number of media with a high frequency of appearance from a set of k media with a high frequency of appearance included in the plurality of owned media groups, extracted in advance based on owned media group information related to a plurality of owned media groups used by a predetermined plurality of users, and associating, for each category of media, a single medium with a medium or a set of media obtained by excluding the medium from the further extracted set of media, where 2≤k≤D; and

[0015] A media recommendation unit is configured to output, as recommendation information, a medium or a set of media that is associated with the single medium accepted by the input acceptance unit in the first table.

[0016] Furthermore, in order to achieve the above-mentioned object, a table creation system according to an aspect of the present invention is a table creation system for creating a table for recommending a medium or a set of media in a game that progresses as a user selects a medium from a owned media group constructed to include D media and places the medium in a game field, the recommended medium or set of media being a medium or a set of media to be added to the medium in order to construct at least a part of the owned media group from the medium, the individual media included in the owned media group being selected respectively from T types of media included in a media set pre-set in the game, the table creation system being characterized in that,

[0017] The table is created by the following operations: from a set of media consisting of k media with high occurrence frequency included in the multiple owned media groups, which is pre-extracted based on owned media group information related to multiple owned media groups used by a specified multiple users, a set of a specified number of media with high occurrence frequency is further extracted for each class of media, and for each class of media, a single medium is associated with a medium or a set of media obtained by excluding the medium from the set of further extracted media, where 2≤k≤D.

[0018] Furthermore, to achieve the above-mentioned object, a media recommendation device according to an aspect of the present invention is a media recommendation device for recommending a medium or a set of media in a game that progresses as a user selects a medium from an owned media group constructed to include D media and places the medium in a game field, wherein the recommended medium or set of media is a medium or set of media to be added to the medium in order to construct at least a part of the owned media group from the medium, wherein each media included in the owned media group is selected from T types of media included in a media set pre-set in the game, and the media recommendation device is characterized by comprising:

[0019] an input accepting unit, configured to accept an input of a single medium from the T types of media;

[0020] a table storage unit for storing a first table created by further extracting, for each category of media, a set of a predetermined number of media with a high frequency of appearance from a set of k media with a high frequency of appearance included in the plurality of owned media groups, extracted in advance based on owned media group information related to a plurality of owned media groups used by a predetermined plurality of users, and associating, for each category of media, a single medium with a medium or a set of media obtained by excluding the medium from the further extracted set of media, where 2≤k≤D; and

[0021] A media recommendation unit is configured to output, as recommendation information, a medium or a set of media that is associated with the single medium accepted by the input acceptance unit in the first table.

[0022] Furthermore, to achieve the above-mentioned object, a table creation method according to an aspect of the present invention is a table creation method for creating a table for recommending a medium or a set of media in a game that progresses as a user selects a medium from an owned media group constructed to include D media and places the medium on a game field, the recommended medium or set of media being a medium or a set of media to be added to the medium in order to construct at least a part of the owned media group from the medium, the individual media included in the owned media group being selected from T types of media included in a media set preset in the game, the table creation method being characterized by comprising the following steps:

[0023] For each of k=2 to D, extracting a set of k media with high appearance frequencies included in the plurality of owned media groups based on owned media group information related to a plurality of owned media groups used by a plurality of users, where 2≤k≤D;

[0024] For each of k=2 to D, from the extracted set of k media with high appearance frequencies, further extract a set of a predetermined number of media with high appearance frequencies for each type of media; and

[0025] For each of k=2 to D, a table is created by associating, for each class of media, a single medium with a medium or a set of media obtained by excluding the medium from the set of further extracted media.

[0026] Furthermore, to achieve the above-mentioned object, a media recommendation system according to an aspect of the present invention is a media recommendation system for recommending media in a game that progresses as a user selects a media from an owned media group configured to include D media and places the media on a game field, the recommended media being media to be added to a set of media in order to construct at least a part of the owned media group from a set of media, wherein each media included in the owned media group is selected from a media set including T types of media pre-set in the game, the media recommendation system being characterized by comprising:

[0027] an input accepting unit for accepting an input of a set of k media constituting at least a part of the owned media group, where 2≤k≤D-1;

[0028] The table storage unit is used to store a second table, wherein the second table is generated by the following operation: when using a set of media consisting of k media and a set of media consisting of k+1 media with high appearance frequencies included in the plurality of owned media groups extracted in advance based on owned media group information related to the plurality of owned media groups used by a predetermined plurality of users, the media included in the media set are respectively combined with the pre-extracted sets of media consisting of k media, thereby generating a plurality of sets of media consisting of k+1 media, and for each pre-extracted set of media consisting of k media, The method further comprises: calculating the occurrence frequencies of each of the generated sets of media consisting of k+1 media with reference to the pre-extracted set of media consisting of k+1 media, thereby further extracting a predetermined number of sets of media consisting of k+1 media with high occurrence frequencies from the generated sets of media consisting of k+1 media for each pre-extracted set of media consisting of k media, and associating each pre-extracted set of media consisting of k media with the media to be added in order to generate the further extracted set of media consisting of k+1 media with high occurrence frequencies; and

[0029] A media recommendation unit capable of outputting as recommendation information the media that is associated with the set of media consisting of k media accepted by the input acceptance unit in the second table and is to be added in order to generate a set of media consisting of k+1 media.

[0030] Furthermore, in the present invention, preferably, the second table is generated by the following operation: when using a set of media consisting of k media and a set of media consisting of k+1 media with high occurrence frequencies included in the plurality of owned media groups pre-extracted based on owned media group information related to a plurality of owned media groups used by a specified plurality of users, the T types of media included in the media set are respectively combined with each pre-extracted set of media consisting of k media, thereby generating T sets of media consisting of k+1 media, and for each pre-extracted set of media consisting of k media, The occurrence frequencies of each of the generated sets of T media consisting of k+1 media are calculated with reference to the pre-extracted set of media consisting of k+1 media, thereby further extracting a specified number of sets of media consisting of k+1 media with high occurrence frequencies from the generated sets of T media consisting of k+1 media for each pre-extracted set of media consisting of k+1 media, and associating the pre-extracted sets of media consisting of k media with the media to be added in order to generate the further extracted sets of media consisting of k+1 media with high occurrence frequencies, where 2≤k≤D-1.

[0031] Furthermore, in order to achieve the above-mentioned object, a table creation system according to an aspect of the present invention is a table creation system for creating a table for recommending media in a game that progresses as a user selects a medium from a group of owned media constructed to include D media and places the medium in a game field, the recommended media being a medium to be added to a set of media in order to construct at least a part of the owned media group from a set of media, the individual media included in the owned media group being selected separately from a set of media including T types of media pre-set in the game, the table creation system being characterized in that,

[0032] The table is created by the following operation: when using a set of media consisting of k media and a set of media consisting of k+1 media with high appearance frequencies included in the multiple owned media groups extracted in advance based on owned media group information related to the multiple owned media groups used by the specified multiple users, the media included in the media set are respectively combined with the pre-extracted sets of media consisting of k media, thereby generating multiple sets of media consisting of k+1 media, and for each pre-extracted set of media consisting of k media, referring to the pre-extracted sets of media consisting of k+ 1 medium is used to calculate the occurrence frequencies of the multiple sets of media composed of k+1 media generated, thereby further extracting a specified number of sets of media composed of k+1 media with high occurrence frequencies from the multiple sets of media composed of k+1 media generated for each pre-extracted set of media composed of k media, and associating the pre-extracted sets of media composed of k media with the media to be added in order to generate the further extracted sets of media composed of k+1 media with high occurrence frequencies, where 2≤k≤D-1.

[0033] Furthermore, to achieve the above-mentioned object, a media recommendation device according to an aspect of the present invention is a media recommendation device for recommending media in a game that progresses as a user selects a media from a owned media group configured to include D media and places the media on a game field, wherein the recommended media is a media to be added to a set of media in order to construct at least a part of the owned media group from a set of media, wherein each media included in the owned media group is selected from a media set including T types of media pre-set in the game, and the media recommendation device is characterized by comprising:

[0034] an input accepting unit for accepting an input of a set of k media constituting at least a part of the owned media group, where 2≤k≤D-1;

[0035] The table storage unit is used to store a second table, wherein the second table is generated by the following operation: when using a set of media consisting of k media and a set of media consisting of k+1 media with high appearance frequencies included in the plurality of owned media groups extracted in advance based on owned media group information related to the plurality of owned media groups used by a predetermined plurality of users, the media included in the media set are respectively combined with the pre-extracted sets of media consisting of k media, thereby generating a plurality of sets of media consisting of k+1 media, and for each pre-extracted set of media consisting of k media, The method further comprises: calculating the occurrence frequencies of each of the generated sets of media consisting of k+1 media with reference to the pre-extracted set of media consisting of k+1 media, thereby further extracting a predetermined number of sets of media consisting of k+1 media with high occurrence frequencies from the generated sets of media consisting of k+1 media for each pre-extracted set of media consisting of k media, and associating each pre-extracted set of media consisting of k media with the media to be added in order to generate the further extracted set of media consisting of k+1 media with high occurrence frequencies; and

[0036] A media recommendation unit capable of outputting as recommendation information the media that is associated with the set of media consisting of k media accepted by the input acceptance unit in the second table and is to be added in order to generate a set of media consisting of k+1 media.

[0037] Furthermore, to achieve the above-mentioned object, a table creation method according to an aspect of the present invention is a table creation system for creating a table for recommending media in a game that progresses as a user selects a media from a owned media group constructed to include D media and places the media on a game field, the recommended media being media to be added to a set of media in order to construct at least a part of the owned media group from a set of media, the respective media included in the owned media group being selected from a set of media including T types of media pre-set in the game, the table creation method being characterized by comprising the following steps:

[0038] For each of k=2 to D-1, based on owned media group information related to a plurality of owned media groups used by a plurality of users, extracting a set of media consisting of k media with high appearance frequencies included in the plurality of owned media groups, where 2≤k≤D-1;

[0039] For each of k=2 to D-1, combining the media included in the media set with the extracted sets of media consisting of k media, thereby generating a plurality of sets of media consisting of k+1 media, and calculating, for each of the extracted sets of media consisting of k media, the occurrence frequencies of the generated plurality of sets of media consisting of k+1 media with reference to the extracted sets of media consisting of k+1 media, thereby further extracting, for each of the extracted sets of media consisting of k media, a prescribed number of sets of media consisting of k+1 media with high occurrence frequencies from the generated plurality of sets of media consisting of k+1 media; and

[0040] A table is created by associating, for each of k=2 to D-1, each extracted set of k media with media to be added to generate a further extracted set of k+1 media with a high frequency of occurrence.

[0041] Furthermore, in the present invention, preferably,

[0042] The table creation system further includes a medium extraction system configured to extract a combination of any number of mediums with a high appearance frequency included in the owned medium group.

[0043] The media extraction system includes:

[0044] a storage unit, configured to store the owned media group information; and

[0045] an extraction unit capable of calculating the respective occurrence frequencies of a set consisting of m media with reference to the owned media group information stored in the storage unit, and extracting a set consisting of m media with a high occurrence frequency that meets a prescribed standard, wherein 2≤m≤D, and

[0046] The extraction unit is capable of generating multiple sets consisting of m+1 media by respectively combining the media included in the media set into the extracted sets, or is capable of generating multiple sets consisting of m-1 media by respectively excluding a single media from the media constituting the extracted sets from the extracted sets.

[0047] Furthermore, in the present invention, preferably,

[0048] The extracting unit of the medium extracting system periodically extracts a set of 2 to D mediums having a high frequency of occurrence that meets a predetermined standard, and

[0049] The table creation system periodically creates a table based on a set of k media with high appearance frequencies included in the owned media group extracted by the extraction unit, where 2≤k≤D.

[0050] Effects of the Invention

[0051] The present invention makes it possible to recommend media or a collection of media suitable for a user in an online game. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 FIG. 4 is a diagram showing the overall structure of a media recommendation system in one embodiment of the present invention.

[0053] Figure 2 is a block diagram showing the hardware structure of a media recommendation server in one embodiment of the present invention.

[0054] Figure 3 FIG. 1 is a block diagram showing the hardware structure of a game server in one embodiment of the present invention.

[0055] Figure 4 is a block diagram showing the hardware structure of an electronic device in one embodiment of the present invention.

[0056] Figure 5 An example game screen displayed on a display of a user's electronic device is shown.

[0057] Figure 6 FIG. 4 is a functional block diagram of a media recommendation server in the first embodiment of the present invention.

[0058] Figure 7 is a functional block diagram of an electronic device in a first embodiment of the present invention.

[0059] Figure 8 An example first table generated by the table creation unit in the first embodiment is shown.

[0060] Figure 9 An example second table generated by the table creation unit in the second embodiment is shown.

[0061] Figure 10 A first example table creation method of the media recommendation system in the first embodiment of the present invention is shown.

[0062] Figure 11 A second example table creation method of the media recommendation system in the first embodiment of the present invention is shown.

[0063] Figure 12 Describes the additive inference processing of a media extraction system for extracting a combination of k cards, where k represents an arbitrary number.

[0064] Figure 13 Describes global filtering processing.

[0065] Figure 14Describes the subtraction reasoning process of the media extraction system to extract a combination of k cards, where k represents an arbitrary number. DETAILED DESCRIPTION

[0066] Now, a system according to an embodiment of the present invention will be described with reference to the accompanying drawings.

[0067] Figure 1 This is the overall structure of the media recommendation system 1 in one embodiment of the present invention. Figure 1 As shown, the media recommendation system 1 includes a media recommendation server 10, a game server 20, and an electronic device 30, which are connected to a network 2 such as the Internet to be able to communicate with each other.

[0068] Figure 2 This is a block diagram illustrating the hardware structure of a media recommendation server 10 in one embodiment of the present invention. The media recommendation server 10 in this embodiment has a structure similar to that of a conventional server. In one example, the media recommendation server 10 includes multiple computers and server devices. The media recommendation server 10 includes a processor 11, an input device 12, an output device 13, a storage device 14, and a communication device 15. These components are connected via a bus 16. Interfaces are interposed between the bus 16 and the components as needed.

[0069] The processor 11 controls the overall operation of the media recommendation server 10. For example, the processor 11 is a CPU. The processor 11 performs various processes by loading programs and data stored in the storage device 14 and executing these programs. In one example, the processor 11 is composed of multiple processors.

[0070] The input device 12 is a user interface for receiving user input to the media recommendation server 10; for example, the input device 12 is a touch screen, a touchpad, a keyboard, or a mouse. The output device 13 outputs or displays output information of the media recommendation server 10 to the user; for example, the output device 13 is a display or printer for outputting images.

[0071] The storage device 14 includes a main storage device and an auxiliary storage device. The main storage device is a semiconductor memory such as RAM. RAM is a volatile storage medium that allows high-speed reading and writing of information, and is used as a storage area and a working area when the processor 11 processes information. The main storage device may include a ROM as a read-only non-volatile storage medium. In this case, the ROM stores programs such as firmware. The auxiliary storage device stores various programs and data used by the processor 11 when executing each program. The auxiliary storage device is, for example, a hard disk device; however, the auxiliary storage device may be any type of non-volatile storage device or non-volatile memory capable of storing information, which may be removable. For example, the auxiliary storage device stores an operating system (OS), middleware, application programs, various data that can be referenced when executing these programs, and the like.

[0072] The media recommendation server 10 has a function as a database server. The storage device 14 stores data (for example, tables) and programs used for the database, and realizes the database by executing the programs.

[0073] The communication device 15 is a wireless LAN module capable of transmitting and receiving data to and from other computers such as a user terminal or a server via a network. However, the communication device 15 may alternatively be another type of wireless communication device such as a Bluetooth (registered trademark) module, or may be a wired communication device such as an Ethernet (registered trademark) module, or a USB interface.

[0074] Figure 3 1 is a block diagram illustrating the hardware structure of the game server 20 in one embodiment of the present invention. The game server 20 in the present embodiment includes a structure similar to that of a common server. In one example, the game server 20 is a known server computer. The game server 20 includes a processor 21, an input device 22, an output device 23, a storage device 24, and a communication device 25. These various components are connected via a bus 26. An interface is inserted between the bus 26 and the various components as needed. Each element (that is, the processor 21, the input device 22, the output device 23, the storage device 24, and the communication device 25) corresponds to the above-mentioned processor 11, the input device 12, the output device 13, the storage device 14, and the communication device 15, respectively, and has a structure similar to their structure, so their description will be omitted. In one example, the game server 20 is configured to include a plurality of server devices.

[0075] Figure 43 is a block diagram showing the hardware structure of an electronic device 30 according to one embodiment of the present invention. The electronic device 30 in this embodiment is a smart phone. Each electronic device 30 includes a processor 31, an input device 32, an output device 33, a storage device 34 and a communication device 35. These various components are connected via a bus 36. An interface is inserted between the bus 36 and the various components as needed. Each element (i.e., the processor 31, the input device 32, the output device 33, the storage device 34 and the communication device 35) corresponds to the above-mentioned processor 11, the input device 12, the output device 13, the storage device 14 and the communication device 15, respectively, and has a structure similar to their structure, so their description will be omitted. In this embodiment, since the electronic device 30 is a smart phone, the output device 33 is a display 33, and the storage device 34 is a storage device provided in an ordinary smart phone, which includes a RAM as a volatile memory and also includes a ROM as a non-volatile memory. However, the electronic device 30 may alternatively be another type of device having the above-described structure, such as a desktop personal computer, a laptop personal computer, or a tablet computer. In this case, the respective elements (i.e., the processor 31, the input device 32, the output device 33, the storage device 34, and the communication device 35) correspond to the above-described processor 11, input device 12, output device 13, storage device 14, and communication device 15, respectively.

[0076] The game server 20 is connected to the electronic device 30 of each user playing the game via the network 2. In the present embodiment, when each user activates the game App (application) installed in the electronic device 30, the electronic device 30 accesses the game server 20, and the game server 20 accepts access from each electronic device 30 and provides game services via the network 2. At this time, the game server 20 stores game information including the game log of each user in the storage device 24 and manages the game information. In a preferred example, the game server 20 has the function of a database server. In this case, the storage device 24 stores data (e.g., tables) and programs used by the database, and implements the database by executing the program. In the case where the game server 20 is configured to include multiple server devices, one or more of the server devices can be a database server.

[0077] The storage device 24 of the game server 20 stores game programs as applications used for the game, and the game server 20 provides the game by communicating with each user's electronic device 30. The game server 20 executes the game based on the game operation input at the user's terminal device and transmits the execution result to the user's electronic device 30. The game server 20 stores a game log including the content of the operation performed by each user and manages the game log.

[0078] The game provided by the game server in this embodiment is a so-called digital card collection game (DCCG). Specifically, when a user selects cards from a deck of cards constructed to include multiple cards and places these cards in a game field, the game in this embodiment progresses, thereby executing various events according to the combination of cards or classes. A deck of cards is generally referred to as a card pile (card deck). In addition, the game in this embodiment is a battle game in which multiple users fight each other by selecting cards from a deck of cards and placing these cards in a game field 43. For example, each card has card definition information including card ID, card type, and parameters such as life value, attack ability and attributes, and each class has class definition information. Card type is information indicating whether a card represents a character or an item.

[0079] The game server 20 stores each user's game-related information (such as the cards owned by the user) in association with the user ID. The game server 20 allows the user to obtain cards in the game through a prescribed method. The user can obtain multiple cards from all preset card types T available in the game. The obtained cards may include cards of the same type. In one example, the game server 20 performs a lottery process such as gacha in response to user operations, thereby allowing the user to obtain cards. In another example, the game server stores in-game currency associated with the user ID and, in response to a user operation to obtain a desired card by consuming the in-game currency, allows the user to obtain the desired card.

[0080] The game server 20 allows the user to construct a deck by selecting D cards from a deck of cards acquired and owned by the user. In this manner, a deck consisting of D cards is constructed. Therefore, the game in this embodiment progresses as the user selects cards from the deck consisting of D cards and places them on the playing field. In one example, when allowing the user to construct a deck, the game server 20, for example, allows the user to select D cards based on a prescribed rule, such that the user is not allowed to select more than three cards of the same type.

[0081] Figure 5 An example game screen displayed on the display 33 of the user's electronic device 30 is shown. This game screen is a game screen 40 for a card battle between a local user, who is operating the electronic device 30, and another user, who is operating another electronic device 30. Game screen 40 shows a first card deck 42a, which is the local user's hand, and a first card deck 42b, which is the other user's hand. The game is configured so that the local user cannot recognize the contents of the other user's first card deck 42b.

[0082] Each user's deck consists of a first deck 42, which serves as the user's hand, and a second deck 44, which serves as the user's inventory. Whether each user's card 41 is included in the first deck 42 or the second deck 44 is determined based on the progress of the game. The first deck 42 is a deck that the user can select and place on the game field 43, while the second deck 44 is a deck that the user cannot select. Although the deck consists of multiple cards 41, as the game progresses, a card pair may consist of a single card 41. The game progresses as each user places cards 41 from each first deck 42 (42a and 42b) on the game field 43. Note that each user's deck can consist of all cards 41 of different types, or it can be composed of a portion of cards 41 of the same type. Furthermore, the types of cards 41 that make up a user's deck can differ from the types of cards 41 that make up another user's deck. In another example, the deck of cards owned by each user consists of only the first card set 42 .

[0083] Game screen 40 shows a character 45a selected by the local user and a character 45b selected by another user. In this embodiment, the character selected by the user is different from the character associated with the card and defines a class representing the type of card deck. The game in this embodiment is configured so that the cards 41 owned by the user vary according to class. In one example, the game in this embodiment is configured so that the types of cards that can constitute each user's card deck vary according to class. In one example, the game in this embodiment is configured so that the effects produced when a card 41 is placed on the game field 43 vary according to parameters set in the card definition information.

[0084] A deck is a collection of pre-set cards in a game and includes all pre-set card types. In the game provided by the game server 20 of this embodiment, a deck consists of D cards, each selected from the T card types included in the deck. In this embodiment, for ease of explanation, D is assumed to be 40 and T is 1000, but it will be understood that these values ​​are not limited to these. Although for ease of explanation, the T card types are referred to as card-x (1≤x≤T), it will be understood that this notation is used to distinguish the individual cards, and the actual names may be different. The game log includes a user ID, a class, and deck information. The deck information is information related to the D cards that make up the deck used by the user, and is preferably associated with the user ID. The game log does not need to include a class. In one example, the game server 20 stores each deck information stored in the game log in association with at least one of the following: the deck's win rate, the deck's usage rate, and the number of user IDs associated with the deck information.

[0085] Note that the card 41 is an example of a medium, and the card pile (owned card deck) is an example of a owned medium group that is constructed to include multiple mediums owned by the user. The medium can be a character or an item, etc. For example, a six-card set (i.e., a set consisting of six cards) can be represented as a set consisting of six mediums. In addition, a card set preset in the game can be represented as a medium set. For example, in the case where the medium set is composed of mediums including characters and items, the game screen 40 shows the character or item itself as the card 41.

[0086] In Japanese Patent Application 2019-041261, the applicant proposes a media extraction system 4 (not shown) for extracting combinations of any number of media with a high frequency of appearance included in a media group. In a preferred embodiment, a media recommendation system 1 includes the media extraction system 4. The media extraction system 4 uses the game log stored in the game server 20 described above. The media extraction system 4 is configured to extract combinations of any number of cards with a high frequency of appearance included in each user's deck at the start of the game by using card pile information A including at least a portion of the card pile information stored in the game log for a plurality of card piles used by a predetermined number of users. The media recommendation server 10 is configured to be able to communicate with the media extraction system 4 and is configured to be able to receive from the media extraction system 4 the combinations of any number of cards with a high frequency of appearance extracted by the media extraction system 4. The media recommendation server 10 can receive from the media extraction system 4 the combinations of any number of cards with a high frequency of appearance extracted by the media extraction system 4 and the frequency of appearance of the card combinations.

[0087] In this embodiment, for ease of explanation, a card set consisting of k cards (k is an integer greater than or equal to 2) or a card combination pattern consisting of k cards may be referred to as a k-card set. A k-card set with a high frequency of occurrence refers to a k-card set with a high frequency of occurrence in the deck included in the deck information A. Therefore, the level of the k-card set's frequency of occurrence is proportional to the number of instances of the k-card set included in the deck included in the deck information A.

[0088] Next, the media recommendation system 1 in the first embodiment of the present invention will be described. The media recommendation system 1 is configured to be able to recommend cards or card combinations to be added to a card or card combination to constitute at least a part of a deck of cards based on the cards or card combinations in the game of this embodiment. The media recommendation system 1 is capable of outputting the cards or β-1 card sets required to generate a β-card set (2≤β≤D) including card-α (1≤α≤T) desired by the user of the electronic device 30 as recommendation information. The β-card set including card-α can be referred to as a recommended card set (recommended media set).

[0089] Figure 6 is a functional block diagram of the media recommendation server 10 in the first embodiment of the present invention, and Figure 7 is a functional block diagram of the electronic device 30 in the first embodiment of the present invention. The media recommendation server 10 includes a table creation unit 51, a table storage unit 52, a media input unit 53, and a media recommendation unit 54. The electronic device 30 includes an input acceptance unit 61 and a recommended media display unit 62. In this embodiment, these functions are implemented by the processor 11 and the processor 31 executing programs respectively. In this embodiment, similar to other embodiments, since various functions are implemented by loading programs, a part of one component (function) can be set in another component. However, alternatively, these functions can be implemented using hardware by configuring electronic circuits, etc. for implementing part or all of the various functions.

[0090] The table creation unit 51 further extracts the Z with high appearance frequency included in the card pile extracted in advance by the medium extraction system 4 for each type of card. k From a set of k cards of various types (2≤k≤D), a set of k cards with a high frequency of appearance is extracted. The table creation unit 51 creates a first table by associating, for each card type, a single card with a card or card combination obtained by excluding the single card from the further extracted set of k cards.

[0091] In a preferred example, the prescribed number is a numerical value less than or equal to Y. In this case, the prescribed number of k-card sets refers to a k-card set of at most Y types. In one example, Y is 1 to 10. In the following description of the embodiment of the present invention, unless otherwise specifically mentioned, it is assumed that the table creation unit 51 extracts k-card sets of Y types in descending order of appearance frequency for each type of card. The table creation unit 51 extracts k-card sets of Y types in a case where k-card sets of Y types can be extracted in descending order of appearance frequency. In a preferred example, for each type of card, the table creation unit 51 arranges the k-card sets in descending order of appearance frequency, and further extracts Y k-card sets in sequence from the k-card sets with higher appearance frequencies. In a preferred example, Z k (2≤k≤D)=Z, and for example, Z=10000. However, Z may be slightly increased or decreased, for example, within a range of 8000 to 12000, depending on the value of k. In this embodiment, for ease of explanation, it is assumed that Z k (2≤k≤D)=Z.

[0092] The table storage unit 52 stores the first table created by the table creation unit 51 in the storage device 14 .

[0093] Figure 8An example of the first table generated by the table creation unit 51 in the first embodiment is shown. The keys (primary keys) in the first table are T (1000) types of cards card-1 to card-T. Figure 8 In the example shown, the table creation unit 51 further extracts a set of k cards of Y types in descending order of appearance frequency for each type of T types of cards for each value of k=2 to D (40), thereby creating a first table for each value of k=2 to D.

[0094] In the case of k=2, for each of T types of cards, the table creation unit 51 extracts two card sets of Y types that include the card and have a high appearance frequency, and associates the card with a card obtained by excluding the card from the extracted two card sets, thereby creating a first table. 1-1 、card-i 1-2 , ..., and card-i 1-Y ) is associated with card-i 1-1 Indicates one of card-1 to card-T, and card-i 1-1 、card-i 1-2 , ... and card-i 1-Y are different cards. For example, card-i 1-1 、card-i 1-2 , ..., and card-i 1-Y It is card-111, card-2, ..., and card-842. 1-1 Two card sets consisting of card-1 and card-i 1-2 Two card sets consisting of, ..., and card-1 and card-i 1-Y The two card sets that are formed are the two card sets including card-1 arranged in descending order of frequency of appearance in this order. Similarly, with respect to card-2 to card-T, card-x (1≤x≤T) and Y cards (i.e., card-i x-1 、card-i x-2 , ..., and card-i x-Y ) is associated with.

[0095] In the case of k=3, for each of T types of cards, the table creation unit 51 extracts Y types of three-card sets that include the card and have a high appearance frequency, and associates the card with two-card sets obtained by excluding the card from the extracted three-card sets, thereby creating a first table. card-1 is associated with Y two-card sets (i.e., cardset-i(2) 1-1、cardset-i(2) 1-2 , ..., and cardset-i 1-Y ) is associated. cardset-i(k-1) 1-1 represents one of the combinations of k-1 cards, and cardset-i(k-1) 1-1 、cardset-i(k-1) 1-2 , ..., and cardset-i(k-1) 1-Y are different k-1 card sets. For example, cardset-i(2) 1-1 、cardset-i(2) 1-2 , ..., and cardset-i(2) 1-Y It is [(card-9,card-111)], [(card-2,card-842)], ..., and [(card-2,card-3)]. From card-1 and cardset-i(2) 1-1 The three card sets are composed of card-1 and cardset-i(2) 1-2 The three card sets, ..., and the three card sets consisting of card-1 and cardset-i(2) 1-Y The three card sets formed are the three card sets including card-1 arranged in descending order of frequency of appearance in this order. Similarly, with respect to card-2 to card-T, card-x (1≤x≤T) and Y cards (i.e., cardset-i(2) x-1 、cardset-i(2) x-2 , ..., and cardset-i(2) x-Y ) is associated with.

[0096] In the case of k=4 to D, similar to the case of k=3, for each of T types of cards, the table creation unit 51 extracts a set of k cards of Y types that includes the card and has a high appearance frequency, and associates the card with a set of k-1 cards obtained by excluding the card from the extracted set of k cards, thereby creating a first table. For example, in the case of k=40, as Figure 8 As shown, card-1 and Y cards (i.e., cardset-i(39) 1-1 、cardset-i(39) 1-2 , ..., and cardset-i(39) 1-Y ) is associated with card-1 and cardset-i(39) 1-1 A set of 40 cards, consisting of card-1 and cardset-i(39) 1-240 cards consisting of a card set, ..., and card-1 and cardset-i(39) 1-Y The 40-card set is the 40-card set including card-1 arranged in descending order of frequency of appearance in this order. Similarly, for card-2 to card-T, card-x (1≤x≤T) and Y cards (i.e., cardset-i(39) x-1 、cardset-i(39) x-2 , ..., and cardset-i(39) x-Y ) is associated with.

[0097] The input accepting unit 61 accepts input of one of T card types (card-α) and the number of cards (β) that make up the recommended deck (which is the card combination that includes that card) via user operation on the electronic device 30. In this embodiment, the game server 20 provides the game to the electronic device 30 on which the game app has been activated, and the display 33 of the electronic device 30 displays the game screen. The game screen provided by the game server 20 includes a first input accepting screen for obtaining media recommendation information from the media recommendation server 10. This media recommendation information is information related to the recommended media. The first input accepting screen is a screen for accepting input via user operation on the electronic device 30 (such as a touch on the display 33). The input accepting unit 61 accepts input of the card type (card-α) and the order (β) of the recommended deck (card combination) via the first input accepting screen. The electronic device 30 transmits the information accepted by the input accepting unit 61 to the media recommendation server 10. For ease of explanation, the number of cards that make up the card combination will be referred to as the order of the card combination. For example, a three-card set has an order of three.

[0098] The medium input unit 53 receives the information accepted by the input acceptance unit 61 from the electronic device 30 , and accepts input of the type of card card-α and the order β of the recommended card set.

[0099] The media recommendation unit 54 determines the Y cards or card combinations associated with card-α received by the media input unit 53 as first media recommendation information by referring to the first table where k=β in the first table stored in the table storage unit 52. The media recommendation unit 54 transmits the determined first media recommendation information to the electronic device 30.

[0100] The recommended media display unit 62 generates a first recommended media screen based on the first media recommendation information received from the media recommendation server 10, and displays the first recommended media screen on the display 33. The first recommended media screen is a screen for presenting the Y cards or card combinations associated with card-α to the user. In one example, the first recommended media screen is a screen for presenting the Y cards or card combinations to the user in descending order of frequency of occurrence. In another example, the first recommended media screen is a screen for presenting the most frequently appearing card or card combination to the user, and presenting the second most frequently appearing card or card combination to the user in response to a user operation.

[0101] Next, the media recommendation system 1 in the second embodiment of the present invention will be described. The media recommendation system 1 is configured to be able to recommend cards to be added to the card combination based on the card combination in the game in this embodiment to constitute at least a part of the card pile. The media recommendation system 1 can output a card card-α (1≤α≤T) that the user of the electronic device 30 wants to add to the β card set as recommendation information. The functional block diagram of the media recommendation system 1 in the second embodiment of the present invention is the same as Figure 6 The following description will focus on the differences from the first embodiment.

[0102] The table creation unit 51 uses a set of k cards (2≤k≤D-1) of Z types with high frequency of appearance within the deck and a set of k+1 cards of Z types, pre-extracted by the media extraction system 4. For each of the pre-extracted k card sets of Z types, the table creation unit 51 generates a set of k+1 cards of T types by connecting the individual cards included in the card set. Thus, the table creation unit 51 generates Z*T sets of k+1 cards. For each of the pre-extracted k card sets of Z types, the table creation unit 51 calculates the frequency of appearance of each of the generated sets of k+1 cards of T types by referencing the pre-extracted k+1 card sets of Z types with high frequency of appearance. For each of the pre-extracted k card sets of Z types, the table creation unit 51 further extracts a predetermined number (Y) of k+1 card sets with high frequency of appearance from the generated k+1 card sets. The table creation unit 51 creates a second table by associating each of the pre-extracted k card sets of Z types with the cards to be added to generate the k+1 card sets with higher frequency of appearance. In a preferred example, for each of the pre-extracted k card sets of Z types, the generated k+1 card sets are sorted in descending order of frequency of appearance, and Y k+1 card sets are further extracted in order from the card sets with higher frequency of appearance.

[0103] The table storage unit 52 stores the second table created by the table creation unit 51 in the storage device 14 .

[0104] Figure 9 An example second table generated by the table creation unit 51 in the second embodiment is shown. The key (primary key) in the second table is a set of k cards of Z types extracted in advance by the media extraction system 4. Figure 9 In the example shown, the table creation unit 51 creates a second table for each of k=2 to D-1(39) by further extracting (k+1) card sets of Y types in descending order of appearance frequency for each of the k card sets of Z types.

[0105] When k=2, the table creation unit 51 creates a second table by extracting three card sets of Y types with high appearance frequencies, including the two card sets, for each of the two card sets of Z types, and associating the two card sets with the cards to be added to generate the extracted three card sets.

[0106] cardset-e(k)-1, cardset-e(k)-2, ..., and cardset-e(k)-Z are respectively a set of k cards of Z types pre-extracted by the media extraction system 4. cardset-e(2)-1 and Y cards (i.e., card-e 1-1 、card-e 1-2 , ..., and card-e 1-Y ) is associated with card-e(k) 1-1 Represents one of card-1 to card-T, and card-e(k) 1-1 、card-e(k) 1-2 , ..., and card-e(k) 1-Y They are different cards. 1-1 Three card sets are formed, consisting of cardset-e(2)-1 and card-e 1-2 The three card sets, ..., and the cardsets-e(2)-1 and card-e 1-Y The three card sets formed are the three card sets including cardset-e(2)-1 arranged in descending order of frequency of appearance in this order. Similarly, regarding cardset-e(2)-2 to cardset-e(2)-Z, cardset-e(2)-x (1≤x≤Z) and Y cards (i.e., card-e(2) x-1 、card-e(2) x-2 , ..., and card-e(2) x-Y ) is associated with.

[0107] In the case of k=3 to D-1, similar to the case of k=2, for each of the k card sets of Z types, the table creation unit 51 extracts k+1 card sets of Y types that include the k card set and have a high frequency of appearance, and associates the k card sets with the cards to be added to generate the extracted k+1 card sets, thereby creating a second table. For example, in the case of k=39, as Figure 9 As shown, cardset-e(39)-1 and Y cards (i.e., card-e(39) 1-1 、card-e(39) 1-2 , ..., and card-e (39) 1-Y ) is associated with cardset-e(39)-1 and card-e(39) 1-1 A 40-card set consisting of cardset-e(39)-1 and card-e(39) 1-2 The 40-card set, ..., and the cardset-e(39)-1 and card-e(39) 1-Y The 40-card set is a 40-card set including cardset-e(39)-1 arranged in descending order of frequency of appearance in this order. Similarly, regarding cardset-e(39)-2 to cardset-e(39)-39, cardset-e(39)-x (1≤x≤Z) and Y cards (i.e., card-e(39) x-1 、card-e(39) x-2 , ..., and card-e (39) x-Y ) is associated with.

[0108] The input acceptance unit 61 accepts input of a set of γ cards constituting at least a portion of the deck of cards via user operation on the electronic device 30. In this embodiment, the game screen provided by the game server 20 includes a second input acceptance screen for obtaining media recommendation information from the media recommendation server 10, which is information related to the recommended media. The second input acceptance screen is a screen for accepting input via input operation on the electronic device 30 (such as a touch on the display 33). The input acceptance unit 61 accepts input of γ cards constituting the set of γ cards via the second input acceptance screen. The electronic device 30 sends the information accepted by the input acceptance unit 61 to the media recommendation server 10.

[0109] The media input unit 53 receives the information accepted by the input acceptance unit 61 from the electronic device 30 , and accepts input of γ cards constituting the set of γ cards.

[0110] The media recommendation unit 54 determines whether the set of γ cards received by the media input unit 53 is a primary key in the second table with k=γ stored in the table storage unit 52. If the set of γ cards received by the media input unit 53 is a primary key, the media recommendation unit 54 determines the Y cards associated with the set of γ cards as the second media recommendation information. If the set of γ cards is not a primary key, the media recommendation unit 54 determines that no recommended information exists as the second media recommendation information. The media recommendation unit 54 transmits the determined second media recommendation information to the electronic device 30.

[0111] The recommended media display unit 62 generates a second recommended media screen based on the second media recommendation information received from the media recommendation server 10, and displays the second recommended media screen on the display 33. The second recommended media screen is a screen for presenting Y cards associated with a set of γ cards to the user. In one example, the second recommended media screen is a screen for presenting Y cards arranged in descending order of frequency of appearance to the user. In another example, the second recommended media screen is a screen for presenting the user with the card with the highest frequency of appearance, and presenting the user with the card with the second highest frequency of appearance in response to the user's operation. In the absence of recommendation information as the second media recommendation information, the second recommended media screen indicates that there are no cards that can be recommended.

[0112] Next, the media recommendation system 1 according to the third embodiment of the present invention will be described. The table creation unit 51 creates the first table created in the first embodiment and the second table created in the second embodiment. The table storage unit 52 stores the first and second tables created by the table creation unit 51 in the storage device 14.

[0113] The game screen provided by the game server 20 and displayed on the display 33 includes a first input acceptance screen and a second input acceptance screen. The input acceptance unit 61 accepts input of the card type card-α and the card order β in the recommended card set via the first input acceptance screen. The input acceptance unit 61 accepts input of γ cards constituting the γ-card set via the second input acceptance screen. The electronic device 30 transmits the information received by the input acceptance unit 61 to the media recommendation server 10.

[0114] The media input unit 53, the media recommendation unit 54, and the recommended media display unit 62 have the same functions as those in the first embodiment. Therefore, in a case where the electronic device 30 sends information received via the first input acceptance screen to the media recommendation server 10, the recommended media display unit 62 generates a first recommended media screen based on the first media recommendation information received from the media recommendation server 10, and displays the first recommended media screen on the display 33. On the other hand, the media input unit 53, the media recommendation unit 54, and the recommended media display unit 62 have the same functions as those in the second embodiment. Therefore, in a case where the electronic device 30 sends information received via the second input acceptance screen to the media recommendation server 10, the recommended media display unit 62 generates a second recommended media screen based on the second media recommendation information received from the media recommendation server 10, and displays the second recommended media screen on the display 33.

[0115] Next, a table creation method of the media recommendation system 1 in the first embodiment of the present invention will be described. Figure 10 A first example table creation method of the media recommendation system 1 in this embodiment is shown.

[0116] In step (S) 101, the media extraction system 4 uses card deck information A, which includes a plurality of predetermined card deck information, to extract, for each of k = 2 to D, a set of k cards of Z types with high appearance frequencies (2 ≤ k ≤ D) included in card deck A. In step 102, the table creation unit 51 further extracts, for each of k = 2 to D, a set of k cards of a predetermined number (Y types) with high appearance frequencies for each card type from the set of k cards of Z types with high appearance frequencies extracted in step 101. In step 103, the table creation unit 51 creates a first table for each of k = 2 to D by associating, for each card type, a single card with a card or card combination obtained by excluding the single card from the further extracted set of k cards.

[0117] Next, a table creation method of the media recommendation system 1 in the second embodiment of the present invention will be described. Figure 11 A second example table creation method of the media recommendation system 1 in this embodiment is shown.

[0118] In step 201 , the media extraction system 4 uses deck information A including a plurality of predetermined deck information to extract Z types of card sets with high appearance frequencies in deck A for each of k=2 to D-1 (2≤k≤D-1).

[0119] In step 202, for each of k=2 to D-1, the table creation unit 51 combines the cards included in the card set with each of the k card sets of Z categories extracted in step 101, thereby generating k+1 card sets of T categories. Then, for each of the k card sets of Z categories extracted in step 201, the table creation unit 51 calculates the frequency of appearance of each of the generated k+1 card sets of T categories by referring to the k+1 card sets of Z categories with high appearance frequencies extracted in step 201. Then, for each of the k card sets of Z categories extracted in step 201, the table creation unit 51 further extracts a predetermined number (Y categories) of k+1 card sets with high appearance frequencies from the generated k+1 card sets. In step 203 , the table creation unit 51 creates a second table for each of k=2 to D-1 by associating each of the k card sets of Z types extracted in step 201 with the cards to be added to generate a further extracted k+1 card set with a high appearance frequency.

[0120] Next, preferred examples of the first table and the second table in the first to fifth embodiments will be described. In a preferred example, a hash index is implemented as a database index for the first table and the second table. The hash index can be easily implemented, for example, by using a key-value store such as Redis.

[0121] In the first and second tables of this example, the keys are stored as 32-bit or 64-bit integers and therefore always have a fixed length, and the values ​​are always limited to a maximum of Y elements and therefore always have a fixed length. The upper limit of the number of rows in the first table is T, and the upper limit of the number of elements included in the values ​​of the first table is K. Therefore, for each of k=2 to D, the size of the first table is T×K. In the case of D=40, Z=10000, T=1000, and Y=10, the table creation unit 51 converts the 40×1000 data table into a 40×1000×10 index. The upper limit of the number of rows in the second table is Z, and the upper limit of the number of elements included in the values ​​of the second table is K. Therefore, for each of k=2 to D-1, the size of the second table is Z×K. In the case of D=40, Z=10000, and Y=10, the table creation unit 51 converts the 40×10000 data table into a 40×1000×10 index. Since the hash index can be implemented using a memory of approximately 60 megabytes, the media recommendation system 1 can perform recommendation processing in an in-memory manner.

[0122] Next, a media extraction system 4 which is a technology disclosed in Japanese Patent Application No. 2019-041261 will be described below.

[0123] The media extraction system 4 includes a control device 70 and multiple computing devices 80, which are connected to a network so as to be able to communicate with each other. The control device 70 and the multiple computing devices 80 each have a structure similar to that of a typical server or PC. The control device 70 includes a processor 71, an input device 72, an output device 73, a storage device 74, and a communication device 75. These various components are connected via a bus 76. The computing devices 80 each include a processor 81, an input device 82, an output device 83, a storage device 84, and a communication device 85. These various components are connected via a bus 86. The processors 71 and 81 preferably include a GPU for general-purpose processing that does not include graphics rendering. Note that in some cases, multiple k-card sets extracted or compiled by either the control device 70 or the computing device 80 are referred to as multiple types of k-card sets to distinguish them from a single k-card set.

[0124] The storage device 74 of the control device 70 stores card pile information A. The storage device 74 also stores the types T of all cards. The control device 70 controls the parallel processing of the various operations of the multiple computing devices 80, thereby distributing the data to be processed by each of the multiple computing devices 80 and causing the processing to be executed. Card pile information A is stored in the storage device 84 of each of the s computing devices 80. In one example, card pile information A is extracted from the card pile information stored in the game log based on at least one of the win rate and usage rate of the card pile associated with the card pile information. For example, card pile information A is created by extracting card pile information with a win rate exceeding 50% from the card pile information stored in the game log. In one example, card pile information A is extracted from the card pile information stored in the game log based on the number of user IDs associated with the card pile information. In one example, card pile information A and the types T of all cards are stored in a CSV file. In one example, the game server 20 or the control device 70 extracts and creates card pile information A.

[0125] Conventionally, when constructing a deck by selecting 40 cards from 1,000 card types, it is difficult to exhaustively extract k sets of cards (where k represents an arbitrary number) that appear frequently, such as frequently occurring combinations of two cards, three cards, four cards, ..., and 39 cards. In particular, it is difficult to extract sets of 10 to 30 cards that appear frequently. On the other hand, it is relatively easy to calculate frequently occurring single cards, as well as frequently occurring sets of two cards, 39 cards, and 40 cards.

[0126] The media extraction system 4 is configured to be able to perform addition inference processing and subtraction inference processing. These types of processing are realized by the processors of the control device 70 and the calculation device 80 respectively executing programs.

[0127] In the additive inference process, the media extraction system 4 extracts a single card with a high appearance frequency, and then uses the information of the extracted card and the deck information A to extract a set of two cards with a high appearance frequency. Furthermore, the media extraction system 4 uses the information of the extracted two card sets and the deck information A to extract a set of three cards with a high appearance frequency. In this way, the media extraction system 4 extracts a combination of cards with a high appearance frequency, ranging from a single card set to a set of 20 cards (D / 2 cards).

[0128] In the subtraction inference process, the media extraction system 4 extracts a set of 40 cards with a high appearance frequency, and then extracts a set of 39 cards with a high appearance frequency by using the information of the extracted 40 cards and the card pile information A. Furthermore, the media extraction system 4 extracts a set of 38 cards with a high appearance frequency by using the information of the extracted 39 cards and the card pile information A. In this way, the media extraction system 4 extracts a combination of cards with a high appearance frequency from the set of 40 cards to the set of 21 cards (D / 2 + 1 card).

[0129] The media extraction system 4 in this embodiment presupposes that in a card game, a small number of specific cards or card combinations account for the majority of the appearance frequency, that is, there is a high frequency unevenness (Zipf structure).

[0130] The media extraction system 4 in this embodiment achieves exhaustive extraction of combinations by filtering card combination candidates step by step and recursively performing reasoning, while also suppressing combinatorial explosion by utilizing the Zipf structure of the data. In additive reasoning, when the media extraction system 4 filters to retain only frequently occurring combinations, as the order of reasoning advances, the diversity of the elements that make up the combinations is lost, and only combinations derived from specific combinations remain. As a result, the Zipf structure is rapidly lost, the frequency gradient disappears, and filtering no longer works effectively. Therefore, in the media extraction system 4 of this embodiment, global filtering is introduced to maintain the Zipf structure regardless of the order.

[0131] Figure 12 The additive inference processing of the media extraction system 4 for extracting a combination of k cards is described, where k represents an arbitrary number. The storage device 74 of the control device 70 stores the data corresponding to X. k Data related to k types of card sets. This data is related to X k Data related to sets of k cards of different types are arranged in order; for example, the data has kThe data structure is a one-dimensional array in which data related to a set of k cards of different types are arranged in sequence. Note that although for ease of explanation, this specification describes the case of sending or receiving cards or card combinations, it should be understood that this refers to sending or receiving data related to cards or card combinations.

[0132] The control device 70 disrupts X k The control device 70 divides the shuffled k card set into s pieces (step 302), and divides the divided X pieces into s pieces. k / s types of card combinations are distributed and sent to each of the s servers (step 303).

[0133] Each computing device 80 receives X from the control device 70 k / s types of k card sets, for X k For each of the k card sets of s types, the occurrence frequency of each of the received k card sets is calculated by referring to the card pile information A stored in the storage device 24 (step 304). Each calculation device 80 extracts k card sets with high occurrence frequencies that meet the first criterion through local filtering (step 305), and sends these k card sets to the control device 70 (step 306).

[0134] The control device 70 aggregates the k card sets of multiple types received from the computing devices 80 (step 307). The control device 70 performs a global filtering process to cluster the aggregated k card sets of multiple types into multiple clusters, and extracts, for each cluster, a k card set with a high frequency of appearance that meets the second criterion (step 308).

[0135] The control device 70 determines whether to extract a combination of cards that is one more than the number of cards (step 309). If it is determined in step 309 that no card combination will be extracted, the additive inference process is terminated. If it is determined in step 309 that a card combination will be extracted, the control device 70 generates a set of k+1 cards that is a combination of cards that is one more than the number of cards (step 310). Assuming that X is extracted in step 308, k 'K card sets of different types, the control device 70 combines all types of cards included in the card set into X k ' From each of the k card sets of one type, a set of k+1 cards of T types is generated (step 310). As described above, the control device generates a set of k+1 cards of T types from a set of k cards of one type. As a result, the control device 70 generates X k '×T types (X k+1Then, the control device 70 can extract the k+1 card set by executing steps 301 to 310.

[0136] Next, the local filtering process will be described. The first criterion is: when k card sets are sequentially selected from k card sets with higher appearance frequencies, k card sets are selected until the sum of the appearance frequencies of the k selected card sets reaches a specified ratio of the sum of the appearance frequencies of the k card sets calculated by the calculation device 80 in step 304.

[0137] In one example, in step 304, each computing device 80 calculates the X included in the card pile included in the card pile information A. k / s types of k card sets of each number, and rearrange these X in descending order of the calculated number k In this case, for example, each calculation device 80 rearranges the X cards in descending order of the calculated number by using a one-dimensional array data structure. k K card sets of s types are stored in the storage device 24. The counts calculated for each k card set correspond to the respective frequencies of appearance of the k card sets within the deck included in the deck information A. Each calculation device 80 calculates the sum α1 of all calculated counts. Each calculation device 80 selects k card sets in descending order of the calculated counts and calculates the sum of the counts of the selected k card sets as α2. Assuming that the prescribed ratio is T1 (0 ≤ T1 ≤ 1), the k card sets selected by each calculation device 80 before α2 reaches α1 × T1 are used as the k card sets that meet the first criterion.

[0138] Preferably, in the local filtering process, before performing the extraction process using the first criterion, each computing device 80 extracts a set of k cards by reflecting the deck construction rules in the game, thereby reducing the number of combinations. Examples of extraction that reflects the deck construction rules include extraction that excludes combinations of cards that are not allowed in a specific battle, and extraction that excludes types of cards that cannot be included based on class.

[0139] Next, the global filtering process will be described. In step 307, the control device 70 assembles the k card sets received from each computing device 80 by combining them and sorting and arranging the results. In one preferred embodiment, the control device 70 rearranges the k card sets of multiple categories in descending order of frequency of occurrence. In this case, for example, the control device 70 uses a one-dimensional array data structure to store the k card sets of multiple categories, rearranged in descending order of frequency of occurrence, in the storage device 14.

[0140] In the clustering in step 308, the control device 70 clusters the k card sets of Q types that have been aggregated into clusters (subsets) of P types. Note that at this time, P < Q. In a preferred example, as the clustering algorithm, the control device 70 uses agglomerative hierarchical clustering. As the algorithm used for agglomerative hierarchical clustering, a known algorithm is used. In the case of using this agglomerative hierarchical clustering, the control device 70 starts clustering from the state where each cluster is composed of k card sets located at adjacent positions in the array, with the array of k card sets sorted in descending order of frequency. The control device 70 sequentially combines clusters when multiple adjacent k card sets include no less than a specified number or a specified proportion of common cards according to the clustering level, thereby generating a cluster layer. The control device 70 repeats recursively applying this process until the number of clusters becomes a specified number or until there is no cluster group that includes no less than a specified proportion of common cards. The control device 70 divides the k card sets of multiple types aggregated in step 308 into multiple clusters by using the clusters in the layer where the clustering algorithm has stopped or in a specified layer.

[0141] In step 308, for each cluster, the control device 70 calculates the sum of the occurrence frequencies of the k card sets of multiple types set in the cluster. The control device 70 calculates the cluster frequency ratio for each cluster by using the calculated sum of the occurrence frequencies. The second criterion is: when selecting k card sets in descending order of occurrence frequency, select k card sets until the sum of the occurrence frequencies of the selected k card sets reaches a specified value based on the cluster frequency ratio.

[0142] Figure 13 Describe the global filtering process. Figure 13 The left diagram in shows the state where the k card sets of multiple types before the global filtering process have been sorted in descending order of occurrence frequency and have been clustered into multiple clusters. Figure 13 The right diagram in shows the state where the k card sets of multiple types after the global filtering process have been sorted in descending order of occurrence frequency and are still clustered into multiple clusters. Preferably, multiple k card sets are set in one cluster; however, only one k card set can be set in one cluster. The multiple k card sets set in the same cluster include no less than a specified number of common cards.

[0143] In one example, the control device 70 calculates the sum β1 of all the occurrence frequencies of the k card sets of multiple types before the global filtering process. The control device 70 sets M individual clusters in a distinguishable manner, such as CL1, CL2,..., CL MThe control device 70 calculates the total appearance frequencies γ1, γ2, ..., γ of the k-card sets set in each cluster. M For example, the control device 70 calculates the total sum β1 (=γ1+γ2…γ M ) of each cluster γ1, γ2, ..., γ M , calculating the cluster frequency ratio as the ratio of each cluster. The control device 70 pre-sets T2 as the target number of card combinations after global filtering, and sets the product of T2 and the cluster frequency ratio for each cluster as the upper limit of the appearance frequency of the k card sets in that cluster. For cluster CL1, while selecting k card sets in descending order of appearance frequency, the control device 70 calculates the sum of the frequencies of the selected k card sets as δ1. Assuming that the upper limit of the appearance frequency of the k card sets set in cluster CL1 is UL1, the k card sets selected by the control device 70 until δ1 = UL1 is satisfied are used as the k card sets that meet the second criterion.

[0144] Threshold T2 corresponds to Z as described above. The media extraction system 4 (control device 70) can determine thresholds T1 and T2 used in the filtering process by using the results of the immediately preceding inference process (i.e., the results of extracting the k-1 card set in the additive inference process). The media extraction system 4 (control device 70) maintains exhaustiveness with respect to the total number by exploiting the high unevenness between the frequencies of individual combinations in the Zipf structure while significantly reducing the number of card combinations.

[0145] Next, the initial process in the additive inference process will be described. As described above, in the initial process in the additive inference process, the media extraction system 4 extracts a single card with a high appearance frequency.

[0146] The control device 70 shuffles the order of T types (X1 types) of individual cards (step 301). The control device 70 divides the shuffled individual cards into s types (step 302), and distributes and sends the divided T / s types of cards to each of the s servers (step 303).

[0147] Upon receiving T / s types of cards from the control device 70, each calculation device 80 calculates the frequency of appearance of each received card by referring to the card deck information A stored in the storage device 24 (step 304). Each calculation device 80 extracts cards with high frequency of appearance that meet predetermined criteria through local filtering (step 305), and sends the extracted cards to the control device 70 (step 306).

[0148] The control device 70 aggregates the multiple cards received from the computing devices 80 (step 307). The control device 70 skips steps 308 and 309 and generates a two-card set consisting of a number of cards greater than one (step 310). Assuming that X1 types of cards are aggregated in step 307, the control device 70 combines all the cards of each type included in the card set with cards of X1 types, thereby generating a two-card set of T types (step 310). Since the control device generates two-card sets of T types from a single card in this manner, the control device 70 ultimately generates two-card sets of X1×T types (X2 types). Note that the control device 70 can be configured to execute step 308 by adjusting the threshold T2 used in the filtering process.

[0149] Then, the control device 70 can extract a set of k cards by repeatedly executing steps 301 to 310 , where k represents an arbitrary number.

[0150] A prescribed standard similar to the first standard is that, when selecting cards in descending order of appearance frequency, cards are selected until the sum of the appearance frequencies of the selected cards reaches a prescribed ratio of the sum of the appearance frequencies of the cards calculated by the calculation device 80 in step 304.

[0151] Figure 14 The subtraction reasoning process using the media extraction system 4 for extracting a combination of k cards is described, where k represents an arbitrary number.

[0152] The control device 70 disrupts X k The control device 70 divides the shuffled k card set into s pieces (step 402), and divides the divided X pieces into s pieces. k / s types of card combinations are distributed and sent to each of the s servers (step 403).

[0153] Each computing device 80 receives X from the control device 70 k / s types of k card sets, for X k For each of the k card sets of s types, the calculation device 80 calculates the frequency of appearance of each of the k card sets received by referring to the card deck information A stored in the storage device 24 (step 404). Each calculation device 80 extracts k card sets with high frequency of appearance that meet the first criterion through local filtering (step 405), and sends the extracted k card sets to the control device 70 (step 406).

[0154] The control device 70 aggregates the k card sets of multiple types received from the computing devices 80 (step 407). The control device 70 performs a global filtering process to cluster the aggregated k card sets of multiple types into multiple clusters, and extracts, for each cluster, a card combination with a high frequency of occurrence that meets the second criterion (step 408).

[0155] The control device 70 determines whether to extract a combination of cards that is further reduced by 1 (step 409). If it is determined in step 409 that no card combination will be extracted, the subtraction reasoning process is terminated. If it is determined in step 409 that a card combination will be extracted, the control device 70 generates a set of k-1 cards that is a combination of cards that is reduced by 1 (step 410). Assuming that X is collected in step 408 k 'K card sets of different types, the control device 70 will form X k A single card in a set of k cards of one type is excluded from each of the k card sets, thereby generating a new set of k-1 cards of k types. As described above, the control device 70 generates a set of k-1 cards of k types from a set of k cards of one type. As a result, the control device 70 generates X k '×k types (X k-1 A set of k-1 cards of different types.

[0156] The local filtering process and the global filtering process are similar to those in the case of the additive inference process. Note that in the subtractive inference process, the thresholds T1 and T2 are determined similarly to the case of the additive inference process, and the threshold T2 corresponds to Z described above.

[0157] Next, the initial processing in the subtraction inference process will be described. As described above, in the initial processing in the subtraction inference process, the media extraction system 4 extracts a 39-card set from a 40-card set.

[0158] The control device 70 shuffles the arrangement order of the 40 card sets of the number of users included in the card pile information A (step 401). The control device 70 divides the shuffled 40 card sets into s (step 402), and divides the divided X cards into s. 40 / s types of card combinations are distributed and sent to each of the s servers (step 403).

[0159] Each computing device 80 receives X from the control device 70 40 / s 40-card deck of different types, targeting X 40For each of the 40 card sets of each type, the calculation device 80 calculates the frequency of occurrence of each of the received 40 card sets by referring to the card pile information A stored in the storage device 24 (step 404). Each calculation device 80 extracts k card sets with high frequency of occurrence that meet the first criterion through local filtering (step 405), and sends the extracted k card sets to the control device 70 (step 406). However, each calculation device 80 may skip step 405.

[0160] The control device 70 collects the 40 card sets of the plurality of types received from the respective computing devices 80 (step 407). The control device 70 skips steps 408 and 409, and the control device 70 generates a 39-card set that is a combination of cards with the number of cards less than 1 (step 410). Assume that X is collected in step 407. 40 'A set of 40 cards of different types, the control device 70 will select a single card from the cards constituting the set of 40 cards from X 40 ' Each of the 40 card sets of each type is excluded, thereby generating a new set of 39 cards of 40 types. Since the control device 70 generates a set of 39 cards of 40 types from a set of 40 cards of one type in this way, as a result, the control device 70 generates X 40 '×40 types(X 39 Note that the control device 70 may be configured to perform step 408 by adjusting the threshold value T2 used in the filtering process.

[0161] Then, the control device 70 can extract a set of k cards by repeatedly executing steps 401 to 410 , where k represents an arbitrary number.

[0162] As described above, by performing both local and global filtering, the media extraction system 4 can execute additive and subtractive reasoning in real time, even in situations where combinatorial explosions occur. Specifically, by employing two types of filtering, the number of card combination patterns can be kept within a specific range while maintaining the trend of the original set in terms of global variation, thereby suppressing combinatorial explosions. Furthermore, by combining these two types of filtering with progressive filtering using additive or subtractive reasoning, k sets of cards can be extracted without discarding mathematically significant combinations. Thus, using the above configuration, the media extraction system 4 can exhaustively extract any number of card combinations even in data processing that could potentially lead to combinatorial explosions.

[0163] In one example, card pile information A is created by extracting card pile information having a winning rate exceeding a specified winning rate from the card pile information stored in the game log. In this case, the media extraction system 4 can exhaustively extract any number of card combinations with high frequency of appearance used by highly skilled users.

[0164] The storage device 74 of the control device 70 stores Z types of data for each of the 2 to D sets of cards that appear frequently. In a preferred embodiment, the media extraction system 4 periodically performs batch processing to extract Z types of data for each of the 2 to D sets of cards that appear frequently and stores the data in the storage device 74.

[0165] When creating the first table or the second table, the media recommendation server 10 receives Z types of data for each of 2 to D card sets with high appearance frequencies from the media extraction system 4 .

[0166] When the media extraction system 4 is represented as a functional block, it includes a storage unit and an extraction unit. The storage unit stores card pile information A. The extraction unit, implemented by utilizing the information processing of the control device 70 and the computing device 80 described above, can calculate the frequency of occurrence of each of m card sets (2≤m≤D) by referring to the card pile information A stored in the storage unit, and can extract m card sets with high frequency of occurrence that meet a specified standard. Furthermore, the extraction unit can generate multiple m+1 card sets by combining cards included in the card set with each of the extracted m card sets, or can generate multiple m-1 card sets by extracting a single card from each of the extracted m card sets. In a preferred embodiment, the extraction unit periodically and individually extracts 2 to D card sets with high frequency of occurrence that meet a specified standard, and the media recommendation server 10 periodically creates the first table and the second table based on the extracted 2 to D card sets with high frequency of occurrence.

[0167] Next, the main operations and advantages of the media recommendation system 1 in the embodiment of the present invention will be described.

[0168] Due to the continued use of recent DCCGs for several years, the number of items available to users in the game sometimes exceeds thousands. In such DCCGs, users often obtain unintended cards, and the utility of individual cards often changes depending on the game environment (such as the properties of newly added cards). In these situations, it is difficult for users, especially new users, to identify which types of cards are available in the game and which combinations are available. It is even difficult to know the existence of cards they do not own. Therefore, it is very useful for users to know which types of strong card piles can be formed from the cards they currently own, or which cards to add to form strong card combinations from their current cards. However, it is difficult to apply existing information recommendation technologies used on e-commerce websites (such as those for recommending products) to technologies used to recommend cards or card combinations suitable for users in DCCGs. Furthermore, when DCCGs include elements that lead to combinatorial explosion, it is impossible to obtain the card combinations that appear in the top Z frequency of occurrence in any given card combination.

[0169] In the embodiment, as described above, the media extraction system 4 exhaustively extracts combinations of any number of cards with a high frequency of appearance. The media recommendation system 1 generates at least a first table or a second table based on a set of k cards with a high frequency of appearance pre-extracted by the media extraction system 4. Based on the first table, the media recommendation system 1 recommends a set of N cards with a high frequency of appearance based on a single card specified by the user. Based on the second card, the media recommendation system 1 recommends a single card with a high frequency of appearance based on the N cards specified by the user. As described above, in the embodiment, the media extraction system 4 can exhaustively extract combinations of any number of cards that may lead to a combinatorial explosion. In addition, the media recommendation system 1 can learn combinations of cards with a high frequency of appearance and output recommendation information indicating cards to be added, etc. In addition, since the media recommendation system 1 performs processing by using a table, recommendation information indicating cards to be added, etc. can be output at a high speed.

[0170] Furthermore, in a preferred embodiment, the media extraction system 4 exhaustively extracts combinations of any number of cards that appear frequently, as used by highly skilled users. This configuration allows the media recommendation system 1 to rapidly recommend cards or card combinations, ensuring that the card combinations used by highly skilled users are formed from the cards currently in their possession. This allows for the rapid recommendation of cards or card combinations that are suitable for the user. For example, "deck building know-how," which until now relied on external resources, can be provided to users as formal information.

[0171] The media recommendation system 1 is a system that implements high-speed card recommendations based on 1-to-N and N-to-1 relationships by using two types of tables. Therefore, this technology can be applied to a general technology for learning trends in combinations of limited content and recommending information.

[0172] Furthermore, in a preferred example, the first table and the second table are tables whose number of rows and elements is limited, and are tables that implement hash indexes (hash tables). In this case, with D=40, Z=10000, T=1000, and Y=10 as a practical assumption, the total size of both the first table and the second table is approximately 60 megabytes. Therefore, these tables can have a size that can be loaded in each of the media recommendation server 10 and the electronic device 30 in an in-memory manner, and implementation is relatively easy. Furthermore, with this configuration, since the information processing using the information recommendation unit 54 can be implemented as a process for searching a hash table with a computational complexity of O(1), the execution time can be made extremely short, which enables frequent information recommendations to be made in the game. Therefore, a small number of server devices constituting the media recommendation server 10 is sufficient. Note that the media recommendation system 1 can create the first table and the second table in approximately 20 hours.

[0173] In another preferred embodiment, the media extraction system 4 periodically performs batch processing to extract Z types of data for each of 2 to D frequently appearing card sets, and stores this data in the storage device 74. In this case, the media recommendation server 10 can also periodically perform batch processing to regularly update the first and second tables. With this configuration, the media recommendation system 1 can recommend cards or card combinations suitable for the user based on the current gaming environment.

[0174] Unless otherwise specifically stated, the above-described operations and advantages also apply similarly to other embodiments and other examples.

[0175] An embodiment of the present invention may be the above-mentioned media recommendation system 1 or media recommendation server 10 (media recommendation device). Another embodiment of the present invention may be a table creation system or table creation method for implementing the following information processing, in which at least any one of the first table and the second table included in the media recommendation system 1 is created. Another embodiment of the present invention may be a program for implementing the functions or information processing shown in the flowchart of the above-mentioned embodiment of the present invention, or a computer-readable storage medium storing the program. In addition, another embodiment of the present invention may be a method for implementing the functions or information processing shown in the flowchart of the above-mentioned embodiment of the present invention. In addition, another embodiment of the present invention may be a server capable of providing a computer with a program for implementing the functions or information processing shown in the flowchart of the above-mentioned embodiment of the present invention. In addition, another embodiment of the present invention may be a virtual machine for implementing the functions or information processing shown in the flowchart of the above-mentioned embodiment of the present invention.

[0176] An example application of an embodiment of the present invention will be described below.

[0177] In the first example application, the game server 20 provides a card pile construction screen for allowing the user to construct a card pile by selecting cards one by one from a plurality of cards owned by the user, and the electronic device 30 displays this card pile construction screen on the display 33 as the game progresses. In this example, the card pile construction screen also serves as a second input acceptance screen for obtaining media recommendation information.

[0178] The input accepting unit 61 accepts the cards selected by the user so far on the deck construction screen as input for a set of γ cards. The electronic device 30 transmits the information received by the input accepting unit 61 to the media recommendation server 10, and the media input unit 53 receives the information received by the input accepting unit 61 from the electronic device 30 to accept input for the γ cards that constitute the set of γ cards. The media recommendation unit 54 determines second media recommendation information using the second table with k = γ stored in the table storage unit 52. The recommended media display unit 62 generates a second recommended media screen based on the second media recommendation information received from the media recommendation server 10 and displays this second recommended media screen on the display 33. The second recommended media screen presents the user with Y cards associated with the set of γ cards. For example, the second recommended media screen is a screen in which a window presenting Y cards arranged in descending order of appearance frequency is displayed superimposed on the game screen provided by the game server 20.

[0179] With this configuration, when building a deck, the second table can be searched using the N cards selected so far by the user as a query, thereby obtaining candidates for the N+1th card. This allows for the recommendation of cards that are more suitable for the user. Furthermore, the card recommendation process can be implemented relatively easily.

[0180] In the second example application, the game server 20 provides a card generation screen for allowing the user to generate new cards that the user does not own, and the electronic device 30 displays the card generation screen on the display 33 as the game progresses. In this example, the card generation screen also serves as the first input acceptance screen for obtaining media recommendation information.

[0181] The input receiving unit 61 receives the cards (card-α1, card-α2, ..., card-α x ) input. The electronic device 30 transmits the information received by the input acceptance unit 61 to the media recommendation server 10, and the media input unit 53 receives the information received by the input acceptance unit 61 from the electronic device 30 to accept input for each card not owned by the user. The media recommendation unit 54 extracts D-1 card sets associated with each card not owned by the user, using the first table (k=D) stored in the table storage unit 52. The media recommendation unit 54 further extracts D-1 card sets consisting of only cards owned by the user from the extracted D-1 card sets, and their associated cards. It should be understood here that the associated cards are the cards (card types) associated in the first table and the cards not owned by the user for which input was accepted by the input acceptance unit 61. The media recommendation unit 54 sorts the D card sets consisting of the further extracted D-1 card sets and their associated cards in descending order of appearance frequency by referring to the D card sets with high appearance frequencies extracted by the media extraction system 4 and their appearance frequencies. The media recommendation unit 54 determines the D card sets consisting of D-1 card sets and their associated cards arranged in descending order of appearance frequency as the third media recommendation information and transmits the determined third media recommendation information to the electronic device 30 .

[0182] The recommended media display unit 62 generates a third recommended media screen based on the third media recommendation information received from the media recommendation server 10 and displays the third recommended media screen on the display 33. The third recommended media screen is a screen for presenting to the user new cards that the user does not own and the D-1 card sets associated with these cards, arranged in descending order of frequency of appearance. For example, the third recommended media screen is a screen in which a window for presenting new cards that the user does not own and the D-1 card sets associated with these cards, arranged in descending order of frequency of appearance, is displayed superimposed on the game screen provided by the game server 20.

[0183] With this configuration, when generating cards that the user does not own, the cards that the user does not own can be sorted in an order that is convenient for the user. This makes it possible to recommend cards that are more suitable for the user.

[0184] In the third example application, the game server 20 provides a screen for suggesting a single card to be modified for the deck constructed by the user, and the electronic device 30 displays the screen on the display 33 in response to the user's operation. In this example, the input acceptance unit 61 accepts input of a set of D cards constituting the deck constructed by the user without using the first input acceptance screen and the second input acceptance screen.

[0185] In this example, the media recommendation server 10 further includes a main control unit 55. When the information received by the input receiving unit 61 is transmitted from the electronic device 30 to the media recommendation server 10, the main control unit 55 generates D-1 card sets of D categories and transmits these D-1 card sets of D categories to the media input unit 53. The media input unit 53 then receives input for each of the D-1 card sets of D categories. The media recommendation unit 54 determines second media recommendation information for each of the D-1 card sets of D categories using the second table with k = D-1 stored in the table storage unit 52. For each of the D-1 card sets of D categories, the media recommendation unit 54 extracts D card sets consisting only of cards owned by the user from the D card sets of Y categories consisting of the D-1 card sets and their associated second media recommendation information (Y cards). The media recommendation unit 54 refers to the D card sets with high appearance frequencies extracted by the media extraction system 4 and their appearance frequencies, and further extracts a predetermined number of D card sets from all the extracted D card sets, which have a higher appearance frequency than the D card sets in the deck constructed by the user. The media recommendation unit 54 determines the D-1 card sets and their associated cards constituting the further extracted D card sets as fourth media recommendation information.

[0186] The recommended media display unit 62 generates a fourth recommended media screen based on the fourth media recommendation information received from the media recommendation server 10, and displays the fourth recommended media screen on the display 33. The fourth recommended media screen is used to present to the user a D-1 card set and a single card to be modified from the user-constructed deck. For example, the fourth recommended media screen is used to present to the user a D-1 card set that constitutes the most frequently appearing card set and a single card to be modified, and to present to the user a D-1 card set that appears next most frequently and a single card to be modified in response to a user operation.

[0187] The main control unit 55 can generate multiple types of τ card sets (2≤τ≤D-1) and pass these τ card sets to the media input unit 53, which can then accept input of each τ card set. Note that in this case, the media recommendation unit 54 also performs the above-mentioned processing using the second table. In addition, the main control unit 55 is configured to preferentially recommend the fourth media recommendation information determined when the value of τ is large.

[0188] With this configuration, when a suitable card combination can be obtained by replacing a single card from the deck constructed by the user, a suitable card modification can be suggested to the user. This makes it possible to recommend cards that are more suitable for the user.

[0189] In the fourth example application, the game server 20 provides a card pile construction screen for allowing the user to construct a card pile by selecting cards one by one from a plurality of cards owned by the user, and the electronic device 30 displays this card pile construction screen on the display 33 as the game progresses. In this example, the card pile construction screen also serves as the first input acceptance screen for obtaining media recommendation information.

[0190] The input accepting unit 61 accepts input of the card (card-α) initially selected by the user on the card stack construction screen. The electronic device 30 transmits the information received by the input accepting unit 61 to the media recommendation server 10, and the media input unit 53 receives the information received by the input accepting unit 61 from the electronic device 30 to accept the input of card-α. The media recommendation unit 54 determines first media recommendation information by using the first table where k=D stored in the table storage unit 52. The recommended media display unit 62 generates a first recommended media screen based on the first media recommendation information received from the media recommendation server 10 and displays the first recommended media screen on the display 33. The first recommended media screen is a screen that presents to the user a set of D-1 cards associated with card-α, arranged in descending order of frequency of appearance. For example, the first recommended media screen is a screen in which a window that presents the set of D-1 cards, arranged in descending order of frequency of appearance, is displayed in an overlay on the game screen provided by the game server 20.

[0191] With this configuration, it is possible to provide the user with information for constructing a meaningful deck including a specific card when building a deck. This makes it possible to recommend cards that are more suitable for the user.

[0192] Now, a modification of the embodiment of the present invention will be described. As long as there is no contradiction, the modification described below can be appropriately combined and applied to any embodiment of the present invention. The description will focus on the differences from the above-mentioned embodiment.

[0193] In one variation, the game server 20 includes the media recommendation server 10. In another variation, the media recommendation server 10 is not connected to a network and is configured to be able to perform P2P communication with the game server 20. In this case, the electronic device 30 receives the first media recommendation information or the second media recommendation information via the game server 20 by performing communication with the game server 20.

[0194] In a variation, the electronic device 30 is installed with a dedicated app for receiving and displaying in-game media recommendation information provided by the game server 20. When the app is activated, the electronic device 30 displays the first input acceptance screen or the second input acceptance screen according to user operation.

[0195] In one variation, the media recommendation system 1 includes a media recommendation server 10 and a game server 20. The media recommendation server 10 includes a table creation unit 51, a table storage unit 52, an input acceptance unit 61, a media recommendation unit 54, and a recommended media display unit 62. The media recommendation server 10 includes a display 13 as an output device 13, and displays an input acceptance screen on the display 13 when a special app is activated. The input acceptance screen includes a first input acceptance screen and a second input acceptance screen.

[0196] The input accepting unit 61 accepts input of the card type card-α and the card order β in the recommended deck via a first input accepting screen. The media recommendation unit 54 determines Y cards or card combinations associated with the card-α accepted by the input accepting unit 61 as first media recommendation information by referring to a first table with k = β in the first table stored in the table storage unit 52. The recommended media display unit 62 generates a first recommended media screen based on the first media recommendation information determined by the media recommendation unit 54 and displays this first recommended media screen on the display 13.

[0197] The input acceptance unit 61 accepts input of γ cards constituting a γ-card set via a second input acceptance screen. If the γ-card set accepted by the media input unit 53 is a primary key, the media recommendation unit 54 determines Y cards associated with the γ-card set as second media recommendation information. If the γ-card set is not a primary key, the media recommendation unit 54 determines that no recommended information exists as second media recommendation information. The recommended media display unit 62 generates a second recommended media screen based on the second media recommendation information determined by the media recommendation unit 54 and displays this second recommended media screen on the display 13.

[0198] In one variation, a single first table is provided that includes tables for k = 2 to D. In this case, the primary key is a combination of k and key. In another variation, a single second table is provided that includes tables for k = 2 to D-1. In this case, the primary key is a combination of k and key.

[0199] In a variation of the first embodiment, the input acceptance unit 61 accepts input of a single card (card-α) from T types of cards in response to a user operation on the electronic device 30 via a first input acceptance screen. The electronic device 30 sends the information accepted by the input acceptance unit 61 to the media recommendation server 10. The media input unit 53 receives the information accepted by the input acceptance unit 61 from the electronic device 30 to accept the input of the type of card card-α. The media recommendation unit 54 determines Y cards or card combinations associated with the card-α accepted by the media input unit 53 as first media recommendation information by referring to a predetermined first table in the first table stored in the table storage unit 52. The media recommendation unit 54 sends the determined first media recommendation information to the electronic device 30. In a preferred example, the predetermined first table in the first table is a first table with k=40.

[0200] In a variation of the second embodiment, the table creation unit 51 generates a plurality of k+1 card sets by combining cards obtained by excluding cards with an appearance frequency less than a specified value from the cards included in the card set for each of the k card sets of Z types extracted in advance. This configuration can reduce the amount of information processing.

[0201] In a variation of the second embodiment, when the media recommendation unit 54 determines that there is no recommendation information as the second media recommendation information, the following second input acceptance screen is displayed, which prompts the γ-1 card set included in the γ card set accepted by the media input unit 53 as a new input.

[0202] The above-described processes or operations can be freely modified as long as no contradictions arise in the processes or operations (such as a contradiction in which a particular step utilizes data that is not yet available at that step). Furthermore, the above-described examples are provided to illustrate the present invention, and the present invention is not limited to these examples. The present invention can be embodied in various forms as long as they do not deviate from the gist of the present invention.

[0203] Reference Signs List

[0204] 1 Media Recommendation System

[0205] 2 Network

[0206] 4 Media Extraction System

[0207] 10 Media Recommendation Server

[0208] 11, 21, 31, 71, 81 processors

[0209] 12, 22, 32, 72, 82 input devices

[0210] 13, 23, 33, 73, 83 output devices

[0211] 14, 24, 34, 74, 84 storage devices

[0212] 15, 25, 35, 75, 85 communication device

[0213] 16, 26, 36, 76, 86 buses

[0214] 20 Game Servers

[0215] 30 Electronic devices

[0216] 40 Game Screen

[0217] 41 cards

[0218] 42 First Card Set

[0219] 43 Game Field

[0220] 44 Second Card Set

[0221] 45 characters

[0222] 51 Table creation unit

[0223] 52 table storage unit

[0224] 53 Media Input Unit

[0225] 54 media recommendation units

[0226] 55 Main control unit

[0227] 61 Input receiving unit

[0228] 62 recommended media display units

Claims

1. A media recommendation system for recommending a media or a set of media in a game that progresses as a user selects a media from an owned media group configured to include D media and places the media on a game field, wherein the recommended media or set of media is a media or a set of media to be added to the game in order to construct at least a portion of the owned media group from media, wherein each media included in the owned media group is selected from T types of media included in a media set pre-set in the game, the media recommendation system comprising: an input accepting unit, configured to accept an input of a single medium from the T types of media; a table storage unit for storing a first table corresponding to each of k=2 to D, the first table being created by further extracting, for each category of media, a set of a prescribed number of media with a high frequency of appearance from a set of media consisting of k media with a high frequency of appearance included in the plurality of owned media groups, which is pre-extracted based on owned media group information related to a plurality of owned media groups used by a prescribed plurality of users, and associating, for each category of media, a single medium with a medium or a set of media obtained by excluding the medium from the further extracted set of media, where 2≤k≤D; and A media recommendation unit is capable of referring to a first table corresponding to the number of media constituting the media or the set of media to be recommended, and outputting the media or the set of media associated with the single media accepted by the input acceptance unit in the first table as recommendation information.

2. A table creation system for creating a table for recommending a medium or a set of media in a game that progresses as a user selects a medium from an owned media group constructed to include D media and places the medium on a game field, the recommended medium or set of media being a medium or a set of media to be added to the game in order to construct at least a part of the owned media group from media, the individual media included in the owned media group being selected from T types of media included in a media set preset in the game, in, The table corresponding to the respective numbers of media constituting the media or a set of media to be recommended is created by the following operation: from a set of media consisting of k media with high occurrence frequencies included in the multiple owned media groups, which is pre-extracted based on owned media group information related to the multiple owned media groups used by the specified multiple users, a set of a specified number of media with high occurrence frequencies is further extracted for each class of media, and for each class of media, a single medium is associated with the media or a set of media obtained by excluding the medium from the set of media further extracted, where 2≤k≤D.

3. A media recommendation device for recommending a media or a set of media in a game that progresses as a user selects a media from an owned media group configured to include D media and places the media on a game field, wherein the recommended media or set of media is a media or a set of media to be added to the game in order to construct at least a portion of the owned media group from media, wherein each media included in the owned media group is selected from T types of media included in a media set pre-set in the game, the media recommendation device comprising: an input accepting unit, configured to accept an input of a single medium from the T types of media; a table storage unit for storing a first table corresponding to each of k=2 to D, the first table being created by further extracting, for each category of media, a set of a prescribed number of media with a high frequency of appearance from a set of media consisting of k media with a high frequency of appearance included in the plurality of owned media groups, which is pre-extracted based on owned media group information related to a plurality of owned media groups used by a prescribed plurality of users, and associating, for each category of media, a single medium with a medium or a set of media obtained by excluding the medium from the further extracted set of media, where 2≤k≤D; and A media recommendation unit is capable of referring to a first table corresponding to the number of media constituting the media or the set of media to be recommended, and outputting the media or the set of media associated with the single media accepted by the input acceptance unit in the first table as recommendation information.

4. A table creation method for creating a table for recommending a medium or a set of media in a game that progresses as a user selects a medium from an owned media group configured to include D media and places the medium on a game field, the recommended medium or set of media being a medium or a set of media to be added to the game in order to construct at least a portion of the owned media group from the medium, wherein each medium included in the owned media group is selected from T types of media included in a media set preset in the game, the table creation method comprising the following steps: For each of k=2 to D, extracting a set of k media with high appearance frequencies included in the plurality of owned media groups based on owned media group information related to a plurality of owned media groups used by a plurality of users, where 2≤k≤D; For each of k=2 to D, from the extracted set of k media with high occurrence frequencies, further extract a set of a predetermined number of media with high occurrence frequencies for each type of media; as well as For each of k=2 to D, a table corresponding to the respective numbers of media constituting the media or set of media to be recommended is created by associating, for various classes of media, a single medium with a medium or set of media obtained by excluding that medium from the set of media further extracted.

5. A media recommendation system for recommending media in a game that progresses as a user selects a media from an owned media group configured to include D media and places the media on a game field, the recommended media being media to be added to a set of media in order to construct at least a portion of the owned media group from a set of media, wherein each media included in the owned media group is selected from a media set including T types of media pre-set in the game, the media recommendation system comprising: an input accepting unit for accepting an input of a set of k media constituting at least a part of the owned media group, where 2≤k≤D-1; A table storage unit for storing a second table corresponding to each of k=2 to D-1, wherein the second table is generated by the following operation: when using a set of media consisting of k media and a set of media consisting of k+1 media with high appearance frequencies included in the plurality of owned media groups extracted in advance based on owned media group information related to the plurality of owned media groups used by a predetermined plurality of users, the media included in the media set are respectively combined with the pre-extracted sets of media consisting of k media, thereby generating a plurality of sets of media consisting of k+1 media, and for each of the pre-extracted sets of k media, and calculating the occurrence frequencies of each of the generated sets of k+1 media with reference to a pre-extracted set of k+1 media, thereby further extracting a predetermined number of sets of k+1 media with high occurrence frequencies from the generated sets of k+1 media for each pre-extracted set of k media, and associating each pre-extracted set of k media with the media to be added in order to generate the further extracted set of k+1 media with high occurrence frequencies; A media recommendation unit that is capable of referring to a second table corresponding to the number of media constituting the set of media accepted by the input acceptance unit, and outputting as recommendation information the media that is associated with the set of media consisting of k media accepted by the input acceptance unit in the second table and is to be added in order to generate a set of media consisting of k+1 media.

6. The media recommendation system according to claim 5, wherein: The second table is generated by the following operation: when using a set of media consisting of k media and a set of media consisting of k+1 media with high occurrence frequency included in the multiple owned media groups pre-extracted based on owned media group information related to the multiple owned media groups used by the specified multiple users, the T types of media included in the media set are respectively combined with the pre-extracted sets of media consisting of k media, thereby generating T sets of media consisting of k+1 media, and for each pre-extracted set of media consisting of k media, referring to the pre-extracted The occurrence frequencies of each of the generated sets of T media consisting of k+1 media are calculated based on the set of media consisting of k+1 media, thereby further extracting a specified number of sets of media consisting of k+1 media with high occurrence frequencies from the generated sets of T media consisting of k+1 media for each pre-extracted set of media consisting of k+1 media, and associating the pre-extracted sets of media consisting of k media with the media to be added in order to generate the further extracted sets of media consisting of k+1 media with high occurrence frequencies, where 2≤k≤D-1.

7. A table creation system for creating a table for recommending media in a game that progresses as a user selects a media from an owned media group configured to include D media and places the media on a game field, the recommended media being media to be added to a set of media in order to construct at least a part of the owned media group from a set of media, wherein each media included in the owned media group is selected from a set of media including T types of media preset in the game, in, The table corresponding to the respective numbers of media constituting the set of media consisting of k media accepted when recommending the media is created by the following operation: when using a set of media consisting of k media and a set of media consisting of k+1 media with high appearance frequencies included in the multiple owned media groups extracted in advance based on owned media group information related to multiple owned media groups used by a specified multiple users, the media included in the media set are respectively combined with the pre-extracted sets of media consisting of k media, thereby generating a plurality of sets of media consisting of k+1 media, and for each pre-extracted set of k media, the media included in the media set are respectively combined with the pre-extracted sets of media consisting of k media, thereby generating a plurality of sets of media consisting of k+1 media, and The method further comprises the step of: calculating the occurrence frequencies of the generated multiple sets of media consisting of k+1 media with reference to the pre-extracted set of media consisting of k+1 media, thereby further extracting a specified number of sets of media consisting of k+1 media with high occurrence frequencies from the generated multiple sets of media consisting of k+1 media for each pre-extracted set of media consisting of k media, and associating the pre-extracted set of media consisting of k media with the media to be added in order to generate the further extracted set of media consisting of k+1 media with high occurrence frequencies, where 2≤k≤D-1.

8. A media recommendation device for recommending media in a game that progresses as a user selects a media from an owned media group configured to include D media and places the media on a game field, the recommended media being media to be added to a set of media in order to construct at least a portion of the owned media group from a set of media, wherein each media included in the owned media group is selected from a set of media including T types of media pre-set in the game, the media recommendation device comprising: an input accepting unit for accepting an input of a set of k media constituting at least a part of the owned media group, where 2≤k≤D-1; A table storage unit for storing a second table corresponding to each of k=2 to D-1, wherein the second table is generated by the following operation: when using a set of media consisting of k media and a set of media consisting of k+1 media with high appearance frequencies included in the plurality of owned media groups extracted in advance based on owned media group information related to the plurality of owned media groups used by a predetermined plurality of users, the media included in the media set are respectively combined with the pre-extracted sets of media consisting of k media, thereby generating a plurality of sets of media consisting of k+1 media, and for each of the pre-extracted sets of k media, and calculating the occurrence frequencies of each of the generated sets of k+1 media with reference to a pre-extracted set of k+1 media, thereby further extracting a predetermined number of sets of k+1 media with high occurrence frequencies from the generated sets of k+1 media for each pre-extracted set of k media, and associating each pre-extracted set of k media with the media to be added in order to generate the further extracted set of k+1 media with high occurrence frequencies; A media recommendation unit that is capable of referring to a second table corresponding to the number of media constituting the set of media accepted by the input acceptance unit, and outputting as recommendation information the media that is associated with the set of media consisting of k media accepted by the input acceptance unit in the second table and is to be added in order to generate a set of media consisting of k+1 media.

9. A table creation method for creating a table for recommending media in a game that progresses as a user selects a media from an owned media group configured to include D media and places the media on a game field, the recommended media being media to be added to a set of media in order to construct at least a portion of the owned media group from a set of media, wherein each media included in the owned media group is selected from a set of media including T types of media pre-set in the game, the table creation method comprising the following steps: For each of k=2 to D-1, based on owned media group information related to a plurality of owned media groups used by a plurality of users, extracting a set of media consisting of k media with high appearance frequencies included in the plurality of owned media groups, where 2≤k≤D-1; For each of k=2 to D-1, the media included in the media set are respectively combined with the extracted sets of media consisting of each of k media, thereby generating a plurality of sets of media consisting of k+1 media, and for each of the extracted sets of media consisting of k media, the occurrence frequencies of the generated plurality of sets of media consisting of k+1 media are calculated with reference to the extracted sets of media consisting of k+1 media, thereby further extracting a prescribed number of sets of media consisting of k+1 media with high occurrence frequencies from the generated plurality of sets of media consisting of k+1 media for each of the extracted sets of media consisting of k media; as well as By associating each extracted set of k media with the media to be added in order to generate a further extracted set of k+1 media with a high occurrence frequency for each of k=2 to D-1, a table corresponding to the respective numbers of media constituting the set of k media accepted when recommending the media is created.

10. The table creation system according to claim 2 or 7, The table creation system further includes a medium extraction system configured to extract a combination of any number of mediums with a high appearance frequency included in the owned medium group. The media extraction system includes: A storage unit, configured to store the owned media group information; as well as an extraction unit capable of calculating the frequency of occurrence of each set consisting of m media with reference to the owned media group information stored in the storage unit, and extracting a set consisting of m media with a high frequency of occurrence that meets a specified standard, wherein 2≤m≤D, In which, the extraction unit is capable of generating multiple sets consisting of m+1 media by combining the media included in the media set into each extracted set, or is capable of generating multiple sets consisting of m-1 media by excluding a single media from the media constituting the extracted set from each extracted set.

11. The table creation system according to claim 10, in, The extracting unit of the medium extracting system periodically extracts a set of 2 to D mediums having a high frequency of occurrence that meets a predetermined standard, and The table creation system periodically creates a table based on a set of k media with high appearance frequencies included in the owned media group extracted by the extraction unit, where 2≤k≤D.

12. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to claim 4 or 9.

Images