JP7649266B2 - Information Processing System - Google Patents

Description

An embodiment of the present invention relates to an information processing system that manages content information.

Generally, electronic content is traded online. For example, a content display device has been disclosed that allows a user to quickly use other electronic content related to the electronic content that he or she has used (see Patent Document 1).

JP 2016-45902 A

However, since information about content is managed on various sites, etc., users may have difficulty in searching for and finding information about the desired content. For example, if the title of a manga matches the name of a character (e.g., the main character) that appears in the manga, both the manga and the character that match the searched word are displayed as search results. If a user who performed a search is looking for information about content related to only either manga or characters, the user needs to exclude information about content other than the desired content from the search results.
An object of the present invention is to provide an information processing system for managing content information, which reduces the burden on a user when searching for information on a desired content.

The information processing system according to the present invention is an information processing system that manages a plurality of pieces of content information, which is information related to the content of a plurality of creators, and includes a tag information storage unit that stores tags, which are incidental information attached to the content information, a tag type information storage unit that stores tag types, which are information indicating the types of the tags, a content information storage unit that stores the plurality of pieces of content information associated with the tags stored in the tag information storage unit and the tag types stored in the tag type information storage unit, and a search unit that performs a search to extract target content information from the plurality of pieces of content information stored in the content information storage unit based on the tags and the tag types.

The present invention provides an information processing system that manages content information, reducing the burden on users when searching for information on desired content.

1 is a configuration diagram showing a configuration of an information processing system according to a first embodiment of the present invention. 1 is a schematic diagram showing a configuration for creating a portfolio site of an information processing system according to a first embodiment. 4 is a conceptual diagram showing how tags and tag types are assigned to images according to the first embodiment; FIG. 4 is an image diagram showing tags and tag types displayed on a Web screen according to the first embodiment. FIG. 13 is an image diagram showing a Web screen for adding tags and tag types according to the first embodiment. FIG. 13 is an image diagram showing a Web screen for editing a portfolio site according to the first embodiment. FIG. 1 is an image diagram showing a web screen on which a search function is implemented in a portfolio site according to the first embodiment. FIG. 1 is a schematic diagram showing a configuration for searching in an information processing system according to a first embodiment. FIG. 4 is an image diagram of a first search screen according to the first embodiment. FIG. 11 is an image diagram of a second search screen according to the first embodiment. FIG. 11 is a configuration diagram showing the configuration of an information processing system according to a second embodiment of the present invention. FIG. 11 is a schematic diagram showing a configuration for performing machine learning in an information processing system according to a second embodiment. FIG. 13 is a conceptual diagram showing a method for predicting tags and tag types using machine learning according to the second embodiment.

First Embodiment
1 is a diagram showing the configuration of an information processing system 10 according to a first embodiment of the present invention. Note that the same parts in the drawings are given the same reference numerals, and duplicated explanations will be omitted as appropriate.

The information processing system 10 is a system that manages information related to the content of multiple creators. For example, the information processing system 10 is a system that supports creators in publishing their portfolios. Specifically, the information processing system 10 supports creators in creating portfolio sites. For example, the information processing system 10 displays various editing screens that allow creators to create portfolio sites. Creators use the information processing system 10 to create and publish their own portfolio sites. Users use the information processing system 10 to view the portfolio sites created by each creator and obtain information related to the content.

The information processing system 10 is mainly composed of a computer. The information processing system 10 may be composed of multiple computers. The information processing system 10 may include one storage device, or may be composed of multiple storage devices. The storage device may be part of a computer, or may be a device independent of the computer. The various functions and processes in the information processing system 10 are realized by executing a program on a computer.

The computer may be of any type. For example, the computer may be a desktop, laptop, tablet, or other type of computer, or may be a mobile phone, such as a smartphone. The storage device may be of any type. For example, the storage device may be a solid state drive (SSD) or a hard disk drive (HDD).

The information processing system 10 includes a storage device 1 and a computing unit 2. The information processing system 10 is connected to a network 3. Creators and users who use the information processing system 10 connect to the information processing system 10 from their own terminals 4a, 4b via the network 3 so as to be able to communicate with the information processing system 10. The creator terminal 4a is a computer used by the creator. The user terminal 4b is a computer used by the user.

The network 3 may be the Internet, a fixed telephone network, or a mobile telephone network, or may be a combination of these, or may use other communication networks. In addition, the creator terminal 4a and the user terminal 4b may be connected to the information processing system 10 so as to be able to communicate with each other without going through the network 3.

The storage device 1 stores various types of information used in the information processing system 10. Note that some or all of the information in each storage unit constituting the storage device 1 may be organized into a database. Furthermore, the storage device 1 may have any data structure as long as it is structured to store tags and tag types.

The storage device 1 has a member information area 11, a portfolio area 12, and a tag-related information area 13.

The member information area 11 is an area in which information about members is stored. Here, an information processing system 10 in which users who use the information processing system 10 are members-only is described, but as long as it is possible to identify those who are accessing the information processing system 10, the information processing system 10 is not limited to a membership-only system and any system may be applied.

The member information area 11 includes a member information storage unit 111, a creator information storage unit 112, a user information storage unit 113, a portfolio information storage unit 114, and a search history storage unit 115.

The member information storage unit 111 stores member information about members who use the information processing system 10. The member information includes member information and identity verification information. For example, the member information is an account name, etc. Identity verification information is a password for password authentication, or biometric information (fingerprint, etc.) for biometric authentication (fingerprint authentication, etc.). The information processing system 10 determines whether or not a person who has accessed the information processing system 10 is a member based on the member information. The information processing system 10 also determines whether the member who has accessed is a creator or a member based on the member information.

The creator information storage unit 112 stores creator information about creators who are members. For example, the creator information includes personal information. For example, the personal information is real name, pen name, contact information, etc.

The user information storage unit 113 stores user information about users who are members. The user information includes personal information. For example, the personal information is real name, nickname, contact information, etc.

The portfolio information storage unit 114 stores portfolio information that forms a portfolio of the creator's career or achievements. The portfolio information also includes information about the portfolio site created by the creator. The information processing system 10 makes information about the creator's content available to users via the portfolio site. Note that this is not limited to a portfolio site, and any site that makes information about the creator's content available outside the information processing system 10 may be used.

The search history storage unit 115 stores search conditions that the user has used in the past. The search conditions may be stored automatically based on preset conditions, or may be stored based on instructions from the user.

The portfolio area 12 is an area in which information about the works, products, fan communities, etc. to be posted as content on the portfolio site is stored. The portfolio area 12 is assigned to each creator. Creators can freely configure their portfolio sites.

The portfolio site may be configured in any manner so long as it is a site that serves as a creator's portfolio. For example, works, products, and fan communities are not all required content for a portfolio site, and the content may be determined arbitrarily by the creator. For example, the portfolio site may be provided with an optional e-commerce (electronic commerce) function. Furthermore, the information processing system 10 may impose restrictions on the content posted on the portfolio site. For example, it may be mandatory to post works, products, fan communities, etc. on the portfolio site.

The portfolio area 12 includes a work information storage unit 121, a product information storage unit 122, a fan community information storage unit 123, and a content storage unit 124. Each piece of information stored in each of the storage units 121 to 124 of the portfolio area 12 is assigned a tag and a tag type.

Here, a tag is additional information (meta information) that is attached to a work or a product, etc. A tag type indicates the type of tag and is information that is a higher-level concept than the tag. In other words, the tag type corresponds to the meta information of the tag. By attaching tags and tag types to works, etc., users can easily find the desired work, etc. by searching. For example, if the tag attached to an image that is a work is "Hero A" and the tag type is "character", this indicates that the image depicts a "character" called "Hero A". For example, types of tags that indicate the tag type include the work name, character name, author name, the situation of the character depicted in the work, the source of the work, or the name of the fan community, etc.

The work information storage unit 121 stores work information about the creator's works. For example, the work information includes information about a list of works displayed on the creator's portfolio site. The list of works displays tags and tag types along with information about each of the creator's works (for example, the name of the work or a thumbnail, etc.).

The product information storage unit 122 stores product information about the creator's products. For example, the product information includes information about a product list displayed on the creator's portfolio site. The product list displays tags and tag types along with information about each product of the creator (e.g., the work name or thumbnail, etc.). Users can purchase products from the creator's portfolio site.

The fan community information storage unit 123 stores fan community information to realize a fan club function that allows fans to communicate with each other. The target audience of a fan community may be fans of a work, fans of characters appearing in the work, or fans of a creator. Any type of fan community may be established on a portfolio site, and any number of fan communities may be established.

The content storage unit 124 stores content created by creators. For example, the content is a work to be posted on a portfolio site. The content may be anything, such as a book, an image, or a video, and may be not limited to a work of art, but may also be a product.

The tag-related information area 13 is an area in which information about tags and tag types is stored. The tag-related information area 13 includes a tag information storage unit 131 and a tag type information storage unit 132.

The tag information storage unit 131 stores information about tags. For example, the tag information storage unit 131 may store master data of a database. Specifically, the tag information storage unit 131 stores registered tags together with unique numbers (IDs, identification data) assigned to the tags. The information processing system 10 handles tags by the IDs assigned to the tags.

The tag type information storage unit 132 stores information about tag types. For example, the tag type information storage unit 132 may store master data of a database. Specifically, the tag type information storage unit 132 stores registered tag types together with unique numbers (IDs, identification data) assigned to these tag types. The information processing system 10 handles tag types by the IDs assigned to these tag types. Tag types are registered in advance before the information processing system 10 starts operating, but may be added, changed, or deleted during operation. For example, tag types are registered in the tag type information storage unit 132 by the operator.

Each storage unit 121 to 124 in the portfolio area 12 manages various information by referring to the IDs corresponding to the tags and tag types registered in the tag information storage unit 131 and the tag type information storage unit 132.

The calculation unit 2 is a part that performs calculation processing in the information processing system 10. The functions of the calculation unit 2 are realized by the calculation unit 2 executing calculation processing in response to instructions from a program or the like. For example, the calculation unit 2 is a semiconductor chip that performs calculation processing. The calculation unit 2 may be configured in any way as long as it executes calculation processing that uses the tags and tag types stored in the storage device 1. The calculation unit 2 includes a registration processing unit 21 and a search processing unit 22.

The registration processing unit 21 is a process that is executed mainly for creators to use the information processing system 10. For example, the registration processing unit 21 registers various contents of the creators and performs various processes to support the creators in creating a portfolio site.

The search processing unit 22 is a process that is executed mainly for the user to use the information processing system 10. For example, the search processing unit 22 performs a process for the user to search for a desired content from among various contents registered in the information processing system 10.

With reference to FIG. 2, a method in which a creator creates a portfolio site using the information processing system 10 will be described. For the purposes of this description, FIG. 2 is a simplified schematic diagram showing the configuration of the information processing system 10.

The registration processing unit 21 includes an input processing unit 211, a portfolio creation unit 212, and an output processing unit 213.

The input processing unit 211 processes input information from the creator terminal 4a used by the creator and outputs it to the portfolio formation unit 212.

The portfolio creation unit 212 performs processing for creating a portfolio based on the input information processed by the input processing unit 211. For example, the portfolio creation unit 212 performs processing for registering various pieces of content created by the creator in the portfolio. The portfolio creation unit 212 can also register new tags and tag types in order to register various pieces of content. In addition, the portfolio creation unit 212 performs processing for constructing a portfolio site in response to a request from the creator.

The output processing unit 213 outputs various information to the creator terminal 4a based on the processing results by the portfolio formation unit 212.

The creator accesses the information processing system 10 via the creator terminal 4a. The creator logs in to the information processing system 10 and creates a portfolio site. Specifically, the creator registers as a member to use the information processing system 10. When the information processing system 10 determines that the logged-in creator is a member based on the member information stored in the member information storage unit 111, it permits the creator terminal 4a to use the information processing system 10. This allows the creator to create his or her own portfolio in the information processing system 10 using the creator terminal 4a.

Referring to Figure 3, we will explain how a creator assigns tags and tag types to images C1, C2, and C3, which are his or her own works, and stores them in the content storage unit 124.

Here, the first image C1 is an image depicting a character named "Hero A" who appears in a manga titled "Hero A". The second image C2 is an image depicting a character named "Demon King B" who appears in a manga titled "Hero A". The third image C3 is an image depicted on the cover of the 13th volume of the manga titled "Hero A", depicting the characters "Hero A" and "Demon King B" together.

The creator saves the three images C1 to C3 in the content storage unit 124. The creator registers information about each of the three saved images C1 to C3 in the work information storage unit 121. Specifically, the creator assigns a tag and a tag type to each of the three images C1 to C3. The first image C1 is assigned two tags: a tag "Hero A" with a tag type of "character", and a tag "Hero A" with a tag type of "work title". In addition, the creator may register any information about each of the three images C1 to C3 in the work information storage unit 121.

A method for displaying tags and tag types will be described with reference to FIG. 4. FIG. 4 shows an example in which three types of tags Td1, Td2, and Td3 are displayed on a Web (world wide web) screen P1. For example, the Web screen P1 is displayed on a registration screen for a creator to register content or a search screen for a user to search for desired content.

For the first tag Td1, the tag is displayed in the left column Tg1, and the tag type is displayed in the right column Tg2. A delete button Bm for deleting the tag type is provided in the right column Tg2. When the delete button Bm is pressed, the first tag Td1 becomes a tag with no tag type. In other words, the first tag Td1 is displayed like the third tag Td3. For example, the delete button Bm is only displayed on the operation screen of the person who assigned the first tag Td1 (e.g., the creator). This allows only the person who assigned the first tag Td1 to delete the tag type.

For the second tag Td2, the tag is displayed in the left column Tg1, and the tag type is displayed in the right column Tg2. Unlike the first tag Td1, the second tag Td2 does not display a delete button Bm. Therefore, it is not possible to delete the tag type from the second tag Td2.

For the third tag Td3, only the tag is displayed, and the tag type is not displayed. In other words, the third tag Td3 is a tag without a tag type. For example, a tag without a tag type is an onomatopoeia, a coined word, a proper noun, or the like.

Referring to FIG. 5, a method for assigning tags and tag types on the web screen P2 will be described. For example, the web screen P2 is displayed on a registration screen for a creator to register (upload) content to a portfolio site. Note that, although four methods for assigning tags and tag types are described here, all of the methods are reference examples, and tags and tag types may be assigned by methods other than these methods.

For example, the content to which tags and tag types are to be assigned is set in the upper part (not shown) of web screen P2. For example, the creator sets the work to be registered or the name of the work on web screen P2.

The first method is executed by inputting necessary information into display unit M1. The first method is used when it is desired to specify the genre of the content to be registered in information processing system 10. The tag type is specified in advance in display unit M1, and the tag is selected from a list preset in information processing system 10.

For example, tags are selected from a list displayed by a pull-down menu. Specifically, if a creator wants to register a manga or illustration as essential content for a portfolio in the information processing system 10, the tag type is fixed and displayed as "category," and only "manga" and "illustration" are displayed in the tag list. This allows creators to specify and register manga or illustrations as content for their portfolio.

The second method is executed by inputting the necessary information into display unit M2. The second method is used when registering tags without a tag type collectively. Display unit M2 is provided with a text box TB in which multiple tags can be input collectively. The creator inputs the tag names without a tag type to be registered collectively into text box TB. Multiple tag names entered into text box TB are separated by symbols such as commas or spaces. This saves the creator the trouble of setting the tag type to "none", making it easy to register multiple tags without a tag type at once.

The third method is executed by inputting the necessary information into the input section M32 of the display section M3. The third method is used when registering an arbitrary tag. The input section M32 has a section for selecting the tag type and a field for inputting the tag name. The input section M32 also has a button for adding tags so that multiple tags can be registered. The tag display section M31 of the display section M3 displays the tag to be registered. This tag has been generated by inputting the necessary information into the input section M32 up to this point in time.

For example, the tag type is selected from a list displayed by a pull-down menu. The contents of the list change according to the content to be registered. For example, if the content to be registered is a "work", the list will include a "work name" but not a "product name". Note that it is also possible to register a tag without a tag type by not selecting anything from the list, or by selecting "none" from the list and then registering the tag. After selecting the tag type, the desired tag can be generated and registered by entering the tag name.

The fourth method is a method in which tags and tag types are automatically displayed on display unit M4. For example, when logging in to information processing system 10, information processing system 10 can identify the creator to be used. In this way, if information processing system 10 can identify the tag to be assigned to content without the creator having to newly input information related to the tag, the tag is automatically generated. Tags automatically displayed in this way may be made unable to be deleted by the creator. This makes it possible to automatically assign tags (e.g., creator name) that are essential as a portfolio service provided by information processing system 10.

A method for creating a portfolio site using tags and tag types will be described with reference to Figure 6. Figure 6 shows web screens P31 and P32 for editing the portfolio site for this description.

Here, we'll assume that tags and tag types have already been assigned to all content and that it has been registered as a portfolio. In this case, we'll explain how a creator can edit the configuration of the portfolio site.

Certain tag types can be used as labels with specific functions. For example, the tag type "series" identifies the series to which a work belongs.

The left column D31 of the web screen P31 displays a list equivalent to the table of contents of the portfolio site. When an item is selected from this list, tags corresponding to the selected item are displayed in the right column D32. For example, when "Series Settings" for "Works" is selected in the left column D31, the titles (tags) of all series of works are displayed in the right column D32.

When one of the titles in the series displayed in the right column D32 of the web screen P31 is selected, the web screen P32 is displayed. A list of works C31, C32, and C33 belonging to the selected series is displayed in the right column D33 of the web screen P32. The works C31 to C33 to be displayed are automatically determined by the information processing system 10 based on the tags and tag types assigned to the works.

The creator can rearrange or delete the displayed works C31 to C33. The creator may also be able to add works that were not automatically displayed by the information processing system 10 as works belonging to this series. This makes it easier for the creator to edit the portfolio site.

A method for implementing a search function using tags and tag types in a portfolio site will be described with reference to Figure 7. Figure 7 shows an image of a web screen for editing a portfolio site.

To search for tags that have a tag type, a search function that can narrow down searches using the tag type is required. Such a search function can be modularized. By modularizing it, creators can incorporate the search function at any position on their portfolio site.

The left column D41 of the Web screen P4 displays the parts that make up one page (screen) of the portfolio site. By selecting one part from the left column D41 and placing (dropping) it anywhere in the right column D42, the function corresponding to the selected part is implemented.

Here, the search part is a part that modularizes the search function. The work display part is a part that displays works on the screen. The product display part is a part that displays products on the screen.

If a search part is placed in location PS1, outside the range of the work display parts and product display parts, a search will be performed on all content, without distinction between works and products. If a search part is placed in location PS2, within the range of the product display parts, a search will be performed on products only. In other words, a narrowed search for products can be performed. Similarly, by placing a search part in a specific location, a function can be implemented to perform a narrowed search for specific targets, such as works or fan communities.

With reference to FIG. 8, a method in which a user searches within a portfolio site using the information processing system 10 will be described. For the purposes of this description, FIG. 8 is a simplified schematic diagram showing the configuration of the information processing system 10.

The search processing unit 22 includes an input processing unit 221, a search target determination unit 222, a search target extraction unit 223, and an output processing unit 224.

The input processing unit 221 processes input information indicating search conditions from the user terminal 4b used by the user, and outputs it to the search target determination unit 222. The input processing unit 221 also performs processing to display a screen for inputting search conditions on the user terminal 4b.

The search target determination unit 222 determines the search target based on the input information processed by the input processing unit 211. For example, if the tag "Demon King B" with a tag type of "character", the tag "Hero A" with a tag type of "work title", and the tag "illustration" with a tag type of "category" are specified as search conditions, the search target determination unit 222 determines that the search target is an "illustration" depicting a character named "Demon King B" who appears in content with a work title of "Hero A". The search target determination unit 222 outputs the determined search target to the search target extraction unit 223.

The search target extraction unit 223 searches for content that matches the search target determined by the search target determination unit 222 from among the content registered as a portfolio, and extracts information related to the content that matches the search target. The search target extraction unit 223 outputs the extracted information to the output processing unit 224.

The output processing unit 224 performs processing to output the search results to the user terminal 4b based on the information extracted by the search target extraction unit 223.

The user accesses the information processing system 10 using the user terminal 4b. The user logs in to the information processing system 10 and browses the portfolio sites of each creator. Specifically, the user registers as a member to use the information processing system 10. When the information processing system 10 determines that the logged-in user is a member based on the membership information stored in the membership information storage unit 111, it permits the user terminal 4b to use the information processing system 10. This allows the user to use the user terminal 4b to browse the portfolio site of any creator registered in the information processing system 10.

A user performs a search to find the desired content. The search range may be the portfolio sites of all creators, or the portfolio site of a specific creator. For example, if the search range is the portfolio sites of all creators, a search function that allows searching within such a search range is provided on a web screen created by the administrator of the information processing system 10.

The method for displaying the search screen will be described with reference to Figures 9 and 10. Figures 9 and 10 are image diagrams of the search screen.

Screen D52 is an input screen for performing a simple search. In a simple search, only tags are specified as search conditions, and tag types are not specified. A user enters search keywords in input field F1 of screen D52 to perform a search. Here, tags as search conditions correspond to search keywords in general searches. Therefore, a user can perform a search without being aware of tags. Screen D52 may be the initial state of the search screen. Screen D52 allows the display area to be small because there are few search conditions to be specified. Furthermore, with a simple search, a user does not need to determine detailed search conditions, making it easy to perform a search.

Screen D53 is an input screen for performing an advanced search. In an advanced search, a search is performed by specifying tags and tag types. Screen D53 includes an input field F1 for inputting search keywords, and buttons for multiple tag types for specifying as narrowing down conditions. A user inputs a search keyword in input field F1 and presses (e.g., clicks) the button for the tag type to be specified to perform a search. Multiple tag types may be specified in a search. Note that instead of the button for specifying the tag type, an input field for the user to input the tag type may be provided, similar to the input field F1 for search keywords. The simple search screen D52 and the advanced search screen D53 may be able to be switched between by the user.

Note that the search is not limited to the above, and any search conditions may be set. For example, the order in which the search results are displayed may be specified, such as by popularity or by posting date.

The search results of an advanced search are equivalent to search results where a tag is used as a search keyword and then narrowed down by tag type. For example, suppose you perform a search by specifying "Hero A" as the search keyword and "character" as the tag type in the narrowing criteria. In this case, images depicting "Hero A" as a character will be included in the search results, but images that do not depict "Hero A" as a character will not be included in the search results, even if they are from a work with the title "Hero A." In other words, content in which "Hero A" is not treated as a "character" will not be included in the search results.

Screen D51 is a screen that displays the search results from a simple search or an advanced search. Screen D51 is provided with a save button B51 for saving the search conditions, and an additional conditions button B52 for specifying additional search conditions. Information about content that matches the search conditions is displayed as the search results at the bottom of screen D51.

When the user presses the save button B51, the search conditions of the displayed search results are saved. For example, the user saves the search conditions when satisfactory search results are obtained. As shown in screen D55, the saved search conditions are displayed as suggested search conditions when the user focuses on input field F1 (for example, when the cursor is moved near input field F1) during subsequent searches. The user may be allowed to call up the saved search conditions at any time. The user may also be allowed to save the search conditions before searching.

When the user presses the additional conditions button B52, screen D54 is displayed. The user can use screen D54 to add new search conditions to the search conditions of the current search results and perform an additional search. For example, an additional search is an AND search or an exclusion search. An AND search is a search performed by adding a new search keyword to the current search keywords using AND (logical product). An exclusion search is a search performed by adding an exclusion word to the current search conditions. An exclusion search excludes content containing the exclusion word from the exclusion results. By performing an additional search, the current search results can be narrowed down.

Note that screen D51 is the detailed search screen D53 to which a screen for displaying search results has been added, but is not limited to this. For example, screen D51 may display only the search results. Furthermore, the screen for displaying the search results of a simple search may be the simple search screen D52 to which a screen for displaying search results has been added.

According to this embodiment, the information processing system 10 can easily find the desired content by searching by associating a tag having a tag type with the content and managing information related to the content. For example, when the information processing system 10 provides a service of providing portfolio sites for creators, even if portfolio sites with completely different configurations are created for each creator, the desired content can be easily found by searching from the portfolio sites of all creators. Furthermore, by narrowing down the search by tag type, it is possible to easily exclude content that is not desired to be included in the search results. This can improve the convenience for users who use the information processing system 10.

Second Embodiment
FIG. 11 is a diagram showing the configuration of an information processing system 10A according to the second embodiment.

Information processing system 10A, in information processing system 10 according to the first embodiment shown in FIG. 1, calculation unit 2 is replaced with calculation unit 2A, and memory device 1 is replaced with memory device 1A. Calculation unit 2A is obtained by adding a machine learning unit 23 to calculation unit 2 according to the first embodiment. Memory device 1A is obtained by adding a trained model memory unit 14 to memory device 1 according to the first embodiment. Other points are the same as those of the first embodiment.

The information processing system 10A performs machine learning to predict tags and tag types for any content. The predictive model (trained model) created by machine learning is used to improve convenience for creators or users who use the information processing system 10A.

With reference to FIG. 12, a method in which a creator registers new content as a portfolio in the information processing system 10 will be described. For the purposes of this description, FIG. 12 is a simplified schematic diagram showing the configuration for performing machine learning in the information processing system 10A.

The machine learning unit 23 performs machine learning using the information stored in the memory units 121-124 of the portfolio area 12 and the memory units 131-132 of the tag-related information area 13 as learning data. The predictive model generated by machine learning is stored in the trained model memory unit 14. Machine learning by the machine learning unit 23 may be performed automatically based on the settings of the information processing system 10A, or may be performed at any time by the operator of the information processing system 10A.

The calculation unit 2A includes an image analysis unit 214 and a data processing unit 215 in the registration processing unit 21. For example, the image analysis unit 214 and the data processing unit 215 form part of the portfolio creation unit 212.

The image analysis unit 214 analyzes images input from the creator terminal 4a to be registered as content by the creator. Specifically, the image analysis unit 214 analyzes the image data input from the input processing unit 211 using a prediction model stored in the trained model storage unit 14. The image analysis unit 214 predicts tags and tag types for the image by analyzing the image data. The image analysis unit 214 outputs the analysis results to the data processing unit 215.

The data processing unit 215 performs data processing to output data related to the predicted tags and tag types for the image to the creator terminal 4a based on the analysis results by the image analysis unit 214. The data processing unit 215 outputs data related to the predicted tags and tag types to the creator terminal 4a via the output processing unit 213. The data processing unit 215 may automatically assign the predicted tags and tag types to the image.

The creator decides on the tag and tag type to be assigned to the content to be registered, using the predicted tag and tag type received from the calculation unit 2A of the information processing system 10A as one of the candidates. For example, the creator determines whether the predicted tag and tag type are suitable as tags and tag types to be assigned to the content to be registered. The creator may modify the predicted tag and tag type and assign it to the content to be registered.

A method for predicting tags and tag types using machine learning will be described with reference to Figure 13. Figure 13 is a conceptual diagram for this description.

A learning data group DLG is prepared for machine learning by the machine learning unit 23. Each piece of learning data DL1, DL2, DL3, DL4, and DL5 in the learning data group DLG is content to which tags and tag types are assigned and which are stored in the memory units 121 to 124 of the portfolio area 12. Here, the learning data DL1 to DL5 are described as images.

The machine learning unit 23 performs machine learning using the learning data DL1 to DL5, and generates a prediction model MD that predicts tags and tag types from any image. The prediction model MD is used to predict tags and tag types for newly registered images CT.

The image of the first learning data DL1 is given the tag "Hero A in a pinch" of the tag type "situation" and the tag "Volume 10" of the tag type "source".
The image of the second learning data DL2 is given the tag "Demon King B" of the tag type "character."
The image in the third learning data DL3 is given the tag "Hero A" of the tag type "Title of Work."
The image in the fourth learning data DL4 is assigned the tag "Hero A" of the tag type "none," the tag "Yamada X" of the tag type "author," and the tag "Weekly Z" of the tag type "serialized magazine."

The image in the fifth learning data DL5 is tagged with the tag "Sun God" in the tag type "Character" and the tag "Hero A" in the tag type "Title of Work".

When the image CT to be registered is an image that combines images of five pieces of learning data DL1, the tag type and tag assigned to each of the five pieces of learning data DL1 are presented as the prediction result TR using the prediction model MD.

Here, a method for predicting tag types and tags using a prediction model generated by machine learning has been described to improve the convenience of creators, but it may also be used to improve the convenience of users. For example, when a user inputs search conditions, tags and tag types predicted by the prediction model from content previously searched by the user may be displayed to suggest the tags and tag types to the user. In addition, the prediction model generated by machine learning may be applied to any function in the information processing system 10. Furthermore, an API (application programming interface) for image analysis that predicts tags and tag types based on the prediction model may be created. This makes it possible to provide an API that predicts tags and tag types outside the information processing system 10. Therefore, an API that predicts tags and tag types can be easily implemented on a computer outside the information processing system 10.

According to this embodiment, in addition to the effects of the first embodiment, the convenience of creators or users who use the information processing system 10 can be improved by predicting tag types and tags using a prediction model generated by machine learning.

However, additional advantages and modifications may readily occur to those skilled in the art. Therefore, the invention in its broader aspects is not limited to the specific detailed representative embodiments shown and described herein. Accordingly, various modifications may be made without departing from the spirit or scope of the general inventive concept as defined by the appended claims and their equivalents.

1... storage device, 2... computing unit, 3... network, 11... member information area, 12... portfolio area, 13... tag-related information area, 111... member information storage unit, 112... creator information storage unit, 113... user information storage unit, 114... portfolio information storage unit, 115... search history storage unit, 121... work information storage unit, 122... product information storage unit, 123... fan community information storage unit, 124... content storage unit, 131... tag information storage unit, 132... tag type information storage unit.

Claims (7)

An information processing system for managing a plurality of pieces of content information, which is information about the content of a plurality of creators who create manga or illustrations and which is published via the portfolio sites of the respective creators, comprising:
a site creation processing means for performing a process to support the creator in creating his/her portfolio site;
a portfolio information storage unit that stores portfolio information regarding a portfolio including the portfolio site of each of the plurality of creators;
a tag information storage unit that stores tags, which are supplementary information added to the content information;
a tag type information storage unit that stores a tag type, which is information indicating a type of the tag;
a content information storage unit that stores the plurality of pieces of content information associated with the tags stored in the tag information storage unit, the tag types stored in the tag type information storage unit, and the portfolio information stored in the portfolio information storage unit;
and a search means for executing a search to extract a desired content information from the plurality of content information stored in the content information storage unit based on the tag and the tag type, or the tag, the tag type, and the portfolio site of the portfolio information. a registration processing means for performing a process of registering the content information so as to manage the content information in the information processing system;
2. The information processing system according to claim 1, further comprising: a tag-related information designation means for designating the tag and the tag type associated with the content information registered by the registration processing means in order to register the content information. The information processing system according to claim 1, further comprising a machine learning means for performing machine learning to generate a predictive model for predicting the tag and the tag type associated with any content information, using the plurality of content information stored in the content information storage unit as learning data. a registration processing means for performing a process of registering the content information so as to manage the content information in the information processing system;
a tag-related information designation unit that designates the tag and the tag type associated with the content information to be registered by the registration processing unit in order to register the content information;
4. The information processing system according to claim 3, further comprising: The information processing system according to claim 4 , wherein the tag-related information designation means is configured to designate the tag and the tag type predicted by the prediction model generated by the machine learning means. The information processing system according to claim 4, further comprising a tag-related information candidate means for designating the tag and the tag type predicted by the predictive model generated by the machine learning means as candidates for the tag and the tag type to be designated by the tag-related information designation means. The information processing system according to claim 3 , characterized in that an application programming interface created based on the predictive model generated by the machine learning means is implemented in a computer external to the information processing system.

Images