Content management system, content management program, and content management method

The content management system addresses the challenge of managing diverse unstructured content by using machine learning to organize and search content based on semantic similarity, ensuring efficient and relevant content delivery with reduced user burden and flexible usage conditions.

JP7896513B2Active Publication Date: 2026-07-29TOPPAN HOLDINGS INC
View PDF 7 Cites 0 Cited by

Patent Information

Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
TOPPAN HOLDINGS INC
Filing Date
2023-02-22
Publication Date
2026-07-29

AI Technical Summary

Technical Problem

Existing systems struggle with the normalization and centralized management of unstructured content such as articles, photos, and videos due to their varied expression forms and data formats, and there is a growing demand for systems that can uniformly manage and provide relevant content to users.

Method used

A content management system utilizing a reception unit, feature extraction unit, search unit, and output unit, which employs machine learning models to organize and search content based on semantic similarity, along with user management, update management, and history analysis units to facilitate efficient content usage and update notification.

Benefits of technology

Enables unified management and accurate search of unstructured content, reduces user burden in obtaining licenses, allows for flexible content usage conditions, and facilitates quick update notifications, thereby enhancing content relevance and user experience.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007896513000001
    Figure 0007896513000001
  • Figure 0007896513000002
    Figure 0007896513000002
  • Figure 0007896513000003
    Figure 0007896513000003
Patent Text Reader

Abstract

To provide a content management system, method and program, for managing non-uniform content in a unified manner.SOLUTION: A content management system 10 connected to a plurality of content transmission systems including a transmitter device over a network includes a receiving unit 20, a feature extraction unit 21, a retrieval unit 22, and an output unit 23. The receiving unit receives input of data on content from the transmitter device managed by a transmitter of the content. The feature extraction unit uses the content data and a model which has been trained by machine learning so as to output, as features, a concept included in the content, to generate feature data that associates the content with features obtained by applying the content to the model. The retrieval unit uses the feature data to retrieve content associated with features corresponding to a search condition input from a user device managed by a user who is to use the content. The output unit outputs content information corresponding to a retrieval result to the user device.SELECTED DRAWING: Figure 2
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to a content management system, a content management program, and a content management method.

Background Art

[0002] In recent years, in the background of the promotion of the use of ICT (Information and Communication Technology) in urban functions, the development of a system that provides optimal services to people by collectively managing various information obtained in social life has been underway. For example, the system described in Patent Document 1 collects personal information such as the gender and age of users and the location information of users for a plurality of users, and provides each user with information for safe movement.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] If information indicating current or past facts such as personal information and location information is the management target, it is easy to normalize the information into data in a unified format, so it is easy to perform collective management. On the other hand, content including creative works such as articles, photos, and videos has various expression forms and data forms, so the normalization of such content is complicated.

[0005] On the other hand, in modern society, a large amount of content is disseminated through networks by numerous creators using various means. Therefore, there is a growing demand for centralized content management and the provision of information to help users select appropriate content. Consequently, the development of a system capable of uniformly managing unstructured content is highly desirable. [Means for solving the problem]

[0006] A content management system for solving the above problems comprises: a reception unit that receives input of content data from a sender device, which is a device managed by the sender of the content; a feature extraction unit that uses the content data and a machine learning model that outputs concepts contained in the content as features, and applies the content to the model to generate feature data that associates the features obtained with the content; a search unit that uses the feature data to search for content associated with features corresponding to search conditions input from a user device, which is a device managed by a user who plans to use the content; and an output unit that outputs content information, which is information indicating the content corresponding to the search results from the search unit, to the user device.

[0007] A content management program to solve the above problems includes: receiving input of content data from a sender device, which is a device managed by the content sender, on one or more computers; using the content data and a machine learning model that outputs concepts contained in the content as features, applying the content to the model to generate feature data that associates the features obtained with the content; using the feature data to search for content associated with features corresponding to search conditions entered from a user device, which is a device managed by a user who intends to use the content; and providing content information, which is information indicating the content corresponding to the search results, to the user To output to the device and to execute the following.

[0008] A content management method to solve the above problems involves one or more computers receiving input of content data from a sender device, which is a device managed by the content sender; using the content data and a machine learning model that outputs concepts contained in the content as features, applying the content to the model to generate feature data that associates the features obtained with the content; using the feature data to search for content associated with features corresponding to search conditions entered from a user device, which is a device managed by a user who intends to use the content; and outputting content information, which is information indicating the content corresponding to the search results, to the user device.

[0009] According to the above configuration, content is organized according to the meaning it conveys through feature extraction using machine learning models. Therefore, unified management of unstructured content is possible. Furthermore, since content with high semantic similarity to the search criteria is displayed in the search results, it is possible to output information about appropriate content.

[0010] In the system described above, the feature data may be data in a high-dimensional vector space in which each of the multiple contents is represented as a feature vector. According to the above structure, content can be accurately organized and searched based on semantic similarity.

[0011] The above system may further include a user management unit that receives an application from the user for the use of the content and inquires with the sender device whether or not permission for the use of the content is granted. The above configuration makes it possible to reduce the burden on users in obtaining licenses.

[0012] In the above system, the receiving unit may accept input of usage conditions, which are conditions set by the sender for the use of the content, in addition to the content data, and the output unit may output the usage conditions for the content corresponding to the content information to the user device, in addition to the content information.

[0013] The above configuration facilitates the adjustment of conditions regarding content use between users and content creators. Furthermore, since content creators can set usage conditions for each piece of content, content use is encouraged in line with the creator's intentions, and the adjustment of these conditions becomes even smoother.

[0014] The above system may further include an update management unit that, upon receiving notification of an update to previously entered content from the sender device, notifies the user device of the user currently using the content of the update.

[0015] With the above configuration, users can quickly reflect updates to content from the source in the content they are using.

[0016] The above system may further include a history analysis unit that analyzes the usage history of the aforementioned content. The above configuration makes it possible to analyze topics of high user interest and user activity trends. [Effects of the Invention]

[0017] According to the present invention, unstructured content can be managed uniformly. [Brief explanation of the drawing]

[0018] [Figure 1] A diagram showing the overall configuration of a content management system according to one embodiment. [Figure 2] A diagram illustrating the functional configuration of a content management system according to one embodiment. [Figure 3]A diagram showing the processing procedures in the content sorting function and content search function in the content management system of an embodiment. [Figure 4] A diagram showing the processing procedures in the content usage management function in the content management system of an embodiment. [Figure 5] A diagram showing the processing procedures in the content update management function in the content management system of an embodiment.

Mode for Carrying Out the Invention

[0019] Referring to FIGS. 1 to 5, an embodiment of a content management system, a content management program, and a content management method will be described. In the present embodiment, the content is composed of at least one of text, images, audio, and video. The content includes creative works. For example, the content is an article, illustration, photograph, music, video, etc. published on a website.

[0020] [System Overview] As shown in FIG. 1, the content management system 10 is connected to each of a plurality of content distribution systems 50 via a network such as the Internet. In addition, the content management system 10 is connected to one or more user devices 60 via a network such as the Internet. Note that various APIs (Application Programming Interfaces) may be interposed for data input / output between the content management system 10 and the content distribution system 50 or the user device 60.

[0021] The content distribution system 50 is managed by the content distributor and distributes the content to the public via the network. The content distributor manages the content of the distributed content and has the right to use the content or the management authority of such rights. For example, the content distributor includes the content creator.

[0022] The content distribution system 50 consists of one or more information processing devices. One information processing device is a standalone computer. The content distribution system 50 includes, for example, a server for a website where content is provided, a database where content data is stored, and terminals operated by the sender to access these servers and databases. Each of these devices included in the content distribution system 50 is an example of a sender device.

[0023] Specific examples of content dissemination systems 50 include the operation system of a local government's tourism website that publishes articles about tourist spots and events as content, the operation system of a news website that publishes articles about current events as content, and the operation system of a gourmet website that publishes articles about restaurants as content.

[0024] User device 60 is managed by the user who intends to use the content. User device 60 is an information processing device such as a smartphone, tablet terminal, personal computer, server, or database. The user may be an individual who intends to use the content privately, or a service provider who intends to use the content on their own site or other platform to provide a service. If the user is a service provider, user device 60 is managed by the user who intends to use the content. This may include devices within the service provider's service operating system. For example, if a service provider provides information through an information website, the service operating system may include the information website's server, a database storing data for the content posted on the information website, and terminals operated by the service provider to access these servers and databases.

[0025] The content management system 10 consists of one or more information processing devices. The content management system 10 includes, for example, a server and a database. The content management system 10 receives content input from the content transmission system 50 and organizes the input content according to the meaning indicated by the content. Then, the content management system 10 searches for content corresponding to the search criteria entered from the user device 60 from the organized collection of content and outputs content information based on the search results to the user device 60.

[0026] [Content Management System Configuration] Referring to Figure 2, the configuration of the content management system 10 will be described in detail. The content management system 10 has a content organization function, a content search function, a content usage management function, a content update management function, and a content usage history analysis function.

[0027] The content management system 10 includes electronic circuits such as a CPU, MPU, and GPU, memory such as ROM, RAM, registered memory, and unbuffered memory, and storage such as an SSD or HDD. The content management system 10 may also include integrated circuits such as an ASIC or FPGA. Each process performed by the content management system 10 may be performed by software provided by the content management system 10, or it may be realized by a combination of the integrated circuits and software provided by the content management system 10. The above software includes a content management program that causes the computer to perform the processes of each part described below.

[0028] As shown in Figure 2, the content management system 10 functions as a reception unit 20, a feature extraction unit 21, a search unit 22, an output unit 23, a usage management unit 24, an update management unit 25, and a history analysis unit 26, through the execution of the content management program. The content management system 10 also stores registration data 30, feature data 31, and usage history data 32. In other words, the content management system 10 also functions as a data storage unit. The processing of each unit will be described in order below.

[0029] <Reception Department> The reception unit 20 receives content data input from the content transmission system 50. No fixed format is required for the input content, neither in terms of the content's representation format (recognized by the user) nor its data format (recognized by the computer). In other words, the input content is unstructured and unnormalized. To put it another way, multiple pieces of content input from multiple content transmission systems 50 may contain content in different formats.

[0030] In addition, the content distribution system 50 also inputs the terms of use, which are the conditions set by the content provider regarding the use of the content. The reception unit 20 generates registration data 30 for the input content and stores the registration data 30. The registration data 30 includes the content data and terms of use.

[0031] <Feature extraction unit> The feature extraction unit 21 uses the content data and the concepts contained in the content as features. Using a machine learning model trained to produce output, feature data 31 is generated by applying the content to the machine learning model and associating the features obtained with the content. The concepts contained in the content are elements that are directly or indirectly related to the topic, purpose, meaning, etc., contained in the content.

[0032] Specifically, the feature extraction unit 21 represents the content on a vector space using distributed representation. That is, the feature extraction unit 21 uses a machine learning model to extract features from the content, using various concepts contained within the content as features, and calculates the feature vector of the content. In this case, features are obtained as feature quantities from the machine learning model, and through the association between features and content, the content is represented as a feature vector in the feature data 31. In other words, the feature data 31 is data in a high-dimensional vector space in which content is represented as points indicating feature vectors. To put it another way, the feature data 31 is a collection of organized content.

[0033] This allows multiple pieces of content to be represented on a vector space according to their semantic similarity. For example, among multiple pieces of content, those with high semantic similarity from a perspective related to a particular concept are embedded in close proximity within the vector space from the perspective corresponding to that viewpoint. In this way, the semantic positional relationships of each piece of content with respect to other pieces of content and various keywords are defined in a multidimensional manner.

[0034] The feature extraction unit 21 uses known algorithms such as natural language processing, image recognition, and speech recognition. The feature extraction unit 21 only needs to be equipped with a machine learning model appropriate to the technology used. For example, if the content includes text, available technologies include averaging word embeddings or a combination of Word2Vec and TF-IDF, Universal Sentence Encoder, Word Rotator's Distance, Sentence-BERT, etc.

[0035] Furthermore, the processing of the feature extraction unit 21 may differ from the above, as long as a machine learning model is used to output the concepts contained in the content as features and associate the features with the content. For example, the association between features and content may be an association between tags that represent concepts and the content.

[0036] <Search section> The search unit 22 uses the feature data 31 to search for content that has a high semantic similarity to the search conditions input from the user device 60. The search conditions can be, for example, words, sentences, still images, or videos. In other words, the search unit 22 searches for content associated with features corresponding to the search conditions. Specifically, the search unit 22 calculates a feature vector for the search conditions and identifies content that is close to the search conditions in the vector space of the feature data 31. The search unit 22 may identify multiple pieces of content in order of proximity to the search conditions.

[0037] <Output section> The output unit 23 refers to the registered data 30 and outputs the content information Ic and usage conditions of the content identified by the search unit 22 to the user device 60. The content information Ic is information that identifies the content and may include the content data, information indicating the location of the content such as the site where the content is posted, or information indicating an overview of the content. The content information Ic may include multiple of these. The content information Ic may be information included in the registered data 30, or information generated based on the information included in the registered data 30.

[0038] <Usage Management Department> The User Management Unit 24 receives user requests to use content and the content We will manage the use of the content. This management includes monitoring the permission status of content creators regarding the use of their content. Content use includes secondary use such as quoting and reprinting content, as well as commercial use of content for profit.

[0039] The user management unit 24 sends the content usage application information Ia, which is input from the user device 60, to the content distribution system 50 for the target content. The usage application information Ia includes information necessary for determining whether or not to grant permission to use the content, such as the intended user of the content and the form of content use. The user management unit 24 transmits the result of the decision on whether or not to grant permission to use the content, which was received from the content distribution system 50, to the user device 60.

[0040] The usage management unit 24 stores the user's content usage status as usage history data 32 based on the usage application information Ia and the result of the usage permission decision. The usage history data 32 includes information indicating the content the user is currently using and the content the user has used in the past, as well as information such as the usage pattern and usage period of such content.

[0041] <Update management department> The update management unit 25 manages updates to the content. When the update management unit 25 receives notification of a content update from the content distribution system 50, it refers to the usage history data 32 and sends a content update notification to the user device 60 of the user currently using the content.

[0042] <History Analysis Department> The history analysis unit 26 analyzes the usage history of content by referring to the usage history data 32. The analysis includes, for example, aggregating the frequency of use by users for each piece of content, and analyzing cross-user usage of multiple pieces of content by each user, that is, combinations of multiple pieces of content used by the same user.

[0043] [Content Management System Operation] The operation of the content management system 10 will be explained for each function it possesses. The operation of the content management system 10 is, in other words, the content management method implemented by the content management system 10.

[0044] <Content organization function and content search function> Refer to Figure 3 to explain the content organization function and content search function. First, let's explain the content organization function. The content organization function is realized by the processing of the reception unit 20 and the feature extraction unit 21.

[0045] As shown in Figure 3, the content distribution system 50 transmits the content data and usage conditions to the content management system 10 for content that it wishes to register with the content management system 10 (step S10). The information transmitted by the content distribution system 50 may also include information necessary for the content management system 10 to manage the content, such as information about the sender.

[0046] The terms of use, which include, for example, whether permission is required to use the content, the permitted forms of use, and which parts of the content can be used, can be arbitrarily set by the content creator. Furthermore, the content creator can also set the terms of use to allow free use of the content.

[0047] The content management system 10 generates registration data 30 for the content input from the content distribution system 50 and stores the registration data 30 (step S11). The processing in step S11 is performed by the reception unit 20, and this completes the reception of the content.

[0048] In addition to the content data and terms of use, the registration data 30 may also include other information received from the content transmission system 50, such as information about the sender, and information generated for content management based on the information from the content transmission system 50.

[0049] Next, the content management system 10 analyzes the input content data, performs feature extraction on the content, and associates the obtained features with the content (step S12). This generates feature data 31. The process in step S12 is the process of the feature extraction unit 21.

[0050] Next, the content search function will be explained. The content search function is implemented by the processing of the search unit 22 and the output unit 23. When a user is looking for content to use, they input search criteria, such as keywords related to the desired content, into the user device 60. Based on this input, the user device 60 transmits the search criteria to the content management system 10 (step S13). If the user is a service provider and the user device 60 is a device that constitutes the service operation system, the search criteria may be pre-set in that system.

[0051] Upon receiving search criteria, the content management system 10 uses the feature data 31 to search for content that has a high semantic similarity to the search criteria (step S14). The process in step S14 is the process performed by the search unit 22.

[0052] Next, the content management system 10 refers to the registered data 30 and outputs content information Ic and usage conditions corresponding to one or more content items identified as a result of the search to the user device 60 (step S15). The processing in step S15 is the processing of the output unit 23.

[0053] For example, if content information Ic contains content data, the content is output to the user device 60. The content management system 10 may process the content into a different format than the format transmitted by the content transmission system 50 and output it as content information Ic. Examples of content processing include processing into a specific print format such as a booklet, or processing the display design of the content to match the output terminal or program.

[0054] Furthermore, if the content information Ic includes information indicating the location of the content, the user device 60 outputs information indicating the location of the content. Based on this information, the user can access the website operated by the content distribution system 50 to view the content. Also, if the content information Ic includes information indicating an overview of the content, the user device 60 outputs an overview of the content. This allows the user to understand the overview of the content.

[0055] Furthermore, if the user is a service provider, a service utilizing the content is implemented based on the content information Ic, such as posting content on an information site. If the user device 60 is a device that constitutes the service operating system, the service user, who is the ultimate consumer of the content, receives the content by accessing the system using a terminal such as a smartphone or personal computer.

[0056] In the content management system 10, content is multidimensional based on semantic similarity. Because the data is organized systematically, the relevance of content is determined from a perspective corresponding to the concept indicated by the search criteria, resulting in search results that include content containing concepts similar to the search criteria. In other words, the relevance of content is judged from different angles for each search criterion, and different search results are obtained for each angle. Therefore, information on the content desired by the user is accurately provided. As new content is entered, the feature data 31 is updated, so that the most appropriate search results are obtained at the time the user enters the search criteria. Furthermore, since the content usage conditions are output to the user device 60, the user can check the conditions set by the sender regarding the use of the content. Therefore, the user can select the content to use while taking the usage conditions into consideration.

[0057] <Content Usage Management Function> The content usage management function will be explained with reference to Figure 4. The content usage management function is implemented by the processing of the usage management unit 24.

[0058] If the terms of use stipulate that permission is required for the use of the content, a user who wishes to use the content in a form requiring permission enters the usage application information Ia into the user device 60. Based on this, the user device 60 transmits the usage application information Ia to the content management system 10 (step S20).

[0059] Furthermore, the form for entering the application information Ia is sent from the content management system 10 to the user device 60, for example, based on the content management system 10 receiving a notification from the user device 60 indicating that the user wishes to use the content in response to the user's actions.

[0060] Upon receiving the usage application information Ia, the content management system 10 sends the usage application information Ia to the content distribution system 50 managed by the content provider whose content is to be used, and inquires whether permission to use the content has been granted (step S21).

[0061] The sender reviews the usage application information Ia and decides whether or not to grant permission to use the content. Once the decision on whether or not to grant permission is entered into the content transmission system 50, the content transmission system 50 transmits the decision on whether or not to grant permission to use the content to the content management system 10 (step S22).

[0062] The content management system 10 transmits the received license approval decision result to the user device 60 (step S23). The content management system 10 also records the user's usage status of the content as usage history data 32 based on the usage application information Ia and the license approval decision result (step S24). The processes in steps S21, S23, and S24 are the processes of the usage management unit 24.

[0063] Furthermore, the terms of use may be changed by the sender. When the terms of use are changed, the content management system 10 updates the terms of use included in the registration data 30 to the new terms of use, based on the transmission of the new terms of use from the content transmission system 50 to the content management system 10. In addition, the content management system 10 notifies the user device 60 of the user who was licensed to use the service under the previous terms of use of use of the new terms of use. If necessary based on the new terms of use, a new application for use and a decision on whether to grant permission may be made between the user and the sender via the content management system 10.

[0064] Furthermore, the usage history data 32 may include not only the user's content usage history, but also other user behaviors that can be tracked by the content management system 10, such as content search results and content viewing history.

[0065] As described above, since the content management system 10 mediates the licensing of content, the burden on users to obtain licenses is reduced. In particular, when a user wants to use multiple pieces of content from different sources, the user does not need to contact each source individually; they only need to enter the usage application information Ia for each piece of content into the content management system 10, thus significantly reducing the burden.

[0066] Furthermore, since the terms of use are pre-set by the content creator, it is easy to adjust the terms of use between the user and the creator. In addition, since terms of use can be set for each piece of content, creators can set different terms of use for each piece of content they create. Therefore, it is possible to set detailed terms of use that are appropriate for the content, which encourages the use of content in line with the creator's intentions, and also makes it easier to adjust the above conditions.

[0067] <Content Update Management Function> The content update management function will be explained with reference to Figure 5. The content update management function is implemented by the processing of the update management unit 25.

[0068] If the content being transmitted by the content transmission system 50 is updated, the content transmission system 50 sends update data, which is the data of the updated content, to the content management system 10 (step S30).

[0069] Upon receiving update data, the content management system 10 overwrites the content data contained in the registration data 30 with the update data (step S31). The content management system 10 then refers to the usage history data 32 and sends an update notification to the user device 60 of the user currently using the target content, indicating that the content has been updated (step S32). The processes in steps S31 and S32 are the processes performed by the update management unit 25.

[0070] If, regarding the content before the update, the content data was transmitted to the user device 60 as content information Ic, the updated content data, i.e., the update data, may also be transmitted to the user device 60 along with the update notification.

[0071] Furthermore, the feature extraction unit 21 may perform content reorganization on the updated content, or the terms of use may be changed, or a new application for use and permission decision may be made between the user and the sender via the usage management unit 24.

[0072] Furthermore, when content is first input, after the feature data 31 has been generated, the content data may be deleted from the registration data 30. This reduces the amount of data stored in the content management system 10. In this case, the content information Ic output to the user device 60 does not include the content data. When content is updated, it is sufficient for the content transmission system 50 to send a notification to the content management system 10 indicating that the content has been updated; the update data does not need to be sent.

[0073] As described above, since content updates are notified to users currently using the content, users can quickly reflect the updates from the source in the content they are using. Therefore, it is easy to keep the content being used up to date. Note that content updates may also include content deletion.

[0074] <Content Usage History Analysis Function> This section describes the content usage history analysis function. The content usage history analysis function is implemented by the processing of the history analysis unit 26.

[0075] The history analysis unit 26 analyzes the content usage history by referring to the usage history data 32 and outputs the analysis results. The output destination of the analysis results may be a terminal operated by the administrator of the content management system 10, or it may be the content distribution system 50 or user device 60.

[0076] The history analysis unit 26, for example, aggregates the frequency of user usage for each of the multiple content items that make up the feature data 31. This makes it possible to understand which content is frequently used by users and which is not. Therefore, it is also possible to understand topics that are of interest to users and are of high utility, and this information can be used for content creation and the development of new services.

[0077] For example, the history analysis unit 26 may also analyze cross-platform use of multiple content items by each user, that is, combinations of multiple content items used by the same user. This allows for a more accurate understanding of topics of high user interest and user activity trends, and this information can be used to create content and develop new services.

[0078] In the content management system 10, multiple pieces of content originating from different content distribution systems 50 are managed together, making it easy to analyze usage even when users access content from different sites.

[0079] Furthermore, the analysis performed by the history analysis unit 26 may also take into account the semantic similarity of the content as defined in the feature data 31. For example, the history analysis unit 26 may use a set of content with high semantic similarity as the target of the cross-use analysis described above.

[0080] Furthermore, if the usage history data 32 includes the user's search results for content and content viewing history, the history analysis unit 26 may include this information in its analysis. This makes it easier to understand topics of high interest to the user and the user's activity trends. Furthermore, if the user is a service provider, the usage history data 32 may include the usage history of content by users of services provided by that service provider. The history analysis unit 26 may then analyze the content usage history of the service users. The usage history of service users is collected by the content management system 10, which, if necessary, cooperates with the user device 60 and the content distribution system 50.

[0081] As described above, the following effects can be obtained according to this embodiment. (1) Feature extraction using machine learning models organizes content according to the meaning it conveys. Therefore, unified management of unstructured content is possible. Furthermore, since content with high semantic similarity to the search criteria is displayed in the search results, appropriate content can be provided.

[0082] (2) If the feature data 31 is data in a high-dimensional vector space in which each of the multiple contents is represented as a feature vector, then content can be accurately organized and searched according to semantic similarity.

[0083] (3) The content management system 10 accepts user requests to use content and inquires with the sender whether or not permission to use the content is granted. Therefore, the burden on the user to obtain permission to use the content is reduced.

[0084] (4) The terms of use for the content are pre-registered in the content management system 10, and the terms of use are presented to the user from the content management system 10. Therefore, it is easy to adjust the terms of use for the content between the user and the content creator. In addition, since the content creator can set detailed terms of use for each piece of content, the use of the content is encouraged to be in line with the creator's intentions, and the adjustment of the above terms can be made more smoothly.

[0085] (5) When the content management system 10 receives notification of an update to content from the sender, it notifies the user using the content of the update. This allows the user to quickly reflect the content update from the sender in the content they are using.

[0086] (6) The content management system 10 analyzes the user's content usage history. This makes it possible to understand topics of high interest to users and user activity trends.

[0087] [Differentiation] The above embodiment can be implemented with the following modifications. The sender does not need to set any terms of use. In other words, the content sending system 50 does not need to input any terms of use into the content management system 10. The usage management unit 24 may also inquire with the sender whether or not they have granted permission to use all of the content that the user wishes to use.

[0088] The content management system 10 only needs to have a content organization function and a content search function, and does not need to have at least one of the following: a content usage management function, a content update management function, and a content usage history analysis function. [Explanation of Symbols]

[0089] 10…Content Management System 20…Reception Department 21...Feature extraction unit 22… Search section 23…Output section 24…Usage Management Department 25…Update Management Department 26…History Analysis Department 30…Registered data 31…Feature data 32…Usage history data 50…Content distribution system 60...User device

Claims

1. A feature extraction unit that uses data of the unstructured content transmitted to the public via a network by multiple different senders and a machine learning model to extract features from the content, using the concepts contained in the content as features, and generates feature data which is data in a high-dimensional vector space in which each of the multiple content items is represented as a feature vector. A search unit calculates a feature vector of search conditions input from a user device, which is a device managed by a user who intends to use the content, and uses the feature data to search for content associated with the features corresponding to the search conditions. An output unit that outputs content information, which is information indicating content corresponding to the search results obtained by the search unit, and usage conditions, which are conditions set by the sender for the use of the content, to the user device. A user management unit receives the user's request to use the content and inquires with the sender whether or not to grant permission for the use of the content, A content management system equipped with the following features.

2. A feature extraction unit generates feature data that associates the content with the features obtained by applying the content to the model, using data of separate content transmitted by multiple different senders and a machine learning model that outputs concepts contained in the content as features, which are content transmitted to the public via a network. A search unit that uses the aforementioned feature data to search for content associated with features corresponding to search conditions entered from a user device, which is a device managed by a user who intends to use the content, An output unit that processes the content corresponding to the search results from the search unit into a display design different from the form transmitted by the sender and outputs it to the user device, and outputs the usage conditions, which are the conditions set by the sender for the use of the content, to the user device. A user management unit receives the user's request to use the content and inquires with the sender whether or not to grant permission for the use of the content, A content management system equipped with the following features.

3. The output unit processes the content corresponding to the search results from the search unit into a display design different from the form transmitted by the sender and outputs it to the user device. The content management system according to claim 1.

4. The search unit searches for content associated with features corresponding to the search conditions entered from a device, including the operating system for services provided using the content, as the user device. The content management system according to claim 1 or 2.

5. The aforementioned feature data is data in a high-dimensional vector space in which each of multiple content items is represented as a feature vector. The content management system according to claim 2.

6. When the sender notifies the user of an update regarding the aforementioned content, the update management unit further provides the user device of the user using the content to notify the user of the update. A content management system according to any one of claims 1 to 5.

7. The system further includes a history analysis unit that analyzes the usage history of the aforementioned content. A content management system according to any one of claims 1 to 6.

8. On one or more computers, The process involves using data of the unstructured content transmitted to the public via a network by multiple different senders, and a machine learning model to extract features from the content, using the concepts contained within the content as features, and generating feature data that is a high-dimensional vector space where each of the multiple content items is represented as a feature vector. The process involves calculating a feature vector of search conditions input from a user device, which is a device managed by the user who intends to use the content, and using the feature data to search for content associated with the features corresponding to the search conditions. Content information, which is information indicating content corresponding to the results of the search, is output to the user device. The user device outputs the usage conditions, which are the conditions set by the sender for the use of the aforementioned content. The system accepts the user's request to use the content and inquires with the sender whether or not permission to use the content is granted. A content management program that executes [this].

9. On one or more computers, The process involves using data of separate pieces of content transmitted to the public via a network by multiple different senders, and a machine learning model that outputs concepts contained in the content as features, applying the content to the model to generate feature data that associates the features obtained with the content. Using the aforementioned feature data, the system searches for content associated with features corresponding to search conditions entered from a user device, which is a device managed by a user who intends to use the content. The content corresponding to the search results is processed into a display design different from the form transmitted by the sender and output to the user device. The user device outputs the usage conditions, which are the conditions set by the sender for the use of the aforementioned content. The system accepts the user's request to use the content and inquires with the sender whether or not permission to use the content is granted. A content management program that executes [this].

10. One or more computers, The process involves using data of the unstructured content transmitted to the public via a network by multiple different senders, and a machine learning model to extract features from the content, using the concepts contained within the content as features, and generating feature data that is a high-dimensional vector space where each of the multiple content items is represented as a feature vector. The process involves calculating a feature vector of search conditions input from a user device, which is a device managed by the user who intends to use the content, and using the feature data to search for content associated with the features corresponding to the search conditions. Content information, which is information indicating content corresponding to the results of the search, is output to the user device. The user device outputs the usage conditions, which are the conditions set by the sender for the use of the aforementioned content. The system accepts the user's request to use the content and inquires with the sender whether or not permission to use the content is granted. A content management method for executing this.

11. One or more computers, The process involves using data of separate pieces of content transmitted to the public via a network by multiple different senders, and a machine learning model that outputs concepts contained in the content as features, applying the content to the model to generate feature data that associates the features obtained with the content. Using the aforementioned feature data, the system searches for content associated with features corresponding to search conditions entered from a user device, which is a device managed by a user who intends to use the content. The content corresponding to the search results is processed into a display design different from the form transmitted by the sender and output to the user device. The user device outputs the usage conditions, which are the conditions set by the sender for the use of the aforementioned content. The system accepts the user's request to use the content and inquires with the sender whether or not permission to use the content is granted. A content management method for executing this.