Creation support device, creation support method, program, and system
The creation support system addresses the challenge of identifying suitable creators by using tag extraction and semantic matching to create planning documents that meet content intentions.
Patent Information
- Application Number
- JP2025073230
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2025-04-25
- Publication Date
- 2025-07-03
- Estimated Expiration
- 2040-12-25
Smart Images

Figure 2025100922000001_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a creation support device, a creation support method, a program, and a system.
Background Art
[0002] Conventionally, the production (publication, broadcasting, etc.) of content has been carried out through the collaborative work of creators and editors. Specifically, editors grasp the orientation and characteristics of creators, receive the demands of the market and fans, give appropriate advice to creators, and provide content demanded by the market and fans.
Prior Art Documents
Patent Documents
[0003]
Patent Document 1
Summary of the Invention
Problems to be Solved by the Invention
[0004] Those who produce events, broadcasts, and content, such as publishers and content distributors, do not fully grasp who the most suitable creators for their intended events are and where they are located. In addition, with the spread of the Internet, anyone can transmit information, and new creators are emerging one after another. Therefore, the value and importance of editors are increasing.
[0005] Currently, creators directly transmit to their followers through social network services (SNS) and video distribution. However, there has been a problem that it is actually difficult for editors to grasp all the content transmitted by creators on SNS and the like. At present, the followers of each creator are the ones who best understand the orientation and characteristics of the creator.
[0006] Patent Document 1 discloses a matching system that links the authors of e-books with anime and manga creators. Patent Document 1 holds author information including sample images of anime and manga creators in a database, and there is no mechanism to follow the information transmitted by the creators and update the author information.
[0007] The present invention has been made in view of the above, and an object thereof is to assist in creating a planning document that meets the intention.
Means for Solving the Problems
[0008] The creation support device according to one aspect of the present invention is a creation support device that selects a person suitable for a plan from a plurality of persons and supports the creation of a planning document, and acquires the information of the person and the tag assigned to the person from a server that holds tags, which are character strings assigned to each of the plurality of persons. A selection unit that selects a person to whom a tag related to a keyword indicating the planning content described in the template prepared for creating the planning document is assigned, and a creation unit that creates the planning document by describing the information of the selected person in the part where the person information in the template is described.
Effects of the Invention
[0009] According to the present invention, it is possible to assist in creating a planning document that meets the intention.
Brief Description of the Drawings
[0010]
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Mode for Carrying Out the Invention
[0011] [Configuration of Creation Support System] Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[0012] FIG. 1 is a diagram showing an example of the overall configuration of the creation support system of the present embodiment. The creation support system of the present embodiment includes a creation support device 10 and an extraction server 20, and is a system that supports the creation of a planning document.
[0013] It is assumed that the creator routinely posts text, images, music, and videos from the terminal 50 to the SNS service 30 and transmits them to followers. A follower is a person registered so as to be able to receive the transmissions of the creator in the SNS service 30, and is often a fan of the creator, a person who pays attention to the creator, a person the creator is interested in, etc.
[0014] The extraction server 20 assigns tags to the creator based on the content posted on the SNS service 30 of the creator. A tag is additional information representing the attributes, characteristics, areas of interest, or fields of expertise of the creator. For example, if a certain creator's posts on the SNS service 30 often contain a lot of information about dogs, the tag "dog lover" is assigned to that creator. The tags assigned to a creator are not limited to one, and multiple tags may be assigned. Not only information representing the creator, but also information representing the group of followers of the creator may be assigned to the creator as tags. For example, if many of the followers who commented on the content posted by the creator are male in their teens, the tags "male in their teens" or "popular among males in their teens" are assigned to the creator.
[0015] The creation support device 10 inputs the tags assigned to the creator and the template of the proposal, extracts the creator suitable for the plan, and supports the creation of the proposal.
[0016] Next, referring to FIG. 2, the configuration of the creation support device 10 will be described. The creation support device 10 shown in FIG. 2 includes an acquisition unit 11, a matching unit 12, and a creation unit 13. Each unit included in the creation support device 10 may be configured by a computer including an arithmetic processing device, a storage device, etc., and the processing of each unit may be executed by a program. This program is stored in the storage device included in the creation support device 10, and it is also possible to record it on a recording medium such as a magnetic disk, an optical disk, or a semiconductor memory, or to provide it through a network. The creation support device 10 may be configured as a virtual server on the cloud, or may be configured by a computer terminal such as a personal computer or a mobile terminal.
[0017] The acquisition unit 11 acquires the information of the creator and the tags assigned to the creator from the extraction server 20. The extraction server 20 extracts the tags to be assigned to the creator from the posts of the creator acquired from the SNS service 30 and transmits them to the acquisition unit 11. The extraction server 20 may extract the tags using an extraction model representing the relationship between the posts of the creator and the tags, which is created using machine learning. The acquisition unit 11 may have the functions of the extraction server 20.
[0018] The matching unit 12 inputs a template describing the planning content, compares the information of the template with the tags, and extracts the creators to whom tags similar to the template are assigned. Keywords related to the plan may be described in the template. The matching unit 12 quantifies the semantic distance between the keyword and the tag, and extracts the creators to whom tags with a close distance are assigned. For example, when the keyword "pet" is described in the template, not only the creators assigned the tag "pet", but also the creators assigned tags semantically close to "pet" such as the keywords "dog" and "cat" are extracted. Technologies used in natural language processing can be utilized to calculate the distance between the keyword and the tag.
[0019] When a plurality of keywords are described in the template, the matching unit 12 may extract the creators to whom tags close to all of the keywords are assigned, or may extract the creators to whom tags close to some of the keywords are assigned. The matching unit 12 calculates, for each creator, a numerical value indicating the proximity to the keyword for each of the assigned tags, sums up the obtained numerical values to calculate the score of the creator, and may extract the creator with a high score.
[0020] The matching unit 12 may perform morphological analysis on the sentence representing the planning content described in the template to create keywords.
[0021] The creation unit 13 creates a planning document from the template and the information of the extracted creator. For example, the information of the creator is described in a predetermined location in the template.
[0022] [Generation of Extraction Model] FIG. 3 is a diagram showing an example of generation of an extraction model by the extraction server 20. The extraction server 20 inputs the transmission content of the creator and the tag (correct data) assigned to the creator as training data, and generates an extraction model by machine learning. When the transmission content of the creator on the SNS service 30 is input to the extraction model, the tag to be assigned to the creator is output. The extraction server 20 may input the information of the creator's follower group and the tag representing the follower group as training data. Thereby, the characteristics of the people who follow the creator can be assigned as tags of the creator.
[0023] The extraction server 20 periodically acquires the transmission content of the creator from the SNS service 30, inputs the speech content of the creator into the extraction model, and assigns a tag to the creator. For example, in the SNS service 30 where text and photos can be posted, the text and photos posted by the creator are input into the extraction model to obtain the tags to be assigned to the creator. In the SNS service 30 where videos can be posted or live broadcasts can be performed, the text obtained by performing speech recognition from the video, the screen shot of the video, and the video itself are input into the extraction model to obtain the tags to be assigned to the creator. The information of the creator's followers or the information of the followers who replied to the creator's speech may be input into the extraction model to obtain the tags to be assigned to the creator. As the information of the followers, the profile of the followers on the SNS service 30 or the transmission content of the followers can be used. The transmission content of the creator on a plurality of different SNS services 30 may be input into the extraction model. Since each SNS service 30 has its own characteristics, by inputting the transmission content on different SNS services 30, tags reflecting various aspects of the creator can be assigned.
[0024] The extraction server 20 returns the information of the creator and the tag in response to the request of the creation support device 10. An example of the information of the creator and the tag is shown in FIG. 4. As shown in FIG. 4, the tag assigned to the creator is not limited to one.
[0025] [Template of the project plan] Next, an example of the template of the project plan will be described.
[0026] Fig. 5 is a diagram showing an example of the template of the project plan. The template shown in this figure has an outline that briefly explains the plan and five items that are the components of the plan: "when", "where", "who", "what", and "how". The outline briefly explains the plan so that the intention and appeal of the plan can be conveyed. "When" indicates the information of the date and time of the event, "where" indicates the information of the venue or the media for the event (for example, the video site to be used), "who" indicates the information of the participating creators, "what" indicates the content of the plan (for example, what is done in the plan), and "how" indicates the information such as the scale, budget, and expected number of viewers of the plan. Among the five items, the "who" item is blank. Keywords indicating the characteristics of the creators who are desired to participate may be described in the "who" item.
[0027] When the creation support device 10 inputs the template of Fig. 5, it extracts keywords from the content described in the outline of the template. Keywords may also be extracted from each item of the components of the plan. For example, when a planner plans content related to a sweets store where pets are allowed to enter, keywords such as "pet" and "sweets" are described in the template.
[0028] The creation support device 10 calculates the distance between the keyword and the tag assigned to each creator and sums up the calculated distances. For example, creators with tags such as "raising a dog" that are semantically close to "pet" or "sweet tooth" that are semantically close to "sweet" will have a higher total value. The creation support device 10 creates a planning document by listing the creators with high total values in the "Who" item of the template. The creation support device 10 may list multiple creators in the "Who" item, and the planner of the proposal may select a creator. Also, the creation support device 10 may determine the hosting medium according to the selected creator and list it in the "Where" item of the template. For example, the creation support device 10 determines to distribute the content on a video site with a large number of followers of the selected creator, and creates a planning document by applying that video site to the template.
[0029] FIG. 6 is a diagram showing an example of a template for planning broadcast content distributed via a network. The template shown in the figure shows the timeline of the broadcast content and the information of desirable creators for each timeline. The information of the desirable creators includes the content of the plan for the corresponding time period. A keyword representing the desirable creator may be described in the information of the desirable creator.
[0030] When the creation support device 10 inputs the template of FIG. 6, it extracts keywords from the information of the desirable creators for each time period. The creation support device 10 calculates the distance between the keyword and the tag assigned to the creator for each time period and each creator, and sums up the calculated distances. The creation support device 10 creates a planning document listing the creators with high total values as the creators suitable for that time period. FIG. 7 shows an example of a planning document created by the creation support device 10 from the template of FIG. 6.
[0031] Note that the template described here is just an example and is not limited to this.
[0032] Further, an appropriate template may be prepared according to the planning content and the target object, and the creation unit 13 may be configured so that the target object can be generated from the template.
[0033] [Operation of Creation Support System] Next, with reference to the flowchart of FIG. 8, the processing flow of the creation support system will be described.
[0034] In step S11, the extraction server 20 collects the creator's posted content from the SNS service 30 and extracts the tags assigned to the creator. Note that the processing in step S11 may be performed periodically or when creating a planning document with the creation support device 10. The extraction server 20 may hold account information for each SNS service 30 for each creator and collect the creator's posted content from multiple SNS services 30.
[0035] In step S12, the creation support device 10 inputs a template, obtains the creator and the tags assigned to the creator from the extraction server 20, compares the template information with the creator's tags, and selects a creator suitable for the template.
[0036] In step S13, the creation support device 10 creates a planning document using the template and the creator selected in step S12.
[0037] [Application to Crowdfunding] Next, an example of applying the creation support system of the present embodiment to crowdfunding will be described. Crowdfunding is a mechanism for raising a small amount of funds from an unspecified number of people who support a project via the Internet.
[0038] Currently, there are many crowdfunding services, and anyone can become a proposer and seek support for various types of projects. In crowdfunding services, support is being sought for a large number of projects. Among these various projects, there are those that have similarities and are thought to yield better results by being advanced jointly. By using the creation support system of this embodiment, it is possible to support the creation of a new project that combines projects that are thought to yield better results by being advanced jointly. For example, using the creation support system of this embodiment, a proposal is made for a plan that combines a project that wants to hold a live event in Shinjuku and a project that wants to hold a live event in Shibuya. Hereinafter, specific examples of using the creation support system of this embodiment will be described.
[0039] The extraction server 20 extracts tags assigned to each project from the crowdfunding service. For tag extraction, the project description text may be used, or the profile of the project proposer may be used.
[0040] The proposer designates their own project registered in the crowdfunding service and inputs the information of that project into the creation support device 10 as a template. The information input when registering the project in the crowdfunding service may be used as a template.
[0041] The creation support device 10 compares the template with the project tags, selects a project close to the template, and creates a proposal document for a new project that combines the proposer's project and the selected project.
[0042] The proposer sends the proposal document for the new project to the proposer of the project selected by the creation support device 10 and proposes to advance the project jointly.
[0043] In addition, a crowdfunding service may be equipped with the functions of the creation support system, use one of the projects registered in the crowdfunding service as a template, extract projects similar to the project, create a business plan for a new project, send the business plan to the originators of these projects, and propose to promote the projects jointly. Also, when a new project is registered, the crowdfunding service may use the information of the new project to be registered as a template, extract already registered projects similar to the new project, and propose to the originators of these projects to promote the projects jointly.
[0044] [Modification Example] In each of the above embodiments and implementation manners, an example has been described in which the system performs matching for an optimal business plan or an optimal matching with other crowdfunding using tag information of selection candidates such as creators and crowdfunding. In practicing the present invention, not limited to the above tag information, various information (hereinafter referred to as "feature data" including tag information) that selection candidates such as creators and crowdfunding have or are attached to the selection candidates can also be used for matching, and any implementation is included in the present invention.
[0045] As characteristic data, for example, hashtag ("#") attached to messages posted by creators or related parties of crowdfunding on SNS, etc., the field of works, business field, or region of residence to which the creator or crowdfunding belongs, the region where the event is held, the age group, occupation or work history (career) of the creator or related parties of crowdfunding, words and phrases frequently appearing in messages posted by the creator or related parties of crowdfunding on SNS, etc., slogans, policies, orientations, etc. of language information such as declarations made by the creator or related parties of crowdfunding themselves or given by others, photo, video, live distribution posted on SNS, etc., financial information such as sales scale, sales forecast, sales performance, number of copies issued, actual amount of fundraising and target amount of fundraising of the creator or crowdfunding, or information such as other creators, publishers, companies, celebrities, local governments, or number of followers who support the creator, information such as other crowdfundings, publishers, companies, celebrities, or local governments that have conducted fundraising or provided support to the crowdfunding may also be used.
[0046] When selecting a candidate suitable for the project plan from among creator candidates, the affinity with the current project plan may be determined by using the project plan itself adopted by a certain creator in the past as characteristic data.
[0047] In the pairing of crowdfundings, the presence or absence and number of people who are donating to both crowdfundings may be used as characteristic data. Furthermore, when pairing crowdfundings, the affinity may be determined by using the project plans, concept papers, etc. of each crowdfunding as characteristic data.
[0048] Alternatively, information obtained by analyzing image information posted on SNS, etc. may be used as characteristic data. For example, when there are many photos of forests or lawns, it is conceivable that the system determines that there is a good affinity with companies that use green as their corporate color, and that there is a good affinity with fields that are linguistically imagined as green, such as law or tranquility.
[0049] It is also possible to obtain the distance from a template by using the content transmitted by a creator or crowdfunding itself as feature data, or to use the vector representation of the transmitted content as feature data. Any form of data may be used as feature data as long as it is information obtained from the transmitted content.
[0050] As described above, the creation support system according to the present embodiment includes an extraction server 20 that extracts tags assigned to each creator from the content transmitted by each creator to the SNS service 30, and compares the keywords and tags described in the template prepared for creating a project plan, and a matching unit 12 that selects a creator to whom a tag close to the information of the template is assigned, and a creation unit 13 that creates a project plan using the template and the creator selected by the matching unit 12. The creation support system assigns tags representing the creator based on the content transmitted by the creator on a daily basis, and selects a creator to whom a tag close to the information of the template prepared for creating a project plan is assigned, so that it is possible to support the creation of a project plan as intended.
Explanation of Signs
[0051] 10…Creation support device 11…Acquisition unit 12…Matching unit 13…Creation unit 20…Extraction server 30…SNS service 50…Terminal
Claims
1. A creation support device that selects a person suitable for a plan from a plurality of persons and supports the creation of a plan document, an acquisition unit that acquires the information of the person and the tag assigned to the person from a server that holds tags, which are character strings assigned to each of the plurality of persons, a selection unit that selects the person to whom a tag related to a keyword indicating the plan content described in a template prepared for creating the plan document is assigned, and a creation unit that creates the plan document by writing the information of the selected person in a place where the person information in the template is written. Creation support device.
2. The creation support device according to claim 1, wherein the selection unit selects a person to whom the same tag as the keyword is assigned. Creation support device.
3. The creation support device according to claim 1 or 2, wherein the creation unit writes the information of a plurality of the persons in a place where the person information in the template is written. Creation support device.
4. The creation support device according to any one of claims 1 to 3, wherein the template includes plan content for each time period, and the selection unit selects a person to whom a tag related to a keyword indicating the plan content for the time period is assigned for each time period. Creation support device.
5. A creation support method that selects a person suitable for a plan from a plurality of persons and supports the creation of a plan document, wherein a computer acquires the information of the person and the tag assigned to the person from a server that holds tags, which are character strings assigned to each of the plurality of persons, selects a person to whom a tag related to a keyword indicating the plan content described in a template prepared for creating the plan document is assigned, and creates the plan document by writing the information of the selected person in a place where the person information in the template is written. Creation support method.
6. A program for operating a computer as a creation support device that selects a person suitable for a plan from a plurality of persons and supports the creation of a plan document, the process of acquiring the information of the person and the tag assigned to the person from a server that holds tags, which are character strings assigned to each of the plurality of persons, and the process of selecting a person to whom a tag related to a keyword indicating the plan content described in a template prepared for creating the plan document is assigned. A program that causes a computer to execute a process of creating the planning document by writing information of the selected person in a location for writing person information in the template. Program.
7. A system comprising a server and a creation support device for supporting the creation of a planning document, wherein the server inputs posting contents posted on a social network service and extracts tags to be assigned to each of a plurality of persons by inputting posting contents posted by each of the plurality of persons to an extraction model that outputs a tag which is a character string for the posting contents, the creation support device includes an acquisition unit that acquires the information of the person and the tag assigned to the person from the server, a selection unit that selects a person to whom a tag related to a keyword indicating the planning content described in a template prepared for creating the planning document is assigned, and a creation unit that creates the planning document by writing the information of the selected person in a location for writing person information in the template, wherein the extraction model is a machine learning model that has learned, as teacher data, posting contents posted on a social network service and tags representing the posting contents System.
Citation Information
Patent Citations
Fixed document generation device / Method and storage medium recording its control program
JP2000099507A
Method and system for carrying out computerized commerce
JP2001312619A
System and method for supporting project and recording medium
JP2002007655A
Device, system, and program for supporting preparation of marketing tool
JP2003316939A
Creator selection system
JP2004192364A