JP7481772B1 - Information output device, information output program, information output method, and information output system - Google Patents

Abstract

[Problem] To provide service providers with a technology for utilizing information posting sites that can further increase their ability to attract customers [Solution] An information output device 104 includes an information acquisition unit 112 that acquires information posted on the network by other users, an information generation command unit 114 that generates information similar to the information acquired by the information acquisition unit 112, and an output unit 116 that outputs the information generated in response to a command from the information generation command unit 114. [Selected Figure] Figure 2

Description

The present invention relates to an information output device, an information output program, an information output method, and an information output system.

Conventionally, a method for managing information posting sites including SNS (Social Network Service) and/or review sites has been known (for example, Patent Document 1).

JP 2023-056579 A

Recently, in the service industry, for example restaurants, efforts are being made to increase customer attraction by using information posting sites. In particular, many service providers use multiple information posting sites to disseminate information to a variety of customer segments.

The purpose of this disclosure is to provide service providers with technology for utilizing information posting sites that can further increase their ability to attract customers.

According to one aspect of the present disclosure, the device includes an information acquisition unit that acquires information posted on a network by other users, an information generation command unit that generates information similar to the information acquired by the information acquisition unit, and an output unit that outputs the information generated in response to a command from the information generation command unit.

1 shows an overview of an information output system to which an information output device according to an embodiment is applied. A block diagram of the information output device is shown. FIG. 2 is a flow chart showing a series of operations of the information output system. 1 shows an example of an input screen. 1 shows an example of an input screen.

Below, an embodiment of the present disclosure will be described. The information output device according to the embodiment is merely an example, and each configuration can be modified as appropriate without departing from the spirit of the present disclosure. In particular, in the following embodiment, the hardware and functional blocks for realizing the information output device will be clearly explained in detail. However, it is clear that the function that each piece of hardware realizes can be changed as appropriate in accordance with the spirit of the present disclosure. It is also clear that each functional block can be realized by the same hardware or software, or different hardware or software. Note that in the following explanation, it is assumed that a control program installed in the information output device controls each piece of hardware.

Figure 1 shows an overview of an information output system to which an information output device is applied. The information output system 100 includes a user terminal 102, an information output device 104, an SNS server 106, and an information generation device 108. The user terminal 102, the information output device 104, the SNS server 106, and the information generation device 108 are connected via a network 110. The information output device 104 and the SNS server 106 function as application servers, as will be described in detail later. Therefore, a web server and/or a database server, etc. are also required to implement the present disclosure, but for convenience of explanation, these publicly known technologies will be omitted.

The user terminal 102 is configured with a screen-equipped information processing device such as a user's personal computer or smartphone. In this embodiment, a series of information output processes are executed while accessing the information output device 104 using a browser installed on the user terminal 102. In other words, in this embodiment, the user terminal 102 functions only as a user interface, and a series of information output processes are executed on the information output device 104. Note that an application capable of executing a series of information output processes may be installed on the user terminal 102, and the user terminal 102 may directly access the information output device 104, the SNS server 106, and the information generating device 108.

A user refers to a person who uses a user terminal 102. In this embodiment, a user refers to a person who has an SNS account and wishes to post new information to his or her SNS account. For example, if a restaurant manager wishes to post new information to an SNS account in the name of the restaurant that he or she manages, the restaurant manager becomes the user.

The information output device 104 functions as an application server that executes an information output program. The information output device 104 is realized by a hardware processor such as a CPU (Central Processing Unit) executing the information output program (software). When the information output program is executed, various information is displayed on the browser of the user terminal 102 via the network 110. Details of the information output device 104 and the information output program for operating the information output device 104 will be described later.

SNS server 106 is an operation server (i.e., an application server) for the SNS site. An SNS site should be interpreted broadly as a site on which various contributors can post information, regardless of whether it is membership-based or non-membership-based, anonymous or named or real-named. Furthermore, a site includes not only websites displayed by a browser, but also the original data of app screens displayed by application programs. SNS server 106 is a general server that manages information for operating an SNS site.

The information generating device 108 is a device or system that uses AI (Artificial Intelligence). The information generating device 108 uses a language model that has learned information sites including countless SNS sites, and outputs a response to an input request. As the information generating device 108, for example, a device or system in which a tool using a machine learning model such as a large-scale language model (LLM: Large Language Models) such as the well-known Chat GPT can be used. In the embodiment of the present disclosure, the machine learning model is one that is trained to output text generated by a computer based on text information posted to the SNS server 106. Note that the machine learning model is not limited to a large-scale language model, and may be any model that can be automatically generated by AI, and is not particularly limited to this.

The network 110 includes, for example, the Internet, a wide area network (WAN), a local area network (LAN), a provider terminal, a wireless communication network, a wireless base station, a dedicated line, and the like. Note that it is not necessary for all combinations of the devices shown in FIG. 1 to be able to communicate with each other, and the network 110 may include a local network in part.

2 shows a block diagram of the information output device. The information output device 104 includes an information acquisition unit 112, an information generation command unit 114, an output unit 116, a posting unit 118, an SNS information assignment unit 120, and a control unit 122. Some or all of the information acquisition unit 112, the information generation command unit 114, the output unit 116, the posting unit 118, and the SNS information assignment unit 120 may be realized by hardware (including circuitry) such as an LSI (Large Scale Integration), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a GPU (Graphics Processing Unit), or may be realized by a combination of software and hardware. In this embodiment, the information acquisition unit 112, the information generation command unit 114, the output unit 116, the posting unit 118, and the SNS information assignment unit 120 are operated by the control unit 122 supplying commands to each unit based on the information output program.

The information acquisition unit 112 acquires information posted on the network. More specifically, the information acquisition unit 112 acquires various pieces of information posted on the SNS server 106. In order for the information acquisition unit 112 to acquire information posted on the SNS server 106, a publicly known API (Application Programming Interface) technology can be used. APIs are installed in various existing SNS servers, so detailed explanations will be omitted. Information acquired by the information acquisition unit 112 includes information posted on the network 110 by various users. Information posted on the network 110 includes SNS information posted on the SNS server 106. Note that there may be multiple SNS servers on the system 100, and in this case, the information acquisition unit 112 may acquire information from multiple SNS servers. The information acquisition unit 112 may acquire information only from SNSs posted by other users, or may extract information posted by other users from all acquired SNSs. The information acquisition unit 112 may also extract only information posted by people related to the user. A person related to a user refers to a person who may compete with the user's SNS. A person who may compete with the user's SNS is assumed to be a person who posts SNS within a pre-designated area or a person who is engaged in the same type of business as the user's SNS.

Whether or not a person is posting on social media within a pre-specified area can be determined based on whether or not the person is doing business in the same or adjacent administrative district as the address registered by the user, or whether or not the person is doing business within a specified distance from the address registered by the user. Adjacent administrative districts can be easily identified by referring to map data present on the network. In addition, a person posting on social media within a pre-specified area may be searched for using an address that excludes lower administrative districts from the address registered by the user. As an example, if the address registered by the user is "1-1-1 Shibuya, Shibuya-ku, Tokyo," it is assumed that a search will be performed using an address that excludes "1-go," "1-banchi," "Shibuya," and other lower administrative districts. The same type of business refers to a similar business type, a business type with overlapping demanders, or a business type that is a superordinate concept of a specific business type. In addition, a person posting on social media using the same hashtag or similar terms (especially nouns) may be considered to be doing the same type of business. The above determination can be made based on information registered in an existing map application such as Google (registered trademark) Map. When registering a store such as a restaurant in a map application, registration of the store's business type may be required. Therefore, the information acquisition unit 112 can check those who may be in competition with the user's SNS by using the information about the business type and the information about the address registered in the map application. In other words, it can be said that the information acquisition unit 112 functions as a search unit that searches the SNS of those who may be in competition with the user.

In another aspect, the information acquisition unit 112 may utilize AI to acquire information posted on the network. In this case, the information acquisition unit 112 may utilize an external device (e.g., the information generating device 108) to transmit a request to acquire the SNS of nearby competing stores, and receive a reply to that request, thereby acquiring the SNS of potential competitors.

The information generation command unit 114 generates information similar to the information acquired by the information acquisition unit 112. Specifically, the information generation command unit 114 generates a predetermined command (a so-called "prompt"). In other words, the information generation command unit 114 can be said to function as a prompt generation unit. The command (prompt) preferably includes an instruction to generate information with similar content as a context. The content of the context is not limited to this, and may be any instruction to generate information based on information ultimately posted by other users. The command may include the content of the information acquired by the information acquisition unit or information indicating the location of the information (URL: Uniform Resource Locator, etc.). The command may include additional information such as a limit on the number of characters of the information to be generated and an instruction on the writing style (including an instruction by attaching an example sentence). The instruction on the writing style may be an instruction to refer to information posted by the user in the past. By using information that the user himself/herself has created in the past as example sentences, it is possible to prevent the format of the sentences posted by the user (tone of voice, writing style, use of special characters, format, etc.) from changing each time a sentence is posted. In addition, by using information from popular articles as example sentences, it is possible to generate articles with a similar sentence format to popular articles, which is expected to attract more attention. The information generation command unit 114 transmits the generated command to the information generation device 108. The information generated by the information generation device 108 in response to this command is transmitted to the information output device 104.

When multiple SNSs are acquired by the information acquisition unit 112, the information generation command unit 114 may generate an instruction to extract SNSs with a high (large) number of ratings and generate information with content similar to the content of the extracted SNSs. The number of ratings on an SNS refers to at least one of the engagement rate, which indicates the proportion of users who responded to an SNS post, the number of good ratings, and the number of comments. The number of ratings may also refer to SNS information. In this case, the information generation command unit 114 extracts SNSs that include more SNS information that is used frequently (used many times). The SNS information here refers to information indicating keywords (specific nouns, etc.) that indicate the content posted on the SNS, such as hashtags and meta tags.

If the information generating device 108 is a system that uses an image generation AI or a voice generation AI, the command generated by the information generation command unit 114 may include a command to generate related image information or voice information. If the information generating device 108 is a system that uses only a text generation AI, an information generating device other than the information generating device 108 that uses an image generation AI or the like may also be used in combination to generate image information or voice information.

In another embodiment, the information generation command unit 114 may cause the information generating device 108 to acquire information posted on the network. In other words, the information generating device 108 takes over the function of the information acquiring unit 112. In this case, it is not necessary to provide the information acquiring unit 112 in the information output device 104. Also in this case, the information generating device 108 may be provided with the function of the search unit described above.

The output unit 116 outputs the information generated in response to the command from the information generation command unit 114 to the user. Specifically, the output unit 116 transmits the information generated by the information generation device 108 to the user terminal 102. The information transmitted from the output unit 116 is displayed on the browser of the user terminal 102. At this time, it is preferable that the information transmitted from the output unit 116 is in a format that can be edited by the user terminal 102. This allows the user to edit the contents of the information using the user terminal 102.

The posting unit 118 posts information generated by the information generating device 108 or information edited using the user terminal 102 to the SNS server 106 in response to instructions from the user.

The SNS information adding unit 120 adds SNS information to the information generated by the information generating device 108. The SNS information adding unit 120 refers to the tag added to the information acquired by the information acquiring unit 112, and adds the same tag as the referenced tag to the information generated by the information generating device 108. When adding tags, the SNS information adding unit 120 may extract tags added to the information acquired by the information acquiring unit 112, and add only tags that have been used a certain number of times or more (are popular) on the SNS server 106. The tags added by the SNS information adding unit 120 are preferably output in a format that can be edited by the user.

The control unit 122 operates each of the information acquisition unit 112, information generation command unit 114, output unit 116, posting unit 118, and SNS information assignment unit 120 in accordance with the information output program installed in the information output device 104.

Next, a series of operations of the information output system 100 will be described. FIG. 3 is a flow diagram showing a series of operations of the information output system. In the following example, it is assumed that user A runs Cafe X in District A. It is assumed that user A periodically posts information on SNS for the purpose of promoting Cafe X. In this example, the information output system 100 can also be said to be an automatic SNS generation system.

When posting new information on the SNS, User A operates the browser of the user terminal 102. As a result, in step S1, the user terminal 102 accesses the information output device 104 via the network 110.

In step S2, the information output device 104 displays an input screen for SNS information on the user terminal 102. An example of the input screen is shown in FIG. 4. The input screen displays a post display box 202, a post instruction button 204, and an automatic post generation instruction button 206. The post display box 202 is an input box into which the user types the content to be posted on the SNS site. When the post instruction button 204 is clicked while text has been entered in the post display box 202, the content entered in the post display box 202 is sent to the SNS server 106. The automatic post generation instruction button 206 is an instruction section for causing the information output program to generate the content to be posted.

When the automatic post generation instruction button 206 is clicked, the information output device 104 starts generating the post content in the processing from step S3 onwards.

In step S3, the information acquisition unit 112 accesses the SNS server 106 and acquires information posted to the SNS server 106 by other users. When extracting only those who may be competitors with Person A, in step S3 the information acquisition unit extracts the SNSs of those who may be competitors with Person A from among multiple SNSs. Specifically, the information acquisition unit 112 extracts the SNSs of restaurants operating within a specified distance (for example, within a radius of 500 m) from Cafe X, or the SNSs of restaurants operating in District A where Cafe X is located and in administrative districts adjacent to District A.

In step S4, the information generation command unit 114 sends a command to the information generating device 108 to generate information similar to the information acquired by the information acquisition unit 112. If multiple SNSs are extracted in step S3, the information generation command unit 114 may extract the SNS with the largest number of ratings from the multiple SNSs, and cause the information generating device 108 to generate an SNS article with content similar to the extracted information.

In step S5, the information generating device 108 generates an SNS article based on the command sent from the information generation command unit 114. Naturally, the content of the generated SNS article changes from moment to moment depending on the proficiency of the machine learning model of the information generating device 108. Therefore, from a certain perspective, it is possible that the SNS article generated by the information generating device 108 is not actually similar to the SNS article referenced. However, whether the generated SNS article is similar to the referenced SNS article is not an issue of this concept. In other words, what is important in this disclosure is whether the command sent from the information generation command unit 114 includes a command to "generate a similar SNS article." Next, in step S6, the information generating device 108 transmits the generated SNS article to the information output device 104.

In step S7, the SNS information assigning unit 120 assigns tags to the SNS article received from the information generating device 108. The tags assigned to the SNS article at this time are the most popular tags among the tags assigned to the referenced SNS article.

In step S8, the output unit 116 displays the contents of the tagged SNS article on the user terminal 102. FIG. 5 shows an example of a screen displayed on the user terminal. Each area shown in FIG. 5 is the same as the area shown in FIG. 4.

As shown in FIG. 5, the post display box 202 displays the contents of the SNS article generated by the information generating device 108. The contents of the post display box 202 are displayed in a format that can be edited by operating the user terminal 102. This allows the user to change the contents of the SNS article as appropriate. As shown in the figure, the contents of the tags are also displayed in an editable format.

If the SNS article displayed in the post display box 202 does not suit the user's taste, the user can operate the user terminal 102 to click the post automatic generation instruction button 206 again. When the post automatic generation instruction button 206 is clicked, the information output system 100 repeats the processes of steps S3 to S8. This makes it possible to automatically generate an SNS article that suits the user's taste.

When the user decides that he/she wishes to post the SNS article displayed in the post display box 202 on the SNS and clicks the post instruction button 204, the posting unit 118 transmits the SNS article information to the SNS server 106 in step S9. As a result, the generated SNS article is posted and made viewable by third parties.

As described above, according to the embodiment, SNS articles can be generated with simple operations. The generated SNS articles are generated taking into account the popular content at the time of article generation, so the generated SNS articles are expected to attract more attention.

This disclosure is not limited to the above-described embodiments, and each configuration of the embodiments can be modified as appropriate.

The following variations are envisioned:

Image information such as photographs may be stored in an appropriate location on the user terminal 102 or the like, and SNS information may be added to the image information in advance. In this case, when the content of the SNS article generated by the information generating device 108 is displayed in the post display box 202, the image information with the associated tags added may be read out and displayed together with the SNS article.

100 Information output system; 104 Information output device: 108 Information generation device: 112 Information acquisition unit: 114 Information generation command unit: 116 Output unit.

Claims (13)

An information output device that outputs content posted by a user to an SNS,
an information acquisition unit that acquires information posted on a network by users other than the user ;
an information generation command unit that causes an information generation device capable of generating information to be posted on an SNS to generate information having content similar to the information acquired by the information acquisition unit using a machine learning model ;
an output unit that outputs information generated in response to a command from the information generation command unit ;
The information generation command unit generates information having contents similar to information having a certain number of evaluations or more among the information acquired by the information acquisition unit.
Information output device. The information output device according to claim 1, wherein the number of ratings includes at least one of the engagement rate, the number of likes, and the number of comments of the information posted on the network by other users. The information posted on the network by other users is information posted on a social networking site,
The information output device according to claim 1 , wherein the number of evaluations includes a number of times that SNS information of the information posted on a network by other users has been used. The information output device according to claim 3, further comprising an attachment unit that attaches SNS information that has been used a certain number of times or more to information generated in response to a command from the information generation command unit. The information output device according to claim 1, wherein the output unit outputs the information generated in response to a command from the information generation command unit in an editable format. The information output device according to claim 1, wherein the information generation command unit causes a machine learning model trained to output information including computer-generated text based on text information to generate information. The information output device according to claim 1, wherein the information generation command unit causes a machine learning model trained to output information including at least one of computer-generated images, computer-generated videos, and computer-generated sounds to generate information. The information output device according to claim 1, wherein the other users are users who are in a pre-specified area. The information output device according to claim 1, wherein the other user is a user who competes with the user of the information output device. A program for operating an information output device that outputs content posted by a user to an SNS ,
The information output device includes:
Allowing a user other than the user to acquire information posted on a network;
outputting a command to generate information similar to the acquired information to an information generating device capable of generating information to be posted on an SNS using a machine learning model ;
outputting the information generated by the information generating device ;
generating information having contents similar to information having a certain number of evaluations or more among the information acquired by the information output device, using the information generation device;
Information output program. An information output method for outputting content posted by a user on an SNS, the information output method being executed by an information output device having a processor, the information output method comprising:
acquiring information posted on a network by a user other than the user ;
outputting a command to generate information similar to the acquired information to an information generating device capable of generating information to be posted on an SNS using a machine learning model ;
outputting information generated in response to a command from the information generating device ;
generating information having contents similar to information having a certain number of evaluations or more among the acquired information;
An information output method comprising: An information output system that outputs content posted by a user to an SNS,
an information acquisition unit that acquires information posted on a network by users other than the user ;
an information generation command unit that causes an information generation device capable of generating information to be posted on an SNS to generate information having content similar to the information acquired by the information acquisition unit using a machine learning model ;
an information generating unit that generates information in response to a command from the information generation command unit;
an output unit that outputs the information generated by the information generation unit,
The information generation command unit generates information having contents similar to information having a certain number of evaluations or more among the information acquired by the information acquisition unit.
Information output system. An information output device that outputs content posted by a user to an SNS,
an information generation command unit that causes an information generation device capable of generating information to be posted on an SNS to acquire information posted on a network by a user different from the user , and generate information having content similar to the acquired information using a machine learning model;
an output unit that outputs information generated in response to a command from the information generation command unit;
The information generation command unit generates information having contents similar to information having a certain number of evaluations or more among the information acquired by the information generating device.
Information output device.