Information processing device and program

Abstract

This increases the stability of the research process and reduces costs and effort. [Solution] The processor 11 of the information processing device 1 functions as a means 26 for extracting secondary users from primary users based on account type and tertiary users from secondary users based on the number of posts, a means 27 for requesting the SNS server device 4-1 to send posted data of tertiary users, a means 29 for sending a prompt to the generation AI 5 along with the posts to instruct the generation of tertiary users to generate profiles based on the posts, a means 30 for sending a prompt to the generation AI 5 along with the profiles to instruct the cluster classification of tertiary users based on the profiles, a means 31 for sending a prompt to the generation AI 5 along with the profiles to instruct the generation of tertiary users to generate personas for each cluster based on the profiles, and a means 32 for creating a research report using the personas for each cluster.

Description

Technical Field

[0001] The present disclosure relates to an information processing apparatus and a program.

Background Art

[0002] Currently in the field of marketing, a so-called research process for clarifying differences between various consumer groups and developing marketing measures suitable for each is emphasized. Through the research process, it is possible to understand a specific person image rather than an abstract "consumer image", which is very useful for formulating more specific measures. Typically, in the research process, many consumers interested in a certain product are classified into clusters based on the similarity of their person images, and a virtual person image (persona) is generated for each cluster.

[0003] However, narrowing down the target subjects that form the basis of clustering requires appropriately combining various elements such as age, gender, occupation, purchase history, website browsing history, hobbies, etc. according to the purpose of the investigation, which is very difficult. When starting the research process with a wide range of consumers without much narrowing down, not only does the noise and fluctuation of the survey results increase, reducing reliability, but it also requires a huge amount of cost and labor until the final result is obtained.

[0004] Regarding clustering as well, for example, viewpoints such as age, purchase frequency, preference for product categories, income, etc. are diverse, and more advanced specialized knowledge and experience are required, so the results tend to be unstable. Furthermore, the work of generating personas for each cluster from the questionnaire results also becomes unstable due to differences in the skills of the workers.

[0005] Although it is known that generating personas for each cluster is useful in the marketing field, it is difficult to avoid the labor, cost, and above all, the reliability and instability of the survey results, so in fact it is not being used, which is the reality.

Summary of the Invention

[0006] The objective is to increase the stability of the research process while reducing costs and effort. [Means for solving the problem]

[0007] The information processing device according to this embodiment comprises a storage unit for storing a program, a processor for executing the program, and a communication unit for communicating with an SNS / blog server device and a generation AI server device that provide an SNS service or blog service via an internet line. The processor executes a program, A request means for requesting the SNS / blog server device to send the account type and the number of posts for each of several primary users that are highly relevant to a specific research purpose, A first extraction means for extracting multiple secondary users from multiple primary users based on the account type of each of the multiple primary users, A second extraction means for extracting multiple tertiary users from multiple secondary users based on the number of posts by multiple secondary users, A second request means that requests the SNS / blog server device to send the posted data that each of the multiple third-party users has posted, A first prompt generation means generates a first prompt instructing the generation of a profile for each of several third-party users based on the posting data posted by each of the several third-party users, and transmits this first prompt along with the posting data posted by each of the several third-party users to the generation AI server device. A second prompt generation means extracts characteristic words from the profiles of multiple tertiary users generated by the generation AI server device, generates a second prompt instructing the system to classify the multiple tertiary users into multiple clusters (user groups) that share common characteristic words, and sends this second prompt to the generation AI server device along with the profiles. Multiple tertiary users belonging to each of the multiple clusters classified by the generating AI server device TheA third prompt generation means generates a third prompt that instructs the system to extract common information (qualitative information) or the most abundant qualitative information from the profiles related to the user, and to generate a persona (a fictional character) that is created by writing a concrete and realistic character based on the qualitative information, and transmits this prompt to the generation AI server device. It functions as a means of creating a research report by inserting the research objectives and the personas of each of the multiple clusters generated by the AI ​​server into a predetermined template. [Brief explanation of the drawing]

[0008] [Figure 1] Figure 1 is a diagram showing the overall system configuration including the information processing device according to this embodiment. [Figure 2] Figure 2 shows the physical configuration of the information processing device shown in Figure 1. [Figure 3] Figure 3 shows the functional configuration of the processor in Figure 2. [Figure 4] Figure 4 is a flowchart showing the processing procedure performed by the processor in Figure 2. [Figure 5] Figure 5 shows the data changes to supplement the processing procedure in Figure 4. [Figure 6] Figure 6 shows an example of the primary user list in step S13 of Figure 4. [Figure 7] Figure 7 shows an example of a secondary user list in step S14 of Figure 4. [Figure 8] Figure 8 shows an example of a tertiary user list in step S15 of Figure 4. [Figure 9] Figure 9 shows an example of a tertiary user profile in step S18 of Figure 4. [Figure 10] Figure 10 shows an example of a cluster of third-party users in process S19 of Figure 4. [Figure 11] Figure 11 shows an example of a persona for each cluster in process S20 of Figure 4. [Modes for carrying out the invention]

[0009] Embodiments of the present invention will be described below with reference to the drawings. Figure 1 shows the configuration of the overall system including the information processing device according to this embodiment. The information processing device 1 according to this embodiment receives a research request and executes a research process according to its purpose, etc., and is connected via an internet line 2 to a research request terminal 3, an SNS server device 4-1 that provides SNS services, a blog server device 4-2 that provides blog services, and a generative AI server device 5 that operates and provides generative artificial intelligence (generative AI).

[0010] As shown in Figure 2, the information processing device 1 has a processor 11. RAM 12, ROM 13, storage unit 14, input device 15, display 16, and communication unit 17 are connected to the processor 11 via a system bus 10. The processor 11 is composed of, for example, a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit). The processor 11 executes programs loaded into RAM 12 from the storage unit 14 and ROM 13 to perform research process processing. RAM 12 functions as the main memory, work area, etc., of the processor 11. ROM 13 or storage unit 14 stores the BIOS (Basic Input Output System) executed by the processor 11, the operating system program (OS), the "research process" processing program according to this embodiment, programs for realizing various other functions, and various data required for these processes. The input device 15 consists of a keyboard (KB), a mouse, a touch panel, or other pointing device. The display 16 is typically implemented as an LCD (Liquid Crystal Display). In addition to the "research process" processing program, the memory unit 14 stores various data necessary for the said processing.

[0011] FIG. 3 shows the functional configuration of the processor 11. By executing the "Research Process" processing program, the processor 11 functions as a control unit 21, a storage management unit 22 that manages writing to and reading from the storage unit 14, an investigation request receiving unit 23, a search word determination unit 24, a user information collection unit 25, a user extraction unit 26, a post collection unit 27, a post extraction unit 28, a profiling generation unit 29, a clustering processing unit 30, a persona generation unit 31, a report editing unit 32, and a report sending unit 33.

[0012] The investigation request receiving unit 23 receives an investigation request from the investigation request terminal 3 together with the investigation purpose. The search word determination unit 24 applies natural language processing to the investigation purpose and determines the most important word as the search word.

[0013] The user information collection unit 25 requests the SNS server device 4-1 and the blog server device 4-2 to transmit the account, account type, and number of posts of users (primary users) who include the search word in their posts, and receives user information including the user account, account type, and number of posts from the SNS server device 4-1 and the blog server device 4-2.

[0014] More specifically, typical primary users with a high degree of relevance to the investigation purpose are users who have posted posts containing words included in the text of the investigation purpose. For example, when the investigation purpose is "Classify users with a high interest in a specific brand (such as ABC) into clusters and generate a persona for each cluster.", it is a user who has posted a post containing the name of a specific brand (such as ABC) on SNS or the like. The user information collection unit 25 requests the SNS server device 4-1 and the blog server device 4-2 to transmit the account, account type, and number of posts of users (primary users) who have posted posts containing the name of a specific brand (such as ABC) on SNS or the like. In response to this request, the SNS server device 4-1 and the blog server device 4-2 transmit the account, account type, and number of posts of the primary users to the information processing device 1.

[0015] As a primary user, in addition to being a user who posts a post containing the words included in the text for the purpose of the survey on SNS or the like, or instead of being a user who posts a post containing the words included in the text for the purpose of the survey on SNS or the like, it may also be a follower who follows a user who meets the survey purpose. For example, it is a follower (user) who follows the official account of a specific brand (such as ABC).

[0016] Users who post posts containing the important words included in these survey purposes on SNS or the like, or followers who follow users who meet the survey purpose, can be identified without human intervention as users highly relevant to the survey purpose, excluding the investigator's arbitrariness and excluding variable factors such as skills and experience, and can stably extract the survey subjects. Of course, it is also possible to reduce costs and labor.

[0017] The user extraction unit 26 extracts secondary users whose account type is individual from the primary users. The secondary users extracted from the primary users are users whose account type is individual. As the account type, in addition to individual accounts, business accounts (corporate accounts), temporary guest accounts, shared accounts shared by multiple people, and specific-purpose accounts such as developer accounts can be excluded, so it is limited to individual use and excludes business and commercial use, which can contribute to improving the survey accuracy and reducing the man-hours of the survey process.

[0018] The user extraction unit 26 extracts tertiary users whose number of posts is above a predetermined number from the secondary users. The tertiary users are users whose number of posts is above a predetermined number, for example, 150 or more. Users whose number of posts is above a predetermined number have a high affinity and consistency with the survey purpose and are also heavy users, and can improve the accuracy when generating the profile of each tertiary user using the posts. Furthermore, by extracting the latest predetermined number of posts from the posts and generating a profile, the accuracy can be further improved.

[0019] The post collection unit 27 sends a request to the SNS server device 4-1 and the blog server device 4-2 to send data regarding the posts and profiles of each third-party user, along with the user accounts of the third-party users, and receives data regarding the posts of the third-party users from the SNS server device 4-1 and the blog server device 4-2. The post extraction unit 28 extracts a predetermined number of the most recent posts from the collected posts of the third-party users. The profiling generation unit 29 sends a prompt to the generation AI server device 5, along with the extracted post data, instructing it to generate a profile for each third-party user based on the extracted posts, and receives the profile data for each third-party user from the generation AI server device 5.

[0020] The clustering processing unit 30 sends a prompt to the generating AI server device 5, along with the account and profile data of the third-party users, instructing it to classify the third-party users into multiple clusters (user groups) based on the profiles generated by the generating AI server device 5. The generating AI server device 5 then receives, as clustering results, data relating to the characteristics of each of the multiple clusters and data relating to a list of accounts of third-party users belonging to each cluster. Note that this clustering process may also be performed by the processor 11.

[0021] Performing clustering based on the profiles of third-party users improves accuracy compared to simply using basic attributes such as age, purchase frequency, product category preferences, and income. Furthermore, because it does not involve human intervention, the results are not unstable.

[0022] The persona generation unit 31 sends a prompt to the generation AI server device 5, along with the profile data of the third-party users, instructing it to generate personas (virtual character profiles) for each cluster based on the profiles of the third-party users, and receives persona data for each cluster from the generation AI server device 5. This persona generation process may also be performed by the processor 11.

[0023] By generating personas (virtual profiles) for each cluster based on profiles of tertiary users, the resulting profiles are clearer than those created simply based on attributes such as age, purchase frequency, product category preferences, and income. This allows for greater reliability in product planning and other related processes. Of course, since no human intervention is involved, the results are not unstable.

[0024] The report editing unit 32 edits and creates the research report by inserting the research objectives, the characteristics of each of the multiple clusters, and the personas of each cluster into a predetermined template. The report transmission unit 33 transmits the research report data to the research request terminal 3.

[0025] Figure 4 shows the processing procedure by the processor 11, and Figure 5 shows the data changes to supplement the processing procedure in Figure 4. The research request receiving unit 23 receives a research request from the research request terminal 3 along with the research objective (S11). For example, the research objective may be the text "Classify users who are highly interested in a specific brand (ABC, etc.) into clusters and generate personas for each cluster." The text of the research objective is subjected to natural language processing by the search word determination unit 24, analyzed into morphemes, the part of speech of each word is determined, nouns are extracted, the importance of each is determined, and the noun with the highest importance is determined as the search word (S12).

[0026] The user information collection unit 25 requests the SNS server device 4-1 and the blog server device 4-2 to transmit the account, account type, and number of posts related to the user (primary user) who posted a post containing the search term on SNS etc., and receives the account, account type, and number of posts related to the primary user from the SNS server device 4-1 and the blog server device 4-2 as illustrated in Figure 6 (S13).

[0027] Furthermore, when the search term is, for example, a company name or brand (ABC, etc.), the user information collection unit 25 may request the SNS server device 4-1 and the blog server device 4-2 to transmit the account, account type, and number of posts related to the followers (primary users) who follow the official account, and the SNS server device 4-1 and the blog server device 4-2 may receive the account, account type, and number of posts related to the primary user.

[0028] Next, the user extraction unit 26 extracts secondary users from the primary users whose account type is individual, as illustrated in Figure 7 (S14). Business accounts (corporate accounts), temporary guest accounts, shared accounts used by multiple people, and specific purpose accounts such as developer accounts are excluded.

[0029] Furthermore, the user extraction unit 26 extracts from secondary users users who have, for example, 150 or more posts, as tertiary users as exemplified in Figure 8, who are assumed to have a high affinity and consistency with the research objectives (S15).

[0030] The post collection unit 27 collects the posts and profiles (objective attributes such as gender and age registered by the user) posted by each tertiary user from the SNS server device 4-1 and the blog server device 4-2 (S16). Subsequently, the post extraction unit 28 extracts, for example, the 10 most recent posts from each tertiary user's posts based on their posting date (S17).

[0031] The profiling generation unit 29 generates a prompt instructing the generation of a profile for each tertiary user based on each tertiary user's most recent 10 posts and profile. This prompt is sent to the generation AI server device 5 along with each tertiary user's most recent 10 posts and profile, and the generation AI server device 5 receives the profile data for each tertiary user (S18). The profile is created, for example, by subjecting the posts to natural language processing, extracting relatively important words, and then formulating a description of the person based on those words and the profile. Note that this profiling generation process may also be performed by the processor 11.

[0032] Next, the clustering processing unit 30 generates a prompt instructing the generation AI server device 5 to classify tertiary users into multiple clusters (user groups) based on the profiles generated by the generation AI server device 5. This prompt is sent to the generation AI server device 5 along with the tertiary user account and profile data. The generation AI server device 5 receives, as clustering results, data relating to the characteristics of each of the multiple clusters illustrated in Figure 10, and data relating to a list of tertiary user accounts belonging to each cluster (S19). Clustering is a method that, for example, extracts characteristic words from the profiles of tertiary users and classifies tertiary users who share common characteristic words into the same cluster (group). Tertiary users belonging to the same cluster have common characteristics, while those characteristics differ between different clusters. Note that this clustering process may also be performed in the processor 11.

[0033] The persona generation unit 31 creates a prompt instructing the generation of personas (virtual character profiles) for each cluster based on the profiles of tertiary users, and sends this prompt along with the tertiary user profile data to the generation AI server device 5. The generation AI server device 5 then receives data of virtual character profiles (personas) for each cluster, as illustrated in Figure 11 (S20). Persona generation is a process that extracts common items (qualitative information) or the most frequent qualitative information from the profiles of tertiary users belonging to the same cluster, and then uses this to create a concrete and realistic character profile in written form. Regarding quantitative information such as age, gender, occupation, income, and region, the most frequent item for each quantitative item such as age, gender, occupation, income, and region is selected from the profiles of tertiary users belonging to the same cluster. Note that this persona generation process may also be executed in the processor 11.

[0034] The report editing department 32 inserts the survey objectives, the characteristics of each of the multiple clusters, and the personas for each cluster into the corresponding sections of the template, creating the survey report (S21), which is then sent to the survey request terminal 3 by the report transmission department 33 (S22). The survey report can be modified by the worker as needed.

[0035] According to this embodiment, the research process can be executed stably without the intervention of human-induced fluctuations, and the costs and labor required for the process can be reduced.

[0036] While several embodiments of the present invention have been described, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These embodiments can be carried out in a variety of other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims and their equivalents. [Explanation of Symbols]

[0037] 1... Information processing device, 2... Internet connection, 3... Survey request terminal, 4-1... SNS server device, 4-2... Blog server device, 5... Generative AI server device.

Claims

1. It comprises a storage unit for storing a program, a processor for executing the program, and a communication unit for communicating with an SNS / blog server device and a generation AI server device that provide SNS services or blog services via an internet connection. The processor executes the program, A request means for requesting the SNS / blog server device to transmit the account type and the number of posts for each of several primary users that are highly relevant to a specific research purpose, A first extraction means for extracting a plurality of secondary users from the plurality of primary users based on the account type of each of the plurality of primary users, A second extraction means for extracting multiple tertiary users from the multiple secondary users based on the number of posts made by the multiple secondary users, A second request means for requesting the SNS / blog server device to transmit the posted data posted by each of the aforementioned multiple third-party users, A first prompt generation means generates a first prompt instructing the generation of a profile for each of the multiple third-party users based on the posted data posted by each of the multiple third-party users, and transmits this first prompt to the generation AI server device along with the posted data posted by each of the multiple third-party users. A second prompt generation means extracts characteristic words from the profiles of each of the multiple tertiary users generated by the generation AI server device, generates a second prompt instructing the system to classify the multiple tertiary users into multiple clusters (user groups) that share the characteristic words, and transmits this second prompt to the generation AI server device along with the profiles. A third prompt generation means generates a third prompt that instructs the generation AI server device to extract common information (qualitative information) or the most abundant qualitative information from the profiles of the multiple tertiary users belonging to each of the multiple clusters classified by the generation AI server device, and to generate a persona (virtual person profile) which is a concrete and realistic person profile written based on the qualitative information, and transmits it to the generation AI server device. An information processing device that functions as a means for creating a research report by inserting the aforementioned research objectives and the personas of each of the multiple clusters generated by the AI ​​generation server device into a predetermined template.

2. The information processing apparatus according to claim 1, wherein the primary user is a user who posted a post containing words included in the specific research purpose.

3. The information processing apparatus according to claim 1, wherein the primary user is a follower who follows users who match the specific research objective.

4. The information processing apparatus according to claim 1, wherein the secondary user is a user whose account type is individual.

5. The information processing apparatus according to claim 1, wherein the aforementioned third-party user is a user whose number of posts is equal to or greater than a predetermined number.

6. The information processing apparatus according to claim 1, wherein the first prompt generation means extracts a predetermined number of the most recent posts from the posts and transmits the first prompt together with the predetermined number of the most recent posts to the generation AI server device.

7. An information processing device comprising a storage unit for storing a program, a processor for executing the program, and a communication unit for communicating with an SNS / blog server device and a generation AI server device that provide SNS services or blog services via an internet connection, A request means for requesting the SNS / blog server device to transmit the account type and the number of posts for each of several primary users that are highly relevant to a specific research purpose, A first extraction means for extracting a plurality of secondary users from the plurality of primary users based on the account type of each of the plurality of primary users, A second extraction means for extracting multiple tertiary users from the multiple secondary users based on the number of posts made by the multiple secondary users, A second request means for requesting the SNS / blog server device to transmit the posted data posted by each of the aforementioned multiple third-party users, A first prompt generation means generates a first prompt instructing the generation of a profile for each of the multiple third-party users based on the posted data posted by each of the multiple third-party users, and transmits this first prompt to the generation AI server device along with the posted data posted by each of the multiple third-party users. A second prompt generation means extracts characteristic words from the profiles of each of the multiple tertiary users generated by the generation AI server device, generates a second prompt instructing the system to classify the multiple tertiary users into multiple clusters (user groups) that share the characteristic words, and transmits this second prompt to the generation AI server device along with the profiles. A third prompt generation means generates a third prompt that instructs the generation AI server device to extract common information (qualitative information) or the most abundant qualitative information from the profiles of the multiple tertiary users belonging to each of the multiple clusters classified by the generation AI server device, and to generate a persona (virtual person profile) which is a concrete and realistic person profile written based on the qualitative information, and transmits it to the generation AI server device. A program that functions as a means of creating a research report by inserting the aforementioned research objectives and the personas of each of the multiple clusters generated by the AI ​​server device into a predetermined template.

Images