Information processing device, information processing method, and program

Abstract

To provide an information processor or the like capable of accurately classifying questionnaires while reducing the labor of a user.SOLUTION: The information processing apparatus 1 includes the reception unit 20 that receives category information including a description of a category input from the operation unit 120, and the acquisition unit 10 that inputs the category information and an answer to a question in a questionnaire to the generation unit 200 and acquires a result of classifying the answer into a category.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to an information processing apparatus, an information processing method, and a program for classifying answers to questions in a questionnaire.

Background Art

[0002] In Patent Document 1, a questionnaire collection system has been proposed that collects questionnaire answer results from communication terminals used by end-users targeted in the questionnaire via the Internet. Although the analysis of questionnaires collected conventionally has been carried out, consultants and the like analyze the questionnaire results for each answer to the question, and finally present them to the client as a report.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] When entrusting a consultant to analyze the answers to questions in the questionnaire results, there is a problem that it is costly. Also, when analyzing within the company without using a consultant, although the cost of outsourcing can be suppressed, work within the company is generated, and ultimately, labor costs are incurred. Also, regarding the accuracy of the analysis, it depends on the ability of the analyst, and there is a possibility that accurate analysis may not be performed depending on the ability of the consultant or the personnel within the company.

[0005] The present invention provides an information processing apparatus and the like that can accurately classify a questionnaire with less effort on the part of the user.

Means for Solving the Problems

[0006] [Concept 1] The information processing apparatus according to the present invention is A reception unit that receives category information, including a description of the category, from the operation unit, The generation unit inputs the aforementioned category information and the answers to the questions in the survey, and the acquisition unit obtains the results of classifying the answers into categories. It may be provided.

[0007] [Concept 2] In an information processing device based on Concept 1, The acquisition unit may use the responses to the questions in the questionnaire, after they have been categorized, to have the generation unit create a summary of the responses related to that category.

[0008] [Concept 3] In an information processing device based on Concept 2, The acquisition unit may use the responses to the questionnaire questions, after they have been categorized, to have the generation unit divide the responses into subcategories and then create a summary of the responses.

[0009] [Concept 4] In an information processing device according to Concept 2 or 3, The system may also include an output unit that outputs, along with the summary, the answers used to create the summary, or a summary answer that summarizes the answers used to create the summary, as a reference answer.

[0010] [Concept 5] In an information processing device based on any one of concepts 1 to 4, The acquisition unit may have the generation unit generate a title using the answers to the questions.

[0011] [Concept 6] In an information processing device based on any one of concepts 1 to 5, The survey includes multiple questions, The information processing device includes a division unit that divides the answer into questions, The acquisition unit may also acquire the results of classifying the answers to each question into categories.

[0012] [Concept 7] In an information processing apparatus according to any one of Concepts 1 to 6, when there are multiple responses to the questionnaire, the result classified into the categories includes the number or ratio classified into each category, the information processing apparatus may include an output unit that outputs the number or ratio classified into each category.

[0013] [Concept 8] The information processing method according to the present invention includes a step of receiving, by a reception unit, category information including an explanation of a category input from an operation unit, and a step of obtaining, by an acquisition unit, a result of classifying responses into categories by inputting the category information and responses to questions in the questionnaire to a generation unit. It may also be provided with.

[0014] [Concept 9] The program according to the present invention is a program installed in an information processing apparatus, and in the information processing apparatus in which the program is installed, a reception function for receiving category information including an explanation of a category input from an operation unit, and an acquisition function for obtaining a result of classifying responses into categories by inputting the category information and responses to questions in the questionnaire to a generation unit may be realized. [Advantages of the Invention]

[0015] In the present invention, when adopting the mode of receiving category information including an explanation of a category input from an operation unit, inputting the category information and responses to questions in the questionnaire to a generation unit, and obtaining a result of classifying responses into categories, it is possible to accurately classify the questionnaire with less effort on the part of the user.

[0016] If the questionnaire can be accurately classified, the analysis is more likely to be appropriately carried out, and ultimately, the analysis can be performed with high accuracy. Therefore, the present invention leads to high-precision analysis of the questionnaire.

Brief Description of the Drawings

[0017] [Figure 1] Schematic block diagram of the information processing system according to the embodiment. [Figure 2] Diagram showing examples of category names and explanations of categories used in the embodiment. [Figure 3] Diagram showing an example of the pie chart output in the embodiment. [Figure 4] Diagram showing an aspect summarized for each category in the embodiment, including sub-classifications. [Figure 5] Diagram showing an example of the information flow in the information processing system according to the embodiment. [Figure 6] Diagram showing an example of the summary output in the embodiment, including reference answers. [Figure 7] Diagram showing an example of the flow of the information processing method according to the embodiment until the category classification result is output. [Figure 8] Diagram showing an example of the flow of the information processing method according to the embodiment until the summary report is output after FIG. 7.

Embodiments for Carrying Out the Invention

[0018] Embodiment 《Configuration》 [[ID= forty-one ]] The information processing apparatus according to the present embodiment may be composed of one apparatus or a plurality of apparatuses. In the present embodiment, an information processing method using the information processing apparatus, a program installed to generate the information processing apparatus, and a storage medium such as a USB or a DVD storing the program are also provided. The "or" in the present application is a concept including "and", and "A or B" is a concept including A, B, and A and B.

[0019] The information processing device of this embodiment may be installed in any location, may be a server, or may be used in a cloud environment. The information processing device of this embodiment may consist of one device or multiple devices. Furthermore, if the information processing device consists of multiple devices, each device does not need to be located in the same room or other space, but may be located in different rooms, different buildings, different regions, etc. Also, if the information processing device consists of multiple devices, one organization may own and / or manage some of them, and another organization may own and / or manage the rest.

[0020] The information processing device of this embodiment is generated, for example, by installing a program. This program may be delivered by email, obtained by accessing a predetermined URL and logging in, or recorded on a recording medium. The program according to this embodiment is used to generate the information processing device described below, and the recording medium according to this embodiment is used to record the program. Furthermore, the information processing method of this embodiment is performed by the information processing device on which the above program is installed.

[0021] As shown in Figure 1, the information processing device 1 may be able to communicate with multiple user terminals 100. User terminals 100 are devices such as personal computers, smartphones, tablet devices, smartwatches, smart glasses, and other wearable computers, and are typically mobile devices such as smartphones and tablet devices. User terminals 100 may have an operation unit 120 for user input and a display unit 110 for displaying information. In smartphones and tablet devices, the operation display unit will have the functions of both the operation unit 120 and the display unit 110.

[0022] Broadly speaking, there are two types of questionnaires: multiple-choice and open-ended. However, analyzing open-ended questionnaires requires a significant amount of time and effort. The information processing device 1 of this embodiment is primarily used to analyze open-ended questionnaires, particularly to analyze what kinds of opinions are expressed. However, it is not limited to this, and can be used in any manner that categorizes and summarizes the free-form responses (open-ended comments) written by users into categories specified by the user. For example, this embodiment may be used to organize complaints and requests submitted voluntarily by users. In this embodiment, the free-ended comments in a questionnaire will be used as an example for explanation below. If this embodiment is used for purposes other than questionnaires, the effects described later can be obtained in a manner unrelated to questionnaires.

[0023] As shown in Figure 1, the information processing device 1 of this embodiment includes a receiving unit 20 that receives one or more category information input by the user from an operation unit 120, an acquisition unit 10 (see Figure 5) that inputs the category information and one or more answers to questions in a questionnaire to a generation unit 200 and obtains the result of classifying the one or more answers to questions into categories, and a storage unit 80 that stores various information. The storage unit 80 may store category information input by the user, output results from the generation unit 200 acquired by the acquisition unit 10, prompts (commands) to be input to the generation unit 200, etc. Input from the user may be made from a user terminal 100. The generation unit 200 may be a generation AI such as ChatGPT provided separately from the information processing device, or it may be included as a component of the information processing device.

[0024] When summarizing responses to a survey, it is effective to categorize the responses before summarizing them. However, accurate categorization requires expertise, and even for experts, the process is extremely time-consuming. In contrast, by adopting the configuration of this embodiment, responses to the survey can be automatically categorized according to user-specified categories. This is highly beneficial because it allows for categorization that aligns with the user's intentions. If the generation unit 200 were to perform the categorization, there is a high probability that the responses would be categorized in a way that does not align with the user's intentions. However, in this embodiment, the user pre-specifies the categories, and the generation unit 200 then performs the categorization, making it highly beneficial because it allows for categorization that aligns with the user's intentions.

[0025] Furthermore, in this embodiment, since classification is performed by the generation unit 200, even with low expertise in classifying questionnaire responses, it is possible to classify questionnaire responses with high accuracy, and standardization of classification criteria can also be achieved. In addition, since user-initiated category classification can be omitted, significant labor savings can be achieved.

[0026] When multiple responses are input to the generation unit 200, the results of categorization may include the number or percentage (number, percentage, etc.) of responses classified into each category. The output unit 90 may then output the number or percentage of responses classified into each category, which can then be displayed on the display unit 110. This configuration is beneficial because it allows users to see at a glance what number or percentage of the survey responses belong to which category. The number or percentage of responses classified into each category may be generated for each question. The output method may also involve using graphs such as pie charts (see Figure 3) or bar graphs. When the number or percentage (number, percentage, etc.) classified into each category is output and displayed on the display unit 110, it is beneficial because it allows users to instantly grasp the numerical content of the survey results.

[0027] As shown in Figure 2, category information may include a category description (shown as "Details" in Figure 2). Category information may also include a category name (shown as "Category assigned to the answer" in Figure 2), etc. Specifically, category information may include a category name and a detailed description for each category. In this case, a prompt (command) such as "You are entering your answer to question (questionnaire) XX. Please classify your answers to the following questions (questionnaire) using the category names XX, △△, and △▽. The category name XX is used to mean ●●●●. The category name △△ is used to mean ▲▲▲▲. The category name △▽ is used to mean ▲▼▲▼. (Answer 1), (Answer 2), ..., (Answer Z)" may be used to allow the acquisition unit 10 to acquire the results of classifying the answers to the questions into categories from the generation unit 200. In this case, the system may also input the answers actually categorized by a person with expertise and specific examples (past cases) related to the classification results, and then have the acquisition unit 10 acquire the categorized results from the generation unit 200. By adopting this configuration, it is expected that the accuracy of category classification can be further improved.

[0028] Alternatively, each respondent may be required to input their answers to the questionnaire. In this case, a prompt such as "You are entering your answer to question (questionnaire) XX. Please classify your answers to the following questions (questionnaires) using the category names XX, YY, and YYY. The category name XX means ●●●●. The category name △△ means ▲▲▲▲. The category name △YYY means ▲▼▲▼. (Answer)" (where "Answer" refers to the answer to a questionnaire question by a specific respondent) may be provided for each question, and the input for each question or multiple questions at once may be entered into the generation unit 200, and the answer results for each respondent may be divided into categories specified by the user.

[0029] The acquisition unit 10 may use the responses to the questions in the questionnaire, after they have been categorized, to have the generation unit 200 create a summary of the responses related to that category (see Figure 5). In this case, the acquisition unit 10 may use a prompt such as, "The following are the responses to the questions (questionnaire) categorized under the category name XX. The category name XX is used to mean ●●●●. Please summarize the content of the following responses considering the meaning of the category. (Response 1), (Response 2), ..., (Response M)" to obtain the results of summarizing the responses to multiple questions from the generation unit 200. In this case as well, the acquisition unit 10 may also input specific examples (past cases) of questionnaire responses and results actually summarized by experts, and then obtain the summary results from the generation unit 200. By adopting such a configuration, it is expected that the accuracy of category summarization can be further improved. Previously, analyzing 1,000 open-ended responses (categorization and summarization) took approximately 4-5 hours. However, by adopting this method, the work time was reduced to about 10 minutes, and the analysis results were more accurate, resulting in a dramatic improvement.

[0030] The acquisition unit 10 may have the generation unit 200 generate a title separate from the category name using the answers to the questions. In this case, the result of summarizing the answers to the questions may be used. For example, the acquisition unit 10 may use a prompt such as, "The answers to the questions (surveys) classified under the category name XX have been entered as follows. The category name XX is used to mean ●●●●. The following answers have been obtained, and as a result of summarizing the following answers, □□□□□□□ has been obtained. Please generate an appropriate title. (Answer 1), (Answer 2), ..., (Answer M)" to have the generation unit 200 generate a title for the answers classified under a predetermined category. By having the generation unit 200 generate the title in this way, the title can be made to appropriately reflect the content of the category, which is beneficial because it makes it easier for the user to understand the content of the category.

[0031] The questionnaire may include multiple questions. A division unit 30 may be provided to divide the answers into individual questions. The acquisition unit 10 may have the generation unit 200 categorize the answers to each question. The generation unit 200 may also be used to create summaries or titles using the obtained categorization. By employing such a division unit 30, even questionnaires containing multiple questions can be automatically divided into individual questions, and it is beneficial in that categorization and summaries for each question can be obtained.

[0032] As shown in Figure 4, summaries may be created for each category, and summaries may be generated for each category. In this case, after the data has been classified into categories, the acquisition unit 10 may use a prompt such as, "The following are the answers to the questions (surveys) classified under the category name XX. The category name XX is used to mean ●●●●. The following answers have been obtained, but please further subdivide these answers into smaller categories. When doing so, please create a title for each subcategory. (Answer 1), (Answer 2), ..., (Answer M)" to further subdivide the opinions within each category, and the generation unit 200 may generate a title for each subcategory. Dividing the data into smaller categories in this way allows for a more appropriate analysis of the answers to the survey questions. Furthermore, displaying these subcategories on the display unit 110 is beneficial because it makes it easier for the user to categorize the answers. Additionally, generating a title for each subcategory by the generation unit 200 is beneficial because it makes it easier for the user to understand the content of the subdivided answers.

[0033] In this case, you may have them create a summary for each category. For example, "The following responses have been entered for the questions (surveys) categorized under the category name XX. The category name XX is used to mean ●●●●. The responses (Answer 1), (Answer 2), ..., and (Answer M) are divided as follows. Please create a summary that combines these: "[Emphasis on citizen participation] ...[Consideration for children and the elderly] ..." The acquisition unit 10 may use prompts such as "[Long-term budget allocation] ···············[Revitalizing local communities and reflecting their needs] ···············" to cause the generation unit 200 to generate a summary for each category (shown as "Summary" in Figure 4).

[0034] In the configuration shown in Figure 6, the titles are "Budget Reduction and Efficiency Improvement" and "Utilization of Private Funds." Below these, summaries of "Budget Reduction and Efficiency Improvement" and "Utilization of Private Funds" are shown. The answers used to create the summaries are listed as reference answers. Outputting these reference answers from the output unit 90 is beneficial to the user as it makes it easier to analyze the responses to the questionnaire. The reference answers listed may be the actual answers entered, or they may be summaries of the actual answers entered (summary answers). In this case, for example, a prompt such as, "The following are the answers to the questions (surveys) categorized under the category name XX. The category name XX is used to mean ●●●●. The following answers have been obtained. Please classify these answers into A items or less and list the summary results in short sentences. (Answer 1), (Answer 2), ..., (Answer M)" may be used so that the acquisition unit 10 classifies the answers from the generation unit 200 into A items or less and obtains the summary results.

[0035] The output unit 90 may output the summarized answers to the questions, categorized by the answers to each question. More specifically, if there are multiple questions, a summary of the answers may be created for each question. Typically, for each question, the output unit 90 will output answers divided into multiple categories.

[0036] The report may include an overall summary (summary report) for each question. The acquisition unit 10 may also have the generation unit 200 generate the overall summaries for each question. In this case, the acquisition unit 10 may use a prompt such as, "The answers to the questions (surveys) categorized as XX, YY, and □□ are entered as follows. The category name XX is used to mean ●●●●. The category name △△ is used to mean ▲▲▲▲. The category name □□ is used to mean ■■■■. The result of summarizing the category XX is ●●●●●●●●. The result of summarizing the category △△ is ▲▲▲▲▲▲▲▲. The result of summarizing the category □□ is ■■■■■■■■. Please create a document summarizing all of these." to cause the acquisition unit 10 to create an overall summary of the answers to the questions from the generation unit 200.

[0037] The acquisition unit 10 may cause the generation unit 200 to generate a report summarizing the entire questionnaire, which includes multiple questions. In other words, the acquisition unit 10 may cause the generation unit 200 to generate an overall report by summarizing the reports obtained for each question. In this case, the acquisition unit 10 prepares a sentence for each answer that reads, "The answers to the questions (surveys) categorized as XX, YY, and □□ are entered as follows. The category name XX is used to mean ●●●●. The category name △△ is used to mean ▲▲▲▲. The category name □□ is used to mean ■■■■. The result of summarizing the category XX is ●●●●●●●●. The result of summarizing the category △△ is ▲▲▲▲▲▲▲▲. The result of summarizing the category □□ is ■■■■■■■■. The result of summarizing all of these is AAAAAAAAAAAAAA." (Hereafter referred to as "Question Prompt 1", "Question Prompt 2", ..., "Question Prompt P", etc.), and may use a prompt such as "(Question Prompt 1)(Question Prompt 2)...(Question Prompt P) are as follows, please combine these and create an overall report." to have the acquisition unit 10 create an overall summary of the answers to the questions from the generation unit 200.

[0038] Category information entered by the user may be stored in the storage unit 80 in association with user identification information such as the user's user ID. The format output by the output unit 90 may be Word, PowerPoint, etc., and is not particularly limited. The information stored in the storage unit 80 in association with user identification information may be made available for the user to read as needed.

[0039] The following describes an example of the processing method according to this embodiment.

[0040] The user inputs category information, including category descriptions and category names, from the user terminal 100 (see Figure 7). The user also inputs multiple responses to a survey for which a report is to be created. Category descriptions and category names may be entered in pre-prepared fields. Multiple responses to the survey may be entered in a table format, with each response corresponding to a question. In the configuration where the division section 30 is provided, when multiple survey results are input, the responses are automatically separated by question. For example, (Question 1) would be divided into (Response 1-1), (Response 1-2), ..., (Response 1-N); (Question 2) would be divided into (Response 2-1), (Response 2-2), ..., (Response 2-N); ...; (Question R) would be divided into (Response R-1), (Response R-2), ..., (Response RN).

[0041] When the execution button for report creation displayed on the display unit 110 of the user terminal 100 is pressed, the acquisition unit 10 reads a category classification prompt from the storage unit 80, and combines the category information, including the category description and category name entered in the prompt, with the multiple answers to cause the generation unit 200 (shown as "Generation AI" in Figure 7) to generate results that separate the multiple answers into categories. When the generation unit 200 outputs the results that separate the multiple answers into categories, the results are stored in the storage unit 80, linked to user information and creation date and time. Note that in Figure 7, the "Classification Prompt Creation Device" constitutes part of the acquisition unit 10.

[0042] Next, the acquisition unit 10 reads a summarization prompt from the storage unit 80 and, for each category, incorporates category information, including a description of the category and the category name, along with the answer to the question, into the prompt, causing the generation unit 200 to generate a summary for each category (see Figure 8). In this case, the generation unit 200 may generate a summary of the actual input answer (see Figure 6). Alternatively, the data may be divided into subcategories, and the generation unit 200 may generate a summary for each category (see Figure 4). Once the generation unit 200 outputs the summary result (including the summary of the answer) for each category, the result is stored in the storage unit 80, linked to user information and creation date and time. If there are multiple categories, this process is performed for all categories, and a summary is generated for each category. Note that in Figure 8, the "summarization prompt creation device" and the "summary report creation device" constitute part of the acquisition unit 10.

[0043] Next, the acquisition unit 10 may read a prompt for title generation from the storage unit 80 and, for each category, incorporate category information including the category description and category name, along with the summary result generated by the generation unit 200, into the prompt, causing the generation unit 200 to generate a title for each category. Once the generation unit 200 outputs a title for each category, the result is stored in the storage unit 80, linked to user information and creation date and time. If there are multiple categories, this process is performed for all categories, and a title is generated for each category.

[0044] The acquisition unit 10 may read a prompt from the storage unit 80 for creating a summary report for a single question (for generating a summary for each question), and incorporate category information including a description of the category and the category name, the summary result and title generated by the generation unit 200 into the prompt, causing the generation unit 200 to generate an overall summary for the question. When the generation unit 200 outputs an overall summary for the question, the result is stored in the storage unit 80, linked to user information and creation date and time.

[0045] If the survey includes multiple questions, the above process can be automated to apply to the remaining questions as well.

[0046] Once the above process is completed for all questions, the acquisition unit 10 may read a prompt for summary generation (for generating summaries of all questions) from the storage unit 80, and incorporate category information including category descriptions and category names, the summary results generated by the generation unit 200, the title, and the summary for each question into the prompt, causing the generation unit 200 to generate a summary of the questionnaire. When the generation unit 200 outputs the summary, the result is stored in the storage unit 80, linked to user information and creation date and time.

[0047] According to this embodiment, a report including an overall summary for each question, a title and summary result for each category (including a summary of the answers), and an overall summary is output from the output unit 90 and displayed on the display unit 110 of the user terminal 100, and is stored in the storage unit 80 in association with user information and creation date and time. In this case, the results of the category classification (number, percentage, etc. of the population) may also be displayed on the user terminal 100.

[0048] The acquisition unit 10, reception unit 20, splitting unit 30, output unit 90, etc. shown in Figure 1 may be implemented by a single unit (control unit) or by different units. Functions from multiple "units" may be integrated. Furthermore, the acquisition unit 10, reception unit 20, splitting unit 30, output unit 90, etc. may be implemented by a circuit configuration. The information processing device 1 has a processor, and the processor executes a program to realize various functions of the information processing device 1's acquisition unit 10, reception unit 20, splitting unit 30, output unit 90, etc. in this embodiment.

[0049] The above-described embodiments and the disclosure of drawings are merely examples for illustrating the invention described in the claims, and the above-described embodiments or the disclosure of drawings do not limit the invention described in the claims. [Explanation of symbols]

[0050] 10 Acquisition Department 20 Reception Department 30 Division 80 Storage section 90 Output section 100 user terminals 110 Display section 120 Operation section 200 Generation part

Claims

1. A receiving unit that reliably receives category information, including a category and a description of the category, which is input by the user who is the user of each user terminal from the operating unit of each user terminal, The acquisition unit inputs the category information received by the reception unit and the answers to the questions in the questionnaire into the generation unit and obtains the results of classifying the answers into categories. Equipped with, An information processing device that can classify responses to survey questions into different categories for each user.

2. The information processing apparatus according to claim 1, wherein the acquisition unit uses the answers to the questions in the questionnaire after they have been classified into categories to cause the generation unit to create a summary of the answers related to those categories.

3. The information processing apparatus according to claim 2, wherein the acquisition unit uses the answers to the questions in the questionnaire, after they have been classified into categories, to cause the generation unit to divide the answers into subcategories and then create a summary of the answers.

4. The information processing apparatus according to claim 2 or 3, further comprising an output unit that outputs, along with the summary, the answers used in creating the summary or a summary answer which is a summary of the answers used in creating the summary, as a reference answer.

5. The information processing apparatus according to any one of claims 1 to 3, wherein the acquisition unit causes the generation unit to generate a title using the answers to the questions.

6. The survey includes multiple questions, It includes a section for dividing the answers into questions, The reception desk receives category information, including a description of the category, which is entered by the user through the user's operation panel on the user terminal for each question. The acquisition unit acquires the results of classifying the answers for each question into categories, as described in any one of claims 1 to 3.

7. If there are multiple responses to the questionnaire, the results of classifying them into the categories mentioned above include the number or percentage of responses classified into each category. An information processing apparatus according to any one of claims 1 to 3, comprising an output unit that outputs the number or percentage classified into each category.

8. The reception unit is required to receive category information, including categories and descriptions of those categories, from the operation units of multiple user terminals, through the operations of the users of each user terminal. The acquisition unit inputs the category information received by the reception unit and the answers to the questions in the questionnaire into the generation unit, and obtains the result of classifying the answers into categories. Equipped with, An information processing method that allows responses to survey questions to be categorized into different categories for each user.

9. A program installed on an information processing device, The information processing device on which the aforementioned program is installed, A reception function that reliably accepts category information, including categories and descriptions of those categories, entered by the user of each user terminal through the operation of the user terminal's control panel, The acquisition function inputs the category information received by the reception function and the answers to the questions in the survey into the generation unit, and obtains the results of classifying the answers into categories. To make it happen, A program that allows users to categorize their responses to survey questions into different categories.

Images