Support devices, support methods, support systems, and computer programs

Abstract

This technology provides support for extracting survey results relevant to the target profile. [Solution] The support device comprises an input unit, an acquisition unit, an extraction unit, and an output unit. The input unit is configured to accept input of a target image. The acquisition unit is configured to acquire questionnaire data including multiple questionnaire results. The extraction unit is configured to extract one or more questionnaire results related to the target image as panel data based on the questionnaire data. The output unit is configured to output the extracted panel data.

Description

Technical Field

[0006] , , , ,

[0001] This disclosure relates to a support device.

Background Art

[0002] In the marketing field, a technique for analyzing a target using a questionnaire is known. For example, Patent Document 1 discloses a promotion support system that uses the response results of a questionnaire to narrow down a group having predetermined attributes such as age and gender, and distributes promotion information to the narrowed-down group.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

[0007] With this configuration, users can extract survey results relevant to the target profile without having to be aware of the content of the survey questions or results. Therefore, a support device can assist in extracting survey results relevant to the target profile.

[0008] In one aspect of this disclosure, each of the multiple survey results may include the results of responses by the survey respondents to multiple questions. With this configuration, it is possible to analyze survey respondents whose responses correspond to survey results related to the target profile, based on their answers to multiple questions. Therefore, it can support the analysis of the target profile using survey results.

[0009] In one aspect of this disclosure, the output unit may be configured to further output the extraction conditions for panel data by the extraction unit. The extraction conditions may be expressed by a plurality of association combinations. Each of the plurality of association combinations may be a combination of one association question related to the target image from among a plurality of questions, and an association response result which corresponds to one association question and is an answer result related to the target image.

[0010] With this configuration, the panel data output by the support device can be identified by the extraction conditions, which determine the conditions under which it was extracted relative to the input target image.

[0011] In one aspect of this disclosure, the extraction criteria may be expressed by a logical expression using the logical AND and / or OR of a plurality of related combinations.

[0012] With this configuration, the panel data output by the support device can be more clearly understood, based on the logical formula of the extraction conditions, as it shows the conditions under which the data was extracted relative to the input target image.

[0013] One aspect of the present disclosure may further include a change receiving unit configured to receive change instructions indicating at least one change to the target image and extraction conditions. The extraction unit may be configured to re-extract panel data based on the change instructions. The output unit may be configured to output the re-extracted panel data.

[0014] With this configuration, the target image can be modified or the extraction conditions modified based on the output panel data and at least one of the extraction conditions, so that the expected panel data is extracted.

[0015] One aspect of this disclosure may further include an analysis unit configured to analyze the features of the panel data. An output unit may further output the features of the panel data. This configuration allows us to understand the characteristics of the target image based on panel data.

[0016] In one aspect of this disclosure, the analysis unit may be configured to input panel data and a request signal requesting that the panel data be analyzed to artificial intelligence, and to obtain an analysis result generated by the artificial intelligence as a response to the request signal. The output unit may be configured to output the analysis result as a feature of the panel data. This configuration facilitates the analysis of target image features based on panel data.

[0017] In one aspect of this disclosure, the extraction unit may be configured to extract a panel dataset including first panel data, which is panel data, and second panel data, which is different from the first panel data. The output unit may be configured to output the panel dataset.

[0018] With this configuration, more appropriate panel data can be selected from the extracted panel datasets. Therefore, the support device can assist in extracting questionnaire results related to the target profile.

[0019] One aspect of this disclosure is a support system comprising the above-described support device and a display device. The display device is configured to display information output by an output unit. This configuration makes it possible to realize a system that can display the information output by the support device.

[0020] One aspect of this disclosure is a computer-assisted method comprising: receiving input of a target image; acquiring survey data including multiple survey results; extracting one or more survey results related to the target image as panel data based on the survey data; and outputting the extracted panel data. This support method produces the same effect as the support device described above.

[0021] One aspect of the present disclosure is a computer program for causing a computer to perform an operation, the operation including receiving input of a target image, acquiring survey data including multiple survey results, extracting one or more survey results related to the target image as panel data based on the survey data, and outputting the extracted panel data. According to this computer program, it produces the same effect as the aforementioned support device. [Brief explanation of the drawing]

[0022] [Figure 1] It is a block diagram showing the configuration of the support device. [Figure 2] It is a diagram for explaining questionnaire data and questionnaire results. [Figure 3] It is a flowchart showing the processing of the support device. [Figure 4] It is a diagram for explaining the process of extracting panel data from questionnaire data. [Figure 5] It is a diagram for explaining an example of extraction conditions. [Figure 6] It is a diagram for explaining an example of analysis results. [Figure 7] It is a diagram for explaining an example of inputting a change instruction through an interactive user interface. [Figure 8] It is a diagram for explaining the process of extracting a panel data set from questionnaire data.

Embodiments for Carrying Out the Invention

[0023] Hereinafter, exemplary embodiments of the present disclosure will be described with reference to the drawings. [1. First Embodiment] [[ID=​​​​​​​​​​​​​

[0026] Memory 12 is configured to store computer programs. Memory 12 is used as a workspace when the processor 11 executes processing. An example of memory 12 is RAM (Random Access Memory).

[0027] Storage 13 holds computer programs and various data used when executing processes that follow the computer programs. Examples of storage 13 include HDDs (Hard Disk Drives), SSDs (Solid State Drives), and flash memory.

[0028] The user interface 14 is a general term for interfaces that accept various input operations from the user and interfaces that output various information to the user. Examples of user interfaces 14 include keyboards, mice, touch panels, displays, etc.

[0029] The user interface 14 includes a display device configured to display various information output by the processor 11. Part or all of the user interface 14 (e.g., the display device) may be located outside the support device 1.

[0030] The communication interface 15 is an interface that can communicate various types of data with external devices in accordance with a predetermined standard. The support device 1 is configured to communicate with the artificial intelligence 2 and the database 3, respectively, through the communication interface 15.

[0031] Artificial intelligence 2 is configured to execute processing based on a request signal when a request signal is input, and to output the result of the executed processing as a response to the request signal. For example, artificial intelligence 2 may be built on a large language model (LLM). Artificial intelligence 2 may also be a generative AI built on a large language model. Examples of generative AIs include ChatGPT from OpenAI, Gemini from Google, Copilot from Microsoft, and Claude from Anthropic. The request contained in the request signal may be a prompt expressed using natural language.

[0032] Database 3 holds the survey data 21. As shown in Figure 2, the survey data 21 includes multiple survey results 22. Each of the multiple survey results 22 includes the responses of the survey respondents to multiple questions.

[0033] In the example shown in Figure 2, the results (A1, A2, A3, ...) of a survey respondent for each of several questions (Q1, Q2, Q3, ...) are shown. As an example of multiple questions, Figure 2 shows questions about gender, age, occupation, residential area, family members living together, leisure activities, app usage, etc., but the multiple questions are not limited to these and may be any other arbitrary questions.

[0034] The multiple questions shown in Figure 2 are a multiple-choice questionnaire in which respondents select one or more items from a pre-prepared set of options. However, the multiple questions are not limited to multiple-choice questionnaires and may include any other type of questionnaire. For example, the multiple questions may include an open-ended questionnaire in which respondents can freely write their answers, a questionnaire using a Likert scale in which respondents rate items on a 1-5 scale, or a questionnaire using a ranking system in which respondents rank items from 1st to 3rd place.

[0035] [1-2. Processing] The series of processes performed by processor 11 will be explained using the flowchart in Figure 3. When the processor 11 receives instructions from the user through the user interface 14, it starts the processing shown in Figure 3.

[0036] First, in S100, the processor 11 accepts input of a target profile through the user interface 14. The target profile can be an image of a subject with attributes that the user wants to analyze in marketing. For example, a target profile could be "a health-conscious woman in her 20s" or "a self-employed person in their 40s living alone." The user can input the target profile in any format using natural language. Being able to input in any format using natural language increases the user's input flexibility and improves usability.

[0037] Next, in S110, the processor 11 acquires the survey data 21 from the storage 13. The processor 11 may also accept input of the survey data 21 from the user through the user interface 14, or acquire the survey data 21 from outside the support device 1 through the communication interface 15.

[0038] Next, in S120, the processor 11 extracts one or more questionnaire results 22 related to the target image received in S100 as panel data 23, based on the questionnaire data 21, as shown in Figure 4.

[0039] In this embodiment, the processor 11 receives a request signal from the artificial intelligence 2, instructing it to extract one or more questionnaire results 22 related to the target image from the questionnaire data 21, and acquires the extraction results as a response to the request signal output from the artificial intelligence 2 as panel data 23.

[0040] The request signals that processor 11 inputs to artificial intelligence 2 include requests such as, "From the input survey data, please extract questions and answer results related to the following target profile. Target profile: Women in their 20s with a high level of health awareness."

[0041] In this embodiment, the processor 11 also includes in the request signal an extraction condition request that requests the output of extraction conditions 30 for one or more survey results 22 (i.e., panel data 23), and an analysis request that requests the output of analysis results 40 after analyzing the characteristics of the extracted panel data 23. The analysis request may also include a suggestion request that requests suggestions for effective marketing measures for targets having the analyzed characteristics.

[0042] Extraction criteria requests include requests such as, "Please output the extraction criteria used when extracting survey results." An analysis request might include, for example, "Analyze the characteristics of the extracted survey results and create a concise report." The analysis request may also include an insight request, such as, "Based on the analysis results, provide insights that will be helpful in developing effective media planning for the target audience."

[0043] Artificial intelligence 2 outputs the corresponding extraction conditions 30 in response to the extraction condition request. Artificial intelligence 2 outputs the corresponding analysis result 40 in response to the analysis request. Artificial intelligence 2 outputs the corresponding suggestion included in the analysis result 40 in response to the suggestion request.

[0044] Next, in S130, the processor 11 outputs the panel data 23 extracted in S120 through the user interface 14. In this embodiment, the processor 11 also outputs the extraction conditions 30 shown in Figure 5 and the analysis results 40 shown in Figure 6.

[0045] The extraction condition 30 is expressed by a plurality of association combinations 31. Each of the plurality of association combinations 31 is a combination of one association question 32 that is related to the target profile from among the plurality of questions, and an association response result 33 that is an answer result corresponding to the one association question 32 and is an answer result related to the target profile.

[0046] In the example shown in Figure 5(a), multiple related combinations 31 are represented as related questions 32 and related response results 33 related to the target profile of "women in their 20s with high health awareness," which include being female, being between 20 and 29 years old, spending leisure time on exercise, sports, or walking, and using a healthcare app or beauty app. The extraction condition 30 is represented by a logical formula using the logical AND and OR of these multiple related combinations 31. In this case, "gender," "age," "leisure time," and "app usage" correspond to related questions 32, respectively, while "female," "age 20-29," "exercise, sports, or walking," "healthcare app," and "beauty app" correspond to related response results 33, respectively.

[0047] Alternatively, as an example not illustrated, if the answer format for related question 32 is a multi-level questionnaire, then for related question 32, "Frequency of exercise / sports," options corresponding to "three or more times a week," such as "about three times a week" or "almost every day," may be considered to correspond to related response 33.

[0048] In the example shown in Figure 5(b), multiple related combinations 31 are represented as related questions 32 and related response results 33 related to the target profile of "a self-employed person in their 40s living alone," where the age is 40-49, the occupation is self-employed, and there are no family members living together (i.e., single). The extraction condition 30 is represented by a logical formula using the logical AND of these multiple related combinations 31. In this case, "age," "occupation," and "family members living together" correspond to related questions 32, respectively, and "40-49 years old," "self-employed," and "single" correspond to related response results 33, respectively.

[0049] The extraction criteria 30 are not limited to logical formulas; they can be expressed in other ways. For example, the extraction criteria 30 may be expressed in natural language text, such as, "The extraction criteria set are: gender is female, age is 20-29, leisure activities are exercise / sports / walking, and app usage is either a healthcare app or a beauty app."

[0050] In the example shown in Figure 6, the analysis result 40 shows the results of analyzing the characteristics of the extracted panel data 23, along with suggestions for effective marketing measures for the target profile possessing the analyzed characteristics. Since the characteristics shown in analysis result 40 are those of the panel data 23, they may include content that has little direct relevance to the target profile entered by the user. In the example shown in Figure 6, analysis result 40 shows characteristics related to SNS usage rate, video service usage frequency, TV / radio listening frequency, and residential area, which have little direct relevance to the target profile of "women in their 20s with high health awareness." Furthermore, analysis result 40 includes a suggestion for effective marketing measures for the target profile possessing the analyzed characteristics: "Since this target is highly digitally oriented, media planning centered on SNS and video advertising is considered particularly effective."

[0051] Next, in S140, the processor 11 transitions to a state where it can receive a change instruction 50 that instructs a change to at least one of the target image and extraction conditions 30, and determines whether or not it has received the change instruction 50. The processor 11 is configured to receive the change instruction 50 from the user through the user interface 14.

[0052] In the example shown in Figure 7, the processor 11 displays the analysis results 40 to the user through an interactive user interface 14 and accepts modification instructions 50 from the user. The user can input modification instructions 50 based on the analysis results 40 displayed on the user interface 14.

[0053] For example, if the target profile received in S100 is "women in their 20s with a high level of health awareness," then the instruction 50 may be an instruction to change, expand, or narrow the target profile, such as "Change the target from women in their 20s to women in their 30s," "Include men in their 20s in the target," or "Limit the target to people who frequently use video services."

[0054] Alternatively, the modification instruction 50 may be an instruction to change the logical expression of multiple association combinations 31, such as, if the extraction condition 30 output in S130 is "leisure time is spent exercising, playing sports, or walking, AND a healthcare app is being used," then change it to an OR condition such as "'leisure time is spent exercising, playing sports, or walking, OR a healthcare app is being used.'"

[0055] If processor 11 determines in S140 that it has received the change instruction 50 (S140:YES), it proceeds to S150. In S150, the processor 11 re-extracts one or more survey results 22 as panel data 23 based on the change instruction 50. The processor 11 may re-extract one or more survey results 22 corresponding to the change instruction 50 from the survey data 21 obtained in S110, or from the panel data 23 extracted in S120.

[0056] In this embodiment, the processor 11 inputs a request signal to the artificial intelligence 2, requesting it to re-extract one or more questionnaire results 22 from the questionnaire data 21 based on the change instruction 50, and re-acquires the extraction results output from the artificial intelligence 2 as panel data 23 in response to the request signal. Similar to S120, the request signal includes an extraction condition request and an analysis request. The processor 11 also re-acquires the extraction conditions 30 and analysis results 40 from the artificial intelligence 2 in addition to the panel data 23.

[0057] Next, in S160, the processor 11 outputs the panel data 23 re-extracted in S150 through the user interface 14. In this embodiment, the processor 11 also outputs the extraction conditions 30 and analysis results 40 based on the change instruction 50. In the example shown in Figure 7, the corresponding analysis results 40 are displayed on the user interface 14 for the change instruction 50 received from the user. Subsequently, processor 11 will be replaced by S140.

[0058] On the other hand, if the processor 11 determines in S140 that it did not accept the change instruction 50 (S140: NO), it terminates the process shown in Figure 3. The processor 11 may also determine that it did not accept the change instruction 50 if it receives an instruction through the user interface 14 to terminate the process shown in Figure 3.

[0059] In this manner, the processor 11 of the support device 1 receives input of a target image, acquires questionnaire data 21, and extracts one or more questionnaire results 22 related to the target image as panel data 23 based on the acquired questionnaire data 21. The processor 11 outputs the extracted panel data 23 through the user interface 14. If the processor 11 receives a change instruction 50, it re-extracts and outputs the panel data 23 based on the change instruction 50.

[0060] [1-3. Effects] According to the embodiments described above, the following actions and effects can be obtained. (1a) The processor 11 extracts one or more survey results 22 related to the target profile as panel data 23 based on the survey data 21 which includes multiple survey results 22.

[0061] With this configuration, users can appropriately extract survey results 22 related to the target profile from the survey data 21. For example, if a user sets the target profile as "people with high health awareness," there may be no surveys in the survey data 21 that include a question directly asking whether or not the user fits the target profile (in this example, a question asking about their level of health awareness). Even in such cases, the user can easily extract survey results 22 related to the target profile using the support device 1, without having to consider what kind of survey results 22 to extract, or even being aware of the content of the survey questions or survey results 22. Therefore, the support device 1 can assist in the extraction of survey results related to the target profile.

[0062] (1b) Each of the multiple survey results 22 includes the results of the survey respondents' answers to multiple questions. With this configuration, it is possible to analyze survey respondents corresponding to survey results 22 related to the target profile based on their answers to multiple questions. Therefore, it is possible to support the analysis of the target profile using survey results.

[0063] (1c) In S130, the processor 11 outputs the panel data 23 in addition to the extraction conditions 30 which are represented by a plurality of association combinations 31. Each of the plurality of association combinations 31 is a combination of one association question 32 which is related to the target image from among the plurality of questions, and an association response result 33 which is an answer result corresponding to the one association question 32 and is an answer result related to the target image.

[0064] With this configuration, the user can find out, through the extraction conditions 30, the conditions under which the panel data 23 output by the support device 1 was extracted from the input target image (i.e., the basis for the extraction).

[0065] (1d) The extraction condition 30 may be expressed by a logical expression using the logical AND and / or logical OR of multiple association combinations 31. With this configuration, the user can more clearly understand, through the logical expression of the extraction condition 30, the conditions under which the panel data 23 output by the support device 1 was extracted relative to the input target image (i.e., the basis for the extraction).

[0066] (1e) When the processor 11 receives a change instruction 50 that instructs to change at least one of the target image and extraction conditions 30, it re-extracts one or more questionnaire results 22 as panel data 23 based on the change instruction 50 and outputs them.

[0067] With this configuration, the user can extract new panel data 23 by inputting a change instruction 50 based on the outputted panel data 23 and extraction conditions 30 to the support device 1. For example, the user can adjust the system to extract the expected panel data 23 by modifying the target image or modifying the extraction conditions 30 based on the outputted panel data 23 and extraction conditions 30.

[0068] The processor 11 may re-extract one or more survey results 22 corresponding to the change instruction 50 from the survey data 21 acquired in S110, or from the panel data 23 extracted in S120. This configuration makes it possible to improve the accuracy of extracting panel data 23.

[0069] (1f) The processor 11 analyzes the characteristics of the extracted panel data 23 and outputs the analysis results 40.

[0070] With this configuration, users can learn about the characteristics of their target profile based on the analysis results 40 of the panel data 23, without having to analyze the contents of the panel data 23 in detail themselves. For example, when a user sets their target profile to "people with high health awareness," one or more survey results 22 related to "people with high health awareness" are extracted as panel data 23. In this case, the analysis results 40 show the results of an analysis of the characteristics of multiple questions and answers included in the one or more survey results 22 that are extracted as panel data 23, not limited to questions and answers related to health awareness. In other words, the analysis results 40 can be said to show the characteristics of the target profile, "people with high health awareness," based on one or more survey results 22. Therefore, users can learn about the characteristics of their target profile based on the analysis results 40.

[0071] (1g) The processor 11 includes an analysis request in the request signal input to the artificial intelligence 2, which requests that the artificial intelligence 2 analyze the features of the extracted panel data 23 and output further analysis results 40. The processor 11 obtains the analysis results 40 generated by the artificial intelligence 2 in response to the analysis request and outputs the analysis results 40.

[0072] With this configuration, the features of the panel data 23 can be analyzed using artificial intelligence 2. Therefore, the analysis of the features of the target image based on the panel data is made easy.

[0073] (1h) The processor 11 includes a suggestion request in the request signal input to the artificial intelligence 2, which requests suggestions for effective marketing measures for the target profile having the analyzed characteristics. The processor 11 obtains the suggestions generated by the artificial intelligence 2 in response to the suggestion request and outputs the obtained suggestions in the analysis result 40.

[0074] This configuration allows us to gain insights into effective marketing strategies for the target audience.

[0075] [1-4. Correspondence between terms] In the above embodiment, the process in S100 corresponds to an example of a process executed as an input unit, the process in S110 corresponds to an example of a process executed as an acquisition unit, the processes in S120 and S150 correspond to an example of a process executed as an extraction unit and an analysis unit, the processes in S130 and S160 correspond to an example of a process executed as an output unit, and the process in S140 corresponds to an example of a process executed as a change acceptance unit. The user interface 14 corresponds to an example of a display device.

[0076] [2. Other Embodiments] While embodiments of this disclosure have been described above, it goes without saying that this disclosure is not limited to the embodiments described above and can take various forms.

[0077] (2a) In the above embodiment, the processor 11 extracts and outputs one panel data 23. However, as shown in Figure 8, the processor 11 may extract and output a panel dataset 230 that includes multiple panel data 23, not just one panel data 23. The panel dataset 230 may include multiple panel data 23 that are different from each other. That is, the panel dataset 230 may include a first panel data 23a which is panel data 23, and a second panel data 23b which is panel data 23 that is different from the first panel data 23a. The extraction conditions 30 corresponding to each of the multiple panel data 23 included in the panel dataset 230 may be different from each other.

[0078] With this configuration, the user can select the most appropriate panel data 23 from among the multiple extracted panel data 23. Therefore, the support device 1 can assist in extracting questionnaire results related to the target image.

[0079] (2b) Multiple survey results 22 may include one or more sets of surveys representing the results of responses from multiple survey respondents to multiple common questions. For example, each set of surveys could be "a survey of attitudes conducted by Company X with 2,000 people" or "a survey of actual conditions conducted by Organization Y with 1,500 people." Alternatively, each set of surveys could be the results of responses to surveys conducted periodically, such as monthly or annually.

[0080] The support device 1 may be configured to select one or more questionnaire sets from a plurality of questionnaire sets and use them as questionnaire data 21. The selection of questionnaire sets may be performed automatically using artificial intelligence 2 or the like based on the target profile, or it may be performed manually by the user through the user interface 14.

[0081] (2c) In the above embodiment, in S100, the processor 11 receives input of a target image through the user interface 14. However, the processor 11 may be configured to receive other information for extracting questionnaire results in addition to the target image. For example, the processor 11 may further accept input of product information relating to products corresponding to the target image. Examples of product information include the name of the product, the category of the product, and the company that sells the product.

[0082] Alternatively, the processor 11 may accept further input regarding the number of survey results 22 to extract. For example, the processor 11 may accept further input specifying a range of the number of survey results 22 to extract (e.g., 1000 to 2000). The processor 11 may include the other information received in S100 in the request signal that it inputs to the artificial intelligence 2 in S120.

[0083] This configuration improves the accuracy of extracting one or more survey results 22 related to the target profile.

[0084] (2d) In the above embodiment, the request signals that the processor 11 inputs to the artificial intelligence 2 in S120 and S150 include an extraction condition request for obtaining the extraction conditions 30 and an analysis request for obtaining the analysis result 40. The analysis request includes a suggestion request. However, the request signal does not have to include at least one of the extraction condition request and the analysis request. The analysis request does not have to include a suggestion request.

[0085] The processor 11 does not have to output at least one of the extraction conditions 30 and the analysis results 40 in S130 and / or S160. The analysis results 40 do not have to include suggestions that correspond to the suggestion request.

[0086] (2e) In the above embodiment, the processor 11 uses artificial intelligence 2 to extract one or more questionnaire results 22 related to the target profile from the questionnaire data 21. However, the processor 11 may extract the questionnaire results 22 without using artificial intelligence 2. For example, the processor 11 may be configured to extract the questionnaire results 22 based on keywords included in the input target profile. As an example, the processor 11 may be configured to include in the extraction conditions 30 a questionnaire result 22 that includes a combination of related questions 32 and related answer results 33 corresponding to assumed keywords (e.g., male or female) (e.g., a combination where the answer result to a question asking for gender is male or female).

[0087] (2f) Multiple functions of one component in the above embodiment may be realized by multiple components, or one function of one component may be realized by multiple components. Multiple functions of multiple components may be realized by one component, or one function realized by multiple components may be realized by one component. Some of the configurations of the above embodiment may be omitted. At least some of the configurations of the above embodiment may be added to or replaced with the configurations of other above embodiments.

[0088] (2g) The present disclosure can be implemented in various forms other than the support device described above. For example, it can be implemented in the form of a system in which the support device is a component, a computer program for causing a computer to function as the support device, a non-transitional physical recording medium such as a semiconductor memory on which the computer program is recorded, or a support method.

[0089] [Technical concepts disclosed in this specification] [Item 1] An input unit configured to accept input of a target image, An acquisition unit configured to acquire survey data including multiple survey results, An extraction unit configured to extract one or more survey results related to the target image as panel data based on the aforementioned survey data, An output unit configured to output the extracted panel data, Equipped with, Support equipment.

[0090] [Item 2] The support device described in item 1, Each of the aforementioned survey results includes the results of the survey respondents' answers to multiple questions, Support equipment.

[0091] [Item 3] The support device described in item 2, The output unit is configured to further output the extraction conditions for the panel data by the extraction unit. The aforementioned extraction conditions are expressed by multiple related combinations, Each of the aforementioned multiple related combinations is a combination of one related question among the multiple questions that is related to the target image, and a related response result that corresponds to the one related question and is related to the target image. Support equipment.

[0092] [Item 4] The support device described in item 3, The extraction conditions are expressed by a logical expression using the logical AND and / or OR of the multiple association combinations. Support equipment.

[0093] [Item 5] A support device as described in item 3 or item 4, The system further includes a change receiving unit configured to receive change instructions that instruct changes to at least one of the target image and the extraction conditions, The extraction unit is configured to re-extract the panel data based on the change instruction. The output unit is configured to output the re-extracted panel data. Support equipment.

[0094] [Item 6] A support device described in any one of items 1 through 5, The system further includes an analysis unit configured to analyze the characteristics of the panel data, The output unit is configured to further output the aforementioned features. Support equipment.

[0095] [Item 7] The support device described in item 6, The analysis unit is configured to input the panel data and a request signal requesting the analysis of the features to artificial intelligence, and to acquire the analysis result generated by the artificial intelligence as a response to the request signal. The output unit is configured to output the analysis results, as described above. Support equipment.

[0096] [Item 8] A support device described in any one of items 1 through 7, The extraction unit is configured to extract a panel dataset that includes a first panel data which is the panel data, and a second panel data which is different from the first panel data. The output unit is configured to output the panel dataset. Support equipment.

[0097] [Item 9] A support device described in any one of items 1 through 8, A display device configured to display the information output by the output unit, Equipped with, Support system.

[0098] [Item 10] A computer-based assistance method, Accepting input of the target image, Obtaining survey data that includes the results of multiple surveys, Based on the aforementioned survey data, extract one or more survey results related to the target profile as panel data, Outputting the extracted panel data, including, How to help.

[0099] [Item 11] A computer program that causes a computer to perform an action, The aforementioned operation is, Accepting input of the target image, Obtaining survey data that includes the results of multiple surveys, Based on the aforementioned survey data, extract one or more survey results related to the target profile as panel data, Outputting the extracted panel data, including, Computer program. [Explanation of Symbols]

[0100] 1...Support device, 11...Processor, 21...Survey data, 22...Survey results, 23...Panel data, 23a...First panel data, 23b...Second panel data, 230...Panel dataset.

Claims

1. An input unit configured to accept input of a target image expressed in natural language, An acquisition unit configured to acquire survey data including multiple survey results, An extraction unit configured to extract one or more survey results related to the target image as panel data based on the aforementioned survey data, An output unit configured to output the extracted panel data, Equipped with, The extraction unit is configured to extract the panel data based on a plurality of related combinations, Each of the aforementioned multiple related combinations is a combination of questions and answer results related to the target image, At least one of the aforementioned combinations of associations does not include in either the question or the answer result any keywords that constitute part of the natural language representing the target image received by the input unit. Support equipment.

2. The support device according to Claim 1, The extraction unit is configured to extract panel data by inputting a request signal to the artificial intelligence requesting it to extract one or more questionnaire results related to the target image from the questionnaire data, and acquiring the extraction results as a response to the request signal output by the artificial intelligence as panel data. Support equipment.

3. A support device according to claim 1 or claim 2, Each of the aforementioned survey results includes the results of the survey respondents' answers to multiple questions, Support equipment.

4. The support device according to claim 3, The output unit is configured to further output the extraction conditions for the panel data by the extraction unit. The extraction conditions are expressed by the multiple association combinations, Support equipment.

5. The support device according to claim 4, The extraction conditions are expressed by a logical expression using the logical AND and / or OR of the multiple related combinations. Support equipment.

6. The support device according to claim 4, The system further includes a change receiving unit configured to receive change instructions that instruct a change to at least one of the target image and the extraction conditions, The extraction unit is configured to re-extract the panel data based on the change instruction. The output unit is configured to output the re-extracted panel data. Support equipment.

7. A support device according to claim 1 or claim 2, The system further includes an analysis unit configured to analyze the characteristics of the panel data, The output unit is configured to further output the aforementioned features. Support equipment.

8. The support device according to claim 7, The analysis unit is configured to input the panel data and a request signal requesting the analysis of the features to artificial intelligence, and to acquire the analysis result generated by the artificial intelligence as a response to the request signal. The output unit is configured to output the analysis results, as described above. Support equipment.

9. A support device according to claim 1 or claim 2, The extraction unit is configured to extract a panel dataset that includes a first panel data which is the panel data, and a second panel data which is different from the first panel data. The output unit is configured to output the panel dataset. Support equipment.

10. The support device according to claim 1 or claim 2, A display device configured to display the information output by the output unit, Equipped with, Support system.

11. A computer-based assistance method, The system accepts input of a target image expressed in natural language, Obtaining survey data that includes the results of multiple surveys, Based on the aforementioned survey data, extract one or more survey results related to the target profile as panel data, Outputting the extracted panel data, Includes, The extraction includes extracting the panel data based on multiple related combinations, Each of the aforementioned multiple related combinations is a combination of questions and answer results related to the target image, At least one of the aforementioned combinations of associations does not include in either the question or the answer result any keywords that constitute part of the natural language representing the received target image. How to help.

12. A computer program that causes a computer to perform an action, The aforementioned operation is, The system accepts input of a target image expressed in natural language, Obtaining survey data that includes the results of multiple surveys, Based on the aforementioned survey data, extract one or more survey results related to the target profile as panel data, Outputting the extracted panel data, Includes, The extraction includes extracting the panel data based on multiple related combinations, Each of the aforementioned multiple related combinations is a combination of questions and answer results related to the target image, At least one of the aforementioned combinations of associations does not include in either the question or the answer result any keywords that constitute part of the natural language representing the received target image. Computer program.

Images