Information processing device, processing method for information processing device, and program

Abstract

The objective is to provide a mechanism for efficiently generating reports. [Solution] The information processing device comprises: a determination means for determining a report; a first output means for outputting information relating to a plurality of reports determined by the determination means and an instruction for generating a comparison report comparing the contents of the plurality of reports determined; and a first acquisition means for acquiring the comparison report generated based on the instruction output by the first output means.

Description

Technical Field

[0002]

[0001] The present invention relates to an information processing apparatus, a method for processing an information processing apparatus, and a program.

Background Art

[0002] Voices of customers (Voice of Customer, hereinafter VoC) collected through questionnaires and the like are very important information in product planning and marketing. However, since VoC in text format is qualitative data, a lot of time and labor are required for organization and analysis. Furthermore, when there is a large amount of text data, it becomes even more difficult to grasp its content.

[0003] Therefore, a method of analyzing VoC, which is a large amount of text data, by using a natural language processing technology approach has been established so far. Especially in recent years, new analysis methods using generative AI have emerged. In this method, a topic to be analyzed for a certain product is selected, and a report is generated using generative AI. Then, by having a human read and understand the organized report, efficient analysis becomes possible.

[0004] Non-Patent Document 1 describes a method of analyzing by creating a summary for each topic and creating a report using generative AI based on VoC.

Prior Art Documents

Non-Patent Documents

[0005]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] However, Non-Patent Literature 1 does not allow for the comparison of the characteristics of multiple subjects of analysis or the identification of trends in positive and negative opinions regarding a specific topic. Here, "multiple subjects of analysis" refers to different products, product groups, or the same product at different time periods. In other words, it is not possible, for example, to compare two products, to compare the overall trends of multiple product groups, or to compare opinions on a single product at different time periods.

[0007] As a concrete example, if there are competing printer products A and B, comparing the VoC (Voice of Customer) for printer product A with that of printer product B is a useful analysis for marketing printer product A. Specifically, this analysis could involve identifying the strengths and weaknesses of printer product A compared to printer product B from the perspectives of copy function, scan function, and ease of use of the user interface. However, a problem is that such an analysis is not envisioned in Non-Patent Document 1.

[0008] Therefore, it is desirable to provide a method for effectively and easily conducting comparative analysis with competing products when utilizing Voice of the Customer (VoC) to improve a product or formulate a competitive strategy.

[0009] Therefore, the present invention aims to provide a mechanism for efficiently generating reports. [Means for solving the problem]

[0010] The information processing device comprises: a determination means for determining a report; a first output means for outputting information relating to a plurality of reports determined by the determination means and an instruction for generating a comparison report comparing the contents of the plurality of reports determined; and a first acquisition means for acquiring the comparison report generated based on the instruction output by the first output means. [Effects of the Invention]

[0011] According to the present invention, a mechanism for efficiently generating reports can be provided. [Brief explanation of the drawing]

[0012] [Figure 1] This figure shows an example of the functional configuration of a VoC analysis system in an embodiment of the present invention. [Figure 2] This block diagram shows an example of the hardware configuration of a VoC analysis system in an embodiment of the present invention. [Figure 3] This figure shows an example of a VoCDB in an embodiment of the present invention. [Figure 4] This figure shows an example of an analysis report in an embodiment of the present invention. [Figure 5] This figure shows an example of a comparative report in an embodiment of the present invention. [Figure 6] This is a conceptual diagram of the analysis report generation and comparison report generation processes in an embodiment of the present invention. [Figure 7] This flowchart shows an example of the analysis report generation process in an embodiment of the present invention. [Figure 8] This is a flowchart showing an example of the comparison report generation process in an embodiment of the present invention. [Figure 9] This figure shows an example of prompt generation processing in an embodiment of the present invention. [Figure 10] This figure shows an example of the comparative analysis sentence generation process in an embodiment of the present invention. [Figure 11] This figure shows an example of a user interface for generating an analysis report in an embodiment of the present invention. [Figure 12] This figure shows an example of a user interface for generating a comparative report in an embodiment of the present invention. [Modes for carrying out the invention]

[0013] Embodiments of the present invention will be described in detail below with reference to the drawings.

[0014] Figure 1 is a diagram showing an example of the functional configuration of the VoC analysis system 100 in an embodiment of the present invention. The VoC analysis system 100 is an example of an information processing apparatus.

[0015] The VoC analysis system 100 includes an analysis report generation device 110, a VoC DB 120, a topic tagging device 130, and a comparison report generation device 140, and is connected to a client terminal 150.

[0016] The analysis report generation device 110 includes a user input reception unit 111, a search processing unit 112, and an analysis report generation unit 113. The VoC DB 120 is a VoC database. VoC is the voice of the customer (Voice of Customer) collected through questionnaires or the like.

[0017] The comparison report generation device 140 includes an analysis report reception unit 141, a prompt generation unit 142, and a comparison report generation unit 143.

[0018] Figure 2 is a block diagram showing an example of the hardware configuration of the VoC analysis system 100 in FIG. 1. Note that the client terminal 150 in FIG. 1 also has the same hardware configuration as the VoC analysis system 100.

[0019] As shown in FIG. 2, the VoC analysis system 100 is connected to a CPU (Central Processing Unit) 201, a ROM (Read Only Memory) 202, a RAM (Random Access Memory) 203, an input controller 205, a video controller 206, a memory controller 207, and a communication I / F controller 208 via a system bus 204.

[0020] The CPU 201 comprehensively controls each device and controller connected to the system bus 204.

[0021] ROM202 or external memory211 holds computer-readable and executable programs (including data tables) necessary for implementing control programs executed by the CPU201, such as the BIOS (Basic Input / Output System) and OS (Operating System), as well as the flowcharts shown in Figures 7 and 8 described later.

[0022] RAM203 functions as the main memory, work area, etc., of the CPU201. The CPU201 loads the necessary programs, etc., from ROM202 or external memory211 into RAM203, and then executes the loaded programs to perform various operations.

[0023] The input controller 205 controls input from input devices 209, such as keyboards and pointing devices like mice. If the input device 209 is a touch panel, the user can give various instructions by pressing (touching with a finger, etc.) icons, cursors, or buttons displayed on the touch panel.

[0024] Furthermore, the touch panel may be a multi-touch screen or other touch panel capable of detecting the positions of multiple fingers touching it.

[0025] The video controller 206 controls the display to an external output device such as the display 210. The display 210 may include the display of a notebook computer integrated with the main unit. Note that the external output device is not limited to a display, but may also be a projector, for example. Furthermore, for the aforementioned touch-enabled device, an input device 209 is also provided.

[0026] The video controller 206 can control the video memory (VRAM) for display control, and can utilize a portion of the RAM 203 as the video memory area, or it is possible to provide a separate dedicated video memory.

[0027] The memory controller 207 controls access to the external memory 211. The external memory 211 can be an external storage device (hard disk), a flexible disk (FD), or a CompactFlash® memory connected to a PCMCIA card slot via an adapter, which stores boot programs, various applications, font data, user files, editing files, and various other data.

[0028] The communication interface controller 208 connects to and communicates with external devices via a network and performs communication control processing over the network. For example, it can handle communication using TCP / IP, telephone lines such as ISDN, and 3G mobile phone lines.

[0029] The CPU 201 enables display on the display 210 by, for example, performing the process of expanding (rasterizing) outline fonts into the display information area in RAM 203. The CPU 201 also enables user input via a mouse cursor (not shown) on the display 210.

[0030] Each functional component in Figure 1 is realized by the CPU 201 executing a program.

[0031] Figure 3 shows an example of the VoCDB120 in Figure 1. In addition to the main text 301, the VoCDB120 includes a product item 302 to identify which product the VoC is about, and a posting date and time item 303. For example, for VoC record 304, the main text is "Printer product A leaves streaks when copying," indicating that it refers to "printer product A" and was posted on August 9, 2023. Here, items that reflect the content of the main text, such as the product item 302, may be mechanically created from the main text using known techniques such as feature word extraction, or they may be manually judged and entered. Also, for the sake of simplicity, only two items, the product item 302 and the posting date and time item 303, are shown as examples here, but these are not required, and additional items used as part of the VoC analysis system 100, such as the results of positive / negative analysis, may be included.

[0032] Returning to Figure 1, the topic tagging device 130 is a device for determining what topics are mentioned in the body text 301 of each VoC record. For example, in the case of VoC record 304, since it is an opinion about the copy function, the topic tag "copy function" is added. Here, the process of creating a list of selectable topics or determining which topic to add from the VoC body text may be done using known technologies such as feature word extraction or tagging by generation AI, or it may be done manually by a user or record registrant. Furthermore, the topic tagging results only need to be usable in the search processing unit 112 of the analysis report generation device 110, which will be described later. As shown in Figure 1, the configuration may be to add topic tags to records in the VoCDB 120, or it may be to store them in a separate DB or file.

[0033] The analysis report generation device 110 includes a user input receiving unit 111, a search processing unit 112, and an analysis report generation unit 113.

[0034] The user input receiving unit 111 is a device for receiving user input to the analysis report generation device 110. Users can send filtering conditions and analysis topics for selecting analysis targets to the user input receiving unit 111 via a web browser on a client terminal 150. For example, if a user wants to analyze "the copy function of printer product A", they would send the filtering condition "product item: printer product A" and the topic tag "copy function".

[0035] The search processing unit 112 receives user input from the user input receiving unit 111 and is a processing unit for obtaining VoC records from the VoCDB 120 that match the specified filtering conditions and topic. Here, known techniques can be used to search for VoC records that match the filtering conditions and topic, such as full-text search of the main text 301 or filtering on items provided in the VoCDB 120, such as product items 302.

[0036] The analysis report generation unit 113 generates a VoC analysis report based on the VoC records acquired by the search processing unit 112. The analysis report may be created using a generation AI. AI is artificial intelligence. In this embodiment of the present invention, the analysis report generation unit 113 is configured to use a generation AI, but it is not particularly limited, and a large-scale language model, machine learning model, and other methods capable of generating text may be used. Also, in this embodiment, the analysis report generation unit 113 is located inside the VoC analysis system 100, but these may be located outside, for example, by obtaining the generated report via an API.

[0037] Figure 4 shows an example of an analysis report 400 generated by the analysis report generation device 110. The analysis report 400 includes analysis results 410 and metadata 420.

[0038] Analysis Result 410 is a summary of the analysis results in a format that is easy for the user to read. The analysis target of this analysis result is specified by the user's filtering conditions, and the analysis axis is specified by topic. Here, Analysis Result 410 includes Topic 411 and VoC Summary 412 as items. Although Topic 411 and VoC Summary 412 are shown as examples of the minimum configuration to explain this embodiment, additional items to be used as analysis results, such as the number of VoC records used in the summary, may be included.

[0039] Metadata 420 describes the filtering conditions specified by the user. Metadata 420 includes filtering condition 421 as an item. This allows it to retain information about what the user wants to analyze. In the example in Figure 4, metadata 420 shows that the user has filtered by the condition "Product: Printer Product A," indicating that the user wants to analyze the VoC about "Printer Product A."

[0040] Returning to Figure 1, the comparison report generation device 140 includes an analysis report receiving unit 141, a prompt generation unit 142, and a comparison report generation unit 143.

[0041] The analysis report receiving unit 141 accepts multiple analysis reports generated by the analysis report generation device 110 as input. This can be done by having the user send the analysis reports received from the analysis report generation device 110, or by having the generated analysis reports stored in the VoC analysis system 100, from which the user can select multiple analysis reports to compare.

[0042] The prompt generation unit 142 creates prompts for generating comparison reports based on multiple analysis reports received by the analysis report receiving unit 141. In this embodiment, prompts are created using a generation AI. Here, the AI ​​infers what aspects of comparison the user is requesting based on the differences in the metadata of the analysis reports and includes this in the prompt. A prompt is an instruction to the generation AI.

[0043] The comparison report generation unit 143 generates a comparison report based on multiple analysis reports received by the analysis report receiving unit 141 and prompts generated by the prompt generation unit 142. In this embodiment of the present invention, the comparison report generation unit 143 is configured to use a generation AI, but it is not limited to this, and a large-scale language model, machine learning model, or other method capable of generating text may be used. In this embodiment, the comparison report generation unit 143 is located inside the VoC analysis system 100, but these may be located outside, for example, by obtaining the generated reports via an API. Furthermore, in this embodiment, the analysis report generation device 110 and the comparison report generation device 140 are configured as separate devices, but this is not limited to this, and the generation of analysis reports and comparison reports may be performed by the same device. The prompt generation process and the comparison report generation process will be explained in detail later using examples.

[0044] Figure 5 shows an example of a comparison report 500 generated by the comparison report generation device 140. The comparison report 500 includes topic-specific analysis results 510 and overall analysis results 520.

[0045] The topic-specific analysis results 510 include topics 511 and comparative analysis statements 512, allowing users to view the comparative analysis results of VoC for each topic.

[0046] The overall analysis result 520 is the result of a comparative analysis of each analysis target by combining all topics. However, the comparison report 500 may include both the topic-specific analysis results 510 and the overall analysis result 520, or it may include only one of them.

[0047] Next, the overall picture of the process in this embodiment will be explained using Figure 6. The process in this embodiment includes an analysis report generation phase 610 performed by the analysis report generation device 110 and a comparison report generation phase 620 performed by the comparison report generation device 140.

[0048] The analysis report generation phase 610 is the phase in which an analysis report is generated with the analysis target and topic specified by the user. Here, in order to compare the analysis targets in the comparison report generation phase 620, this analysis report generation process is executed multiple times, generating multiple analysis reports 611 and 612.

[0049] The comparison report generation phase 620 takes the multiple analysis reports 611 and 612 output in the analysis report generation phase 610 as input and generates a comparison report 621 that compares the respective analysis targets. For example, when analysis reports 611 and 612 from the previous example are used as input, the generated comparison report 621 compares the "copy function of printer product A and printer product B," and includes information such as "printer product A's copy function is known to have better color reproduction than printer product B's copy function."

[0050] In this embodiment, in order to generate the comparison report 621 in the comparison report generation phase 620, it is assumed that the generated analysis reports 611 and 612 are generated for different analysis targets but for the same topic. For example, if analysis report 611 has the filtering condition "Product: Printer Product A" and the topic "Copy Function," while analysis report 612 has the filtering condition "Product: Printer Product B" and the topic "Copy Function," then these two analysis reports 611 and 612 can be compared. However, if some analysis reports lack descriptions of a particular topic, it is also possible to adopt a configuration in the comparison report generation phase 620 that retrieves the corresponding VoC record with the filtering condition and topic from the VoCDB 120 and supplements the information.

[0051] Next, using the flowchart in Figure 7, the processing method executed by the analysis report generation device 110 in the analysis report generation phase 610 in the embodiment of the present invention will be described.

[0052] In step S701, the user input receiving unit 111 obtains the filtering conditions α specified by the user, and one or more topic groups T that the user wants to analyze.

[0053] In step S702, the search processing unit 112 searches the VoCDB 120 and obtains a group of VoC records R that match the filtering condition α.

[0054] In step S703, the topic tagging device 130 estimates which topic each VoC record in the VoC record group R relates to from among multiple topic options, and adds a topic tag. This result may be stored by adding a topic tag attribute to the VoC record in the VoCDB 120, or it may be stored by other methods such as outputting the VoC record group R with topic tags attached to a separate file. Also, for the sake of explanation, the topic estimation and topic tagging performed in step S703 are shown to be performed on each VoC record in the VoC record group R, but the processing order may be changed so that step S703 is performed before step S702, and this operation is performed on all VoC records included in the VoCDB 120.

[0055] In step S704, the analysis report generation device 110 starts repeating the process for each topic in the analysis target topic group T. Let the topic at this time be t.

[0056] In step S705, the search processing unit 112 extracts and retrieves records with topic t as a topic tag from the group of topic-tagged VoC records R stored in step S703. Let the extracted group of VoC records be R'.

[0057] In step S706, the analysis report generation unit 113 uses the generation AI to summarize the VoC record group R' and generate a single sentence. Here, by repeatedly executing steps S704 to S706, analysis sentences for each topic t of the analysis target topic group T can be generated. If it is not possible to use all of the VoC record group R' as input to the generation AI due to token limitations or other reasons, an appropriate number of records may be sampled from the VoC record group R' and used as input to the generation AI to create the summary sentence.

[0058] In step S707, the analysis report generation unit 113 describes each topic 411 and the VoC summary 412 in the analysis results 410, and the filtering conditions 421 in the metadata 420, and outputs these together as the analysis report 400.

[0059] Next, using the flowchart in Figure 8, the processing method executed by the comparison report generation device 140 in the comparison report generation phase 620 in the embodiment of the present invention will be described.

[0060] In step S801, the analysis report receiving unit 141 acquires multiple analysis reports as a group of analysis reports. Here, the group of analysis reports received is denoted as L.

[0061] In step S802, the prompt generation unit 142 obtains metadata 420 from each analysis report in the analysis report group L. Here, the multiple metadata obtained from the analysis report group L are referred to as metadata group M.

[0062] In step S803, the prompt generation unit 142 takes the metadata group M as input and generates prompts for creating a comparison report using the generation AI. The prompts created here are denoted as prompt P.

[0063] Figure 9 shows an example of the prompt generation process performed in step S803. If the metadata group M includes items with filtering conditions "Product: Printer Product A" 901 and "Product: Printer Product B" 902, it is inferred that the user wants to analyze the strengths and weaknesses of each printer product. The prompt generation unit 142 provides the generation AI 900 with instruction data 903 that includes the metadata contents of "Product: Printer Product A" 901 and "Product: Printer Product B" 902 in the prompt. The generation AI 900 outputs prompt 904. Prompt 904 is prompt P which includes the words, "Clearly indicate the strengths and weaknesses of each product and evaluate them fairly."

[0064] As another example, if the filtering conditions include "Product: Printer Product A, and Period: 2023" and "Product: Printer Product A, and Period: 2024," it can be inferred that the user wants to analyze "the change in reputation of Printer Product A over time." Therefore, the prompt P generated by the prompt generation unit 142 will include wording such as, "Please tell us whether the reputation of Printer Product A has improved / worsened, and why." In other words, the system can infer what aspects the user wants to compare the reports from based on the filtering conditions selected by the user, and reflect the comparison points the user desires in the instructions for generating the comparison report. This allows for the efficient creation of instructions for users to compare multiple reports. Furthermore, even someone unfamiliar with report creation can generate instructions that include comparison points simply by selecting the reports they want to compare, resulting in the creation of high-quality comparison reports.

[0065] For the sake of simplicity, this explanation assumes that only metadata group M is included as input for the prompt generation process. However, other information related to the analysis report, such as topic 411 of the analysis report or tags assigned to the analysis report, may also be used. Furthermore, a step may be implemented in which the user can modify the prompt after it has been automatically generated in step S803. In addition, two prompts may be created: prompt P for generating the analysis results 510 for each topic, and prompt P' for generating the overall analysis results 520.

[0066] Returning to Figure 8, in step S804, the comparison report generation unit 143 extracts common parts from the topic groups of each analysis report in the analysis report group L. Let the extracted topic group be T'. For example, if the analysis report group L consists of analysis report 1 and analysis report 2, and the topics included in analysis report 1 are "copy function, toner replacement function, scan function", and the topics included in analysis report 2 are "copy function, scan function, network connection function", then the extracted topic group T' will be "copy function, scan function".

[0067] Here, if the extracted topic group T' is an empty set, an error may occur and the comparison report generation process may be terminated. Alternatively, as mentioned above, VoC summaries for the missing topics may be added. In other words, following the previous example, analysis report 1 lacks a VoC summary for the "Network Connectivity" topic compared to analysis report 2. Therefore, VoC records matching the filtering conditions of analysis report 1 and the topic "Network Connectivity" may be retrieved from VoCDB120, the summary may be rewritten, and "Network Connectivity" may be added to the extracted topic group T'.

[0068] In step S805, the comparison report generation unit 143 starts iterative processing for each topic in topic group T'. Let this topic be called topic t.

[0069] In step S806, the comparison report generation unit 143 obtains summary sentences related to topic t from each analysis report in the analysis report group L. The multiple obtained summary sentences are called the summary sentence group S. To illustrate the process performed for one analysis report in the example in Figure 4, when topic t is "copy function", the VoC summary 412 of analysis result row 413 is obtained. This process is performed for each analysis report in the analysis report group L.

[0070] In step S807, the comparison report generation unit 143 takes the prompt P and the description of topic t from the summary sentence group S as input and uses the generation AI 900 to generate a comparison analysis sentence about topic t.

[0071] Figure 10 shows an example of the comparative analysis statement generation process performed in step S807. Analysis report 1001 is the analysis report for printer product A. Analysis report 1002 is the analysis report for printer product B. Data 1003 is the data obtained by adding the summary sentence group S, which consists of "summary sentence about the copy function of printer product A" and "summary sentence about the copy function of printer product B," to prompt P. The comparative report generation unit 143 inputs data 1003 to the generation AI 900. As a result, the generation AI 900 outputs a comparative analysis statement 1004 for printer product A and printer product B regarding the topic "copy function." The comparative analysis statement 1004 corresponds to the topic-specific analysis results 510 in Figure 5.

[0072] In step S808, the comparison report generation unit 143 generates a comprehensive comparative analysis statement using the generation AI 900 based on the summary sentences of all topics in the extracted topic group T'. The comprehensive comparative analysis statement corresponds to the comprehensive analysis result 520 in Figure 5. This process uses the prompts created in step S803, similar to step 807, but here the summary sentences of all topics in the analysis report are used as input. The generation AI 900 creates the comprehensive comparative analysis statement using these prompts and summary sentences as input. Note that the topic-specific comparative analysis statements created in steps S804 to S807 and the comprehensive comparative analysis statement created in step S808 may be created individually or both.

[0073] In step S809, the comparison report generation unit 143 combines the comparison analysis statements for each topic generated in step S807, generates topic-specific analysis results 510, and outputs a comparison report 500 by combining these with the overall analysis results 520 of the overall comparison analysis statement generated in step S808.

[0074] Figure 11 shows an example of the user interface displayed by the analysis report generation device 110 on the browser of the client terminal 150. The analysis report generation screen 1100 includes a filtering condition specification component 1101, a VoC record display component 1102, a topic selection component 1103, and an analysis report creation button 1104.

[0075] The user specifies the target of analysis using the filtering condition specification component 1101. The filtering condition specification component 1101 can be in any format as long as it specifies filtering conditions for VoCDB120; it can be in a free-text format or a format where the user selects from provided items.

[0076] The VoC record display component 1102 is a component that displays VoC records obtained by the search processing unit 112 from the VoCDB 120 according to the conditions specified in the filtering condition specification component 1101.

[0077] The user selects the topics they want to analyze in the analysis report using the topic selection component 1103. For the purposes of this explanation, the topic selection component 1103 is shown as a checkbox-based selection from a set of pre-defined topics, but other formats such as a dropdown menu are also acceptable. Furthermore, the multiple topics that serve as options may be pre-defined, or a separate component may be provided to describe the topics of the options.

[0078] The user can start the analysis report generation process of the analysis report generation device 110 by pressing the analysis report creation button 1104.

[0079] Figure 12 shows an example of a user interface displayed by the comparison report generation device 140 on the browser of the client terminal 150. The comparison report generation screen 1200 includes an analysis report input component 1201, a prompt generation button 1202, a prompt display component 1203, and a comparison report creation button 1204.

[0080] The analysis report input component 1201 is a component for the user to input multiple analysis reports to be compared. For illustrative purposes, this example uses a method where the user uploads two analysis reports, but it is also possible to use a method where the user selects analysis reports stored within the VoC analysis system 100. Alternatively, it is possible to use a method where any number of analysis reports (two or more) can be input. Furthermore, in addition to the user selecting analysis reports, the system may be configured to automatically determine analysis reports for specific themes or topics entered by the user.

[0081] By pressing the prompt generation button 1202, the user can start the process of creating a prompt for generating a comparison report, as shown in step S803 of Figure 8.

[0082] The prompt display component 1203 is a component that displays the prompt created during the prompt generation process. For simplicity of explanation, it is described here as a component that displays the prompt, but a function that allows the user to modify the prompt may also be added.

[0083] The user can initiate the comparison report generation process by pressing the comparison report creation button 1204.

[0084] Furthermore, if the user does not have a function to modify the prompt and the prompt created by the prompt generation process is used as is for generating the comparison report, the prompt generation button 1202 and the prompt display component 1203 may be omitted.

[0085] Furthermore, in embodiments of the present invention, additional functions can be implemented for user convenience. For example, to enable users to understand the background of the evaluation more concretely and directly, a function could be implemented in which, when a product difference in a specific topic is selected, the VoC record that formed the basis of the evaluation is displayed as an example. This function can be achieved by retrieving VoC records from VoCDB120 that match the relevant filtering conditions and have the relevant topic tags.

[0086] As described above, in step S801 of Figure 8, the analysis report receiving unit 141 functions as a determination unit, and the analysis report input component 1201 determines the analysis report.

[0087] In step S803, the prompt generation unit 142 functions as an output unit and outputs instructions to the generation AI 900 for generating instructions that include comparison points for comparing the contents of multiple analysis reports determined in step S801, and metadata group M. These instructions and metadata group M are included in the instruction data 903 in Figure 9. The metadata group M includes filtering conditions (search conditions) 421.

[0088] The prompt generation unit 142 then functions as an acquisition unit and acquires the instruction statement generated by the generation AI 900 based on the instructions for generating an instruction statement that includes the above comparison points. This instruction statement corresponds to prompt 904 in Figure 9.

[0089] In step S807, the comparison report generation unit 143 functions as an output unit and outputs to the generation AI 900 information relating to the multiple analysis reports determined in step S801, and an instruction to generate a comparison report comparing the contents of the multiple analysis reports. The information relating to the analysis reports includes, for example, a summary of each topic in the analysis reports. The instruction to generate a comparison report includes a prompt 904.

[0090] The comparison report generation unit 143 then functions as an acquisition unit and acquires topic-specific analysis results 510 for each topic of the comparison report generated by the generation AI 900 based on the instructions output above.

[0091] In step S808, the comparison report generation unit 143 outputs to the generation AI 900 information relating to the multiple analysis reports determined in step S801, and an instruction to generate a comparison report comparing the contents of the multiple analysis reports. The information relating to the analysis reports includes, for example, summaries of all topics in the analysis reports. The instruction to generate a comparison report corresponds to prompt P.

[0092] Then, the comparison report generation unit 143 obtains the overall analysis results 520 of all topics in a portion of the comparison report generated by the generation AI 900 based on the instructions output above.

[0093] The analysis report 400 in Figure 4 is an analysis report generated based on search conditions set by the user in the filtering condition specification component 1101 in Figure 11, and the search conditions based on the topic selected by the user in the topic selection component 1103. The information targeted for the above search is VoC (Voice of the Customer) from VoCDB120.

[0094] Furthermore, the analysis report 400 may also be an analysis report generated based on information retrieved using search criteria that are based on at least one of the following: conditions set by the user in the filtering condition specification component 1101, and topics selected by the user in the topic selection component 1103.

[0095] The conditions set by the user in the filtering condition specification component 1101 are, for example, the type of printer product. The topics selected by the user in the topic selection component 1103 are, for example, the copy function or the scan function.

[0096] According to this embodiment, the VoC analysis system 100 can generate a comparative report that compares and analyzes each analysis target from multiple analysis reports summarizing VoC. This provides a mechanism for efficiently generating reports.

[0097] Furthermore, the VoC analysis system 100 can automatically generate analytical perspectives when performing comparative analysis. This reduces the mental burden on the user. Moreover, even users who lack the skills to translate appropriate analytical perspectives into prompts can create comparative reports of a certain quality because appropriate prompts are automatically generated.

[0098] In addition, if a particular topic is not included in some of the multiple analysis reports, it becomes possible to adopt a structure that complements that information.

[0099] Although embodiments have been described above, the present invention can take the form of, for example, a system, apparatus, method, program, or recording medium. Specifically, it may be applied to a system consisting of multiple devices, or to an apparatus consisting of a single device.

[0100] Furthermore, the program in this invention is a program that a computer can execute using the processing methods shown in the flowcharts in Figures 7 and 8, and the storage medium of this invention stores a program that a computer can execute using the processing methods in Figures 7 and 8. Note that the program in this invention may also be a separate program for each of the processing methods shown in Figures 7 and 8.

[0101] As described above, it goes without saying that the object of the present invention can also be achieved by supplying a recording medium containing a program that realizes the functions of the embodiments described above to a system or device, and by having the computer (or CPU or MPU) of that system or device read and execute the program stored on the recording medium.

[0102] In this case, the program read from the recording medium itself realizes the novel function of the present invention, and the recording medium on which that program is recorded constitutes the present invention.

[0103] For recording media used to supply programs, examples include flexible disks, hard disks, optical disks, magneto-optical disks, CD-ROMs, CD-Rs, DVD-ROMs, magnetic tapes, non-volatile memory cards, ROMs, EEPROMs, silicon disks, and the like.

[0104] Furthermore, it goes without saying that the functions of the aforementioned embodiments are realized not only by the computer executing the program it has read, but also by the operating system (OS) running on the computer performing some or all of the actual processing based on the instructions of that program, thereby realizing the functions of the aforementioned embodiments.

[0105] Furthermore, it goes without saying that this also includes cases where, after a program read from a recording medium is written to the memory of a function expansion board inserted into a computer or a function expansion unit connected to a computer, the CPU or other components of the function expansion board or function expansion unit perform some or all of the actual processing based on the instructions of the program code, and the functions of the aforementioned embodiments are realized through that processing.

[0106] Furthermore, the present invention may be applied to a system composed of multiple devices or to a device consisting of a single device. It goes without saying that the present invention can also be applied when the results are achieved by supplying a program to a system or device. In this case, by reading a recording medium containing a program for achieving the present invention into the system or device, the system or device can enjoy the effects of the present invention.

[0107] Furthermore, by downloading and reading the program for achieving the present invention from a server, database, etc. on a network using a communication program, the system or device can enjoy the effects of the present invention. It should be noted that configurations combining the above-described embodiments and their modifications are also included in the present invention. [Explanation of Symbols]

[0108] 100 VoC Analysis System 110 Analysis Report Generator 111 User input receiving unit 112 Search Processing Unit 113 Analysis Report Generation Department 120 VoCDB 130 Topic Tagging Device 140 Comparison Report Generator 141 Analysis Report Receiving Section 142 Prompt generation unit 143 Comparison Report Generation Unit 150 client terminals

Claims

1. The decision-making process for determining the report, A first output means that outputs information relating to a plurality of reports determined by the determination means and an instruction to generate a comparison report comparing the contents of the plurality of reports determined, An information processing apparatus comprising: a first acquisition means for acquiring a comparison report generated based on an instruction output by the first output means.

2. The information processing device according to claim 1, characterized in that the information relating to the report includes a summary of the report.

3. The information processing device according to claim 2, characterized in that the information relating to the report includes a summary of the topic of the report.

4. The first output means outputs information relating to the plurality of reports and an instruction to generate the comparison report to the generating AI. The information processing apparatus according to claim 1, characterized in that the first acquisition means acquires a comparison report generated by the generating AI.

5. Furthermore, a second output means outputs instructions for generating instructions that include comparison points for comparing the contents of multiple reports determined by the determination means, The system includes a second acquisition means for acquiring an instruction sentence generated based on an instruction for generating an instruction sentence that includes the aforementioned comparison viewpoint, The information processing apparatus according to claim 1, characterized in that the instruction output by the first output means includes an instruction statement acquired by the second acquisition means.

6. The information processing apparatus according to claim 1, characterized in that the report determined by the determination means is a report generated based on information retrieved using search conditions that are based on at least one of the conditions set by the user and the topic selected by the user.

7. The report determined by the aforementioned determination means is a report generated based on information retrieved using search criteria that are based on at least one of the conditions set by the user and the topic selected by the user. The information processing apparatus according to claim 5, characterized in that the second output means outputs an instruction for generating the instruction statement and the search conditions.

8. The information processing device according to claim 6, characterized in that the information subject to the search is customer feedback.

9. The conditions set by the user are the type of printer product, The information processing apparatus according to claim 6, characterized in that the topic selected by the user is a copy function or a scan function.

10. The second output means outputs instructions for generating the instruction sentence to the generation AI, The second acquisition means acquires the instruction text generated by the generation AI, The first output means outputs information relating to the plurality of reports and an instruction to generate the comparison report to the generation AI. The information processing apparatus according to claim 5, characterized in that the first acquisition means acquires a comparison report generated by the generating AI.

11. The information processing apparatus according to claim 1, characterized in that the report determined by the determination means is a report generated based on information retrieved based on conditions set by the user and search conditions based on topics selected by the user.

12. The conditions set by the user are the type of printer product, The information processing device according to claim 11, characterized in that the topic selected by the user is a copy function or a scan function.

13. The decision-making process involves a decision step to determine the report, The first output means outputs information relating to a plurality of reports determined by the determination step and an instruction to generate a comparison report comparing the contents of the plurality of reports determined, A processing method for an information processing apparatus, characterized in that the first acquisition means includes a first acquisition step of acquiring a comparison report generated based on an instruction output by the first output step.

14. A program for causing at least one computer to function as one of the means of an information processing device described in any one of claims 1 to 8.

Images