Information processing device, information processing method, and program

Abstract

To enable a user to easily reach a desired target sentence or the like. [Solution] When an FAQ answer or product description page is presented to a user to achieve the user's goal of answering a FAQ or explaining a product, the target sentence is taken as the target sentence, etc., and the candidate guide sentence intended for the user to reach the target sentence, etc., is stored in a guide sentence dictionary, with the keyword K, the candidate guide sentence, and the target sentence, etc., associated so that the candidate guide sentence is extracted when the keyword K is input.

Description

Technical Field

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

Background Art

[0002] As an auxiliary service for assisting users who receive goods or services, there is a technology related to a so-called FAQ search service that can search for information related to a combination of typical questions and answers (for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in conventional FAQ search services, including the technology of Patent Document 1, even when a general user executes a search with a keyword that comes to mind, it often does not hit the desired search results. Also, it is often impossible to reach the desired information because there are too many hits. Specifically, for example, in the case of a FAQ search service for insurance products, when a search is executed with the keyword "update", a huge amount of search results are displayed, making it difficult to reach the desired information.

[0005] The present invention has been made in view of such a situation, and an object thereof is to enable a user to easily reach a desired target sentence or the like.

Means for Solving the Problems

[0006] To achieve the above object, an information processing apparatus according to an aspect of the present invention is Dictionary generation means that stores in a guided sentence dictionary a predetermined word, phrase, or sentence presented to the user to achieve a predetermined purpose, as the target sentence, etc., and a guided sentence candidate that the user intends to reach the target sentence, etc., which includes a more general word related to the target sentence, etc. or a similar word, etc., in order to extract such a guided sentence candidate when the word is entered, by associating the word, the guided sentence candidate, and the target sentence, etc. It is equipped with.

[0007] Each of the information processing method and program according to one aspect of the present invention corresponds to each of the method and program corresponding to the information processing apparatus according to one aspect of the present invention. [Effects of the Invention]

[0008] According to the present invention, users will be able to easily find the desired target text or other relevant information. [Brief explanation of the drawing]

[0009] [Figure 1A] This figure shows an overview of the service that can be realized by an information processing system to which a server according to one embodiment of the present invention is applied. [Figure 1B] This figure shows an overview of the service that can be realized by an information processing system to which a server according to one embodiment of the present invention is applied. [Figure 1C] This figure shows an overview of the service that can be realized by an information processing system to which a server according to one embodiment of the present invention is applied. [Figure 2A] This figure shows an overview of the service that can be realized by an information processing system to which a server according to one embodiment of the present invention is applied. [Figure 2B] This figure shows an overview of the service that can be realized by an information processing system to which a server according to one embodiment of the present invention is applied. [Figure 3] This figure shows an overview of the service that can be realized by an information processing system to which a server according to one embodiment of the present invention is applied. [Figure 4] It is a diagram showing an example of the configuration of an information processing system to which a server according to an embodiment of the information processing apparatus of the present invention is applied. [Figure 5] It is a block diagram showing an example of the hardware configuration of the server in the information processing system shown in FIG. 4. [Figure 6] It is a functional block diagram showing an example of the functional configuration for executing the question dictionary generation process, related text extraction process, and question dictionary improvement process among the functional configurations of the server in FIG. 5 that constitutes the information processing system of FIG. 4. [Figure 7] It is a diagram showing an example of a search in a conventional FAQ search site. [Figure 8] It is a diagram for explaining the concept of intention expansion in this service. [Figure 9] It is a diagram showing an example of user guidance realized by the intention expansion of this service. [Figure 10] It is a diagram showing an example of user guidance realized by intention prediction in this service. [Figure 11] It is a diagram showing an example of presenting question candidates in this service. [Figure 12] It is a diagram showing an example in which a user selects a question candidate in this service. [Figure 13] It is a functional block diagram showing an example of the functional configuration for executing the process of allowing the user to search for and present target texts and the like among the functional configurations of the server in FIG. 5.

Embodiments for Carrying Out the Invention

[0010] [[ID=*35]] Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[0011] First, referring to FIGS. 1 to 3, an overview of a service (hereinafter referred to as "this service") that can be realized by an information processing system (see FIG. 3 described later) to which a server 1 according to an embodiment of the present invention is applied will be described.

[0012] FIGS. 1 to 3 are diagrams showing an overview of the present service that can be realized by an information processing system to which a server according to an embodiment of the present invention is applied.

[0013] This service supports easier access to a predetermined word, phrase, or sentence presented to the user as a target sentence or the like in order to achieve the user's predetermined purpose. The target sentence or the like may have any content (content) as long as it is presented to the user to achieve the user's predetermined purpose. However, in the following description, the target sentence or the like will be described as being an FAQ and an answer sentence (article) thereto. Such a target sentence or the like will be appropriately referred to as related article or the like W. That is, this service will be described as a service that provides a search function or the like of a FAQ search site.

[0014] This service is a service provided from a service provider (not shown) to a user (not shown). The user is a person who uses this service. The user operates an information processing device 2 such as a smartphone (hereinafter referred to as "user terminal 2") to access a web site for FAQ search (hereinafter referred to as "this FAQ search site") provided by this service. Then, the user inputs a predetermined keyword K into a search window B provided in a part of this FAQ search site and executes a search (step S1).

[0015] Even if the keyword K input into the search window B in this FAQ search site is highly general, the user can reach the sentence on the FAQ screen that the user desires. Here, the "generality" of the keyword K refers to an index indicating the breadth of the range of meanings that can be recalled from the keyword K itself. A keyword K with a high generality has a wide range of meanings that can be recalled from itself, and a keyword K with a low generality has a narrow range of meanings that can be recalled from itself.

[0016] As a concrete example, suppose a user who is working overtime late into the night experiences a problem where their company terminal loses internet connectivity. In such a case, the user uses their smartphone (user terminal 2) to access this FAQ search site. The user then enters a possible keyword K into search box B to quickly find an FAQ that will solve the problem and performs a search.

[0017] For example, suppose a user enters the keyword K1 "cannot connect" into search box B and performs a search. The keyword K1 "cannot connect" suggests a state where something that should be connected in some way is not connected. However, the keyword K1 "cannot connect" alone does not allow us to specifically recall what is not connected and in what way. In other words, the keyword K1 "cannot connect" has a high degree of generality.

[0018] Now, let's assume a user uses a conventional FAQ search service to search for the keyword K1, "cannot connect." In this case, the search results will display numerous FAQs that partially contain the keyword K1. However, these search results will also include many FAQs that are not related to internet connectivity, making it difficult for the user to find the FAQ they are looking for.

[0019] For example, suppose a user enters the keyword K2 "Internet" into search box B and performs a search. The keyword K2 "Internet" evokes the general concept of the internet. However, the keyword K2 "Internet" alone does not evoke any specific details about what the internet does. In other words, the keyword K2 "Internet" has a high degree of generality. Now, let's assume a user uses a conventional FAQ search service to search for the keyword K2 "Internet." In this case, the search results will display numerous FAQs that contain the keyword K2 "Internet" in part. However, these search results will also include many FAQs that are not related to internet connectivity, making it difficult for the user to find the FAQ they are looking for.

[0020] In contrast, when using this FAQ search site, if a common keyword K is entered into the search box B, one or more candidate question sentences C (hereinafter referred to as "candidate question sentence C") that contain keyword K in at least part are first displayed (step S2). Multiple candidate question sentences C are generated in advance and are stored and managed in a predetermined database (for example, the question sentence dictionary DB181 in Figure 5 described later) in a manner that allows for searching and extraction.

[0021] Specifically, as shown in Figure 1A, when the keyword K1 "cannot connect" is entered into the search box B, a suggested question C1 "cannot connect to the company's wireless network" is displayed. For example, as shown in Figure 1B, when the keyword K2 "Internet" is entered into search box B, a suggested question C2, "How do I connect to the company's internet?", is displayed. In the example in Figure 1, only one candidate question C (candidate question C1 and C2) is displayed, but two or more candidate question Cs may be displayed.

[0022] When the user selects their preferred question sentence candidate C from the displayed list of one or more question sentence candidates C, the screen transitions. Then, as shown in Figure 1C, one or more predetermined words, phrases, or sentences W (hereinafter referred to as "related sentences, etc. W") that have a predetermined relationship with keyword K are displayed (step S3). Furthermore, there are no specific limitations on how "relevance" is defined in this service; service providers can define it as they see fit. The service provider then associates keywords K, candidate questions C, and related texts W, which are recognized as being related to each other, to generate a question dictionary. In other words, related texts W that are associated with keyword K in the question dictionary are treated as having "relevance" to that keyword K. The question dictionary is stored and managed in a predetermined database (for example, the question dictionary DB181 shown in Figure 5 later) in a manner that allows for searching and extraction.

[0023] Specifically, for example, as shown in Figure 1A, when the candidate question C1 "I can't connect to the company's wireless network" is selected, the FAQ screen documents related to the keyword K1 "cannot connect" are displayed as related documents W. Specifically, as shown in Figure 1C, the following FAQ screen document is displayed: "Connecting to the company's Wi-Fi. You can connect to the company's Wi-Fi network using the following ID: ·SSID:XXXX ·Password:XXXX Do not share this with anyone outside the company. If you have any connection problems, please contact the following: [Contact] Information Systems Department, Kaneko XX" and related documents are displayed. This allows users to quickly resolve internet connection issues by referring to the text displayed on the FAQ screen.

[0024] For example, as shown in Figure 1B, if the question candidate C2, "How do I connect to the company's internet?", is selected, then documents from the FAQ screen that are relevant to the keyword K2, "Internet," will be displayed as related documents W. Specifically, as shown in Figure 1C, the following text from the FAQ screen is displayed as related text, etc. W: "Connecting to the company's Wi-Fi You can connect to the company's Wi-Fi network with the following ID: ·SSID:XXXX ·Password:XXXX Do not disclose this to anyone outside the company. If you have any problems connecting, please contact the following: [Contact] Information Systems Department, Kaneko XX" This allows users to quickly resolve internet connection issues by referring to the text displayed on the FAQ screen.

[0025] In this service, when keyword K is entered into search box B, one or more candidate questions C containing keyword K in at least part are displayed. When one candidate question C is selected, the text from the FAQ screen related to keyword K is displayed as related text W. This allows the user to easily find the text from the desired FAQ screen.

[0026] There are two points worth highlighting here. First, the FAQ screen displayed as related texts W does not contain the keyword K. In other words, users can easily find the desired FAQ by simply entering a common keyword K that easily comes to mind into the search box B and selecting the displayed question candidate C.

[0027] Another noteworthy point is that even if different keywords K are entered into the search box B, it is possible to reach the desired identical related text W. In other words, multiple question candidates are associated with a single FAQ, and each of these multiple question candidates is associated with a different keyword K. In other words, if a user wants to display the text of the desired FAQ screen, all they have to do is enter a common keyword K that comes to mind and select the displayed question candidate C. That is, if the problem is "unable to connect to the internet" as in the examples in Figures 1A and 1B above, they can simply enter a common keyword K such as "cannot connect" or "internet".

[0028] Furthermore, this FAQ search site provided through this service can display the relevant parties involved in related documents W, as well as the organizations to which those parties belong. Specifically, as shown in Figure 1C, in the display area F1 below the text (related text, etc. W) on the FAQ screen, the name of the person concerned (Kaneko XX) and the organization to which that person concerned belongs (Information Systems Department) are displayed as [Contact Information]. Furthermore, the names of the individuals involved and the organizations to which they belong are all displayed as buttons that provide access to designated web pages containing detailed information. For example, when the display of an individual (e.g., Kaneko XX) is clicked (e.g., tapped), a designated web page containing detailed information (e.g., department, phone number, email address, etc.) as shown in Figure 2A (hereinafter referred to as the "details page") is displayed. Additionally, as shown in Figure 2B, a designated web page related to the details page (hereinafter referred to as the "related page") is displayed. This allows users to quickly identify the appropriate person to contact and inquire about issues that cannot be resolved solely through the FAQ screen, thus enabling them to resolve problems more quickly.

[0029] As mentioned above, this service manages a dictionary of question phrases in a designated database. As a result, when question candidate C is selected, related texts W can be displayed in the manner shown in Figure 1 above, but it is also possible to display related texts W in the manner shown in Figure 3, for example.

[0030] In other words, when keyword K is entered into search box B (step S11), one or more candidate question sentences C are displayed in the display area F11 below it (step S12). Up to this point, it is the same as the example in Figure 1 above, but below the display area F12, a list of related texts, etc. W is displayed (step S13). Specifically, a list of related texts, etc. W as shown in Figure 3 is displayed as a "word hint". The one or more related texts, etc. W displayed as "word hints" are shown in alphabetical order (A through Z) and in Japanese syllabary order. Furthermore, when no keyword K is entered in the search box B, the one or more related texts, etc. W displayed as "word hints" are listed neatly and similarly to each other without any particular differentiation. However, when keyword K is entered in the search box B, the "word hints" (related texts, etc. W) corresponding to that keyword K are highlighted. Specifically, for example, as shown in Figure 3, suppose the keyword K "address" is entered. Then, among the one or more "word hints" (related texts, etc. W) displayed in the display area F12, "purchase," "convenience store," "listing," "address," "anonymous," "change," and "return" are highlighted (shown in bold in the example in Figure 3) as "word hints" (related texts, etc. W) corresponding to "address." The user selects a desired "word hint" (related text, etc. W) from one or more highlighted "word hints" (related text, etc. W) by, for example, tapping it. Then, although not shown in the illustration, the text from the FAQ screen that the user desired will be displayed as the related text, etc. W.

[0031] This FAQ search site will be improved periodically as part of this service. As described above, in this service, keywords K, candidate questions C, and related texts W are associated and managed as a question dictionary. In addition, the user's operation history is also stored and managed as history information. Specifically, in this service, keywords K entered by the user into search box B, candidate questions C selected by the user, and related texts W reached by the user are managed as history information. The history information includes information about candidate questions C that were displayed when keyword K was entered into search box B but were not selected by the user. The history information also includes information about cases where text on the FAQ screen (related texts W) was displayed when candidate question C was selected, but it was not what the user wanted. Specifically, let's say this FAQ search site is used on an e-commerce (EC) site. And let's say many users enter the name of a specific payment method into search box B. In this case, even though one or more candidate questions C are displayed on the user terminal 2 of many users, many users may not select a candidate question C. Also, there may be cases where no candidate questions C are displayed at all. In such cases, it is possible to predict which candidate question C is likely to be desired by many users, and based on that prediction, new candidate questions C are prepared. This will allow us to improve and enhance this FAQ search site. As a result, we can increase user satisfaction with this FAQ search site. In addition, if the user does not select keywords K and related texts W included in the candidate question C that they did not select, it will be possible to identify products and services that the user desires but are not currently offered, thus contributing to the creation of new products and services that users desire.

[0032] Next, with reference to Figure 4, we will describe the configuration of an information processing system to which a server 1, according to one embodiment of the information processing device of the present invention, is applied, which enables the provision of the service described above. Figure 4 shows an example of the configuration of an information processing system to which a server according to one embodiment of the information processing device of the present invention is applied.

[0033] The information processing system shown in Figure 4 is configured to include a server 1 and a user terminal 2. Server 1 and user terminal 2 are interconnected via a predetermined network NW, such as the Internet.

[0034] Server 1 is an information processing device managed by the service provider. Server 1 performs various processes necessary to realize this service while communicating with user terminal 2 as needed.

[0035] User terminal 2 is an information processing device operated by the user. As mentioned above, user terminal 2 consists of a smartphone, tablet, personal computer, etc.

[0036] Figure 5 is a block diagram showing an example of the server hardware configuration in the information processing system shown in Figure 4.

[0037] Server 1 comprises a CPU (Central Processing Unit) 11, ROM (Read Only Memory) 12, RAM (Random Access Memory) 13, a bus 14, an input / output interface 15, an input unit 16, an output unit 17, a storage unit 18, a communication unit 19, and a drive 20.

[0038] The CPU 11 executes various processes according to the program recorded in the ROM 12 or the program loaded from the storage unit 18 into the RAM 13. RAM13 also stores data and other information necessary for the CPU11 to perform various processes.

[0039] The CPU 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output interface 15 is also connected to this bus 14. An input / output interface 15 is connected to an input unit 16, an output unit 17, a storage unit 18, a communication unit 19, and a drive 20.

[0040] The input unit 16 is configured, for example, with a keyboard, and is used to input various types of information. The output unit 17 consists of a display such as an LCD and a speaker, and outputs various information as images and sounds. The memory unit 18 is composed of DRAM (Dynamic Random Access Memory) and stores various types of data. The communication unit 19 communicates with other devices (for example, user terminal 2 in Figure 4) via a network NW, including the Internet.

[0041] A removable media 40, such as a magnetic disk, optical disk, magneto-optical disk, or semiconductor memory, is appropriately mounted in the drive 20. Programs read from the removable media 40 by the drive 20 are installed in the storage unit 18 as needed. Furthermore, the removable media 40 can store various types of data stored in the storage unit 18, just as the storage unit 18 does.

[0042] Although not shown in the diagram, user terminal 2 in Figure 4 can have a configuration that is basically the same as the hardware configuration shown in Figure 5. Therefore, the explanation of the hardware configuration of user terminal 2 will be omitted.

[0043] Through the cooperation of various hardware and software components of the information processing system shown in Figure 4, including Server 1 in Figure 5, it becomes possible to execute various processes, including question dictionary generation and related text extraction. As a result, the service provider can provide the aforementioned service to the user. "Question word dictionary generation process" refers to the process of generating the question word dictionary mentioned above. "Related text extraction processing" refers to the process of extracting related texts W using the question word dictionary mentioned above. "Question word dictionary improvement process" refers to the process of improving the question word dictionary mentioned above. The following describes the functional configuration for executing the question word dictionary generation process, related text extraction process, and question word dictionary improvement process, which are performed on Server 1 in Figure 5, which constitutes the information processing system in Figure 4.

[0044] Figure 6 is a functional block diagram showing an example of the functional configuration of the server in Figure 5 that constitutes the information processing system in Figure 4, specifically for executing the question word dictionary generation process, the related text extraction process, and the question word dictionary improvement process.

[0045] As shown in Figure 6, when Server 1 executes the question text dictionary generation process, the dictionary generation unit 101 functions in CPU 11. When Server 1 executes the related text extraction process, the keyword reception unit 102, the question text candidate extraction unit 103, the question text candidate presentation unit 104, the question text confirmation unit 105, and the related text presentation unit 106 function in CPU 11. When Server 1 executes the question text dictionary improvement process, the history management unit 107 and the dictionary improvement unit 108 function in CPU 11.

[0046] Furthermore, a question dictionary DB181 is provided in one area of ​​the storage unit 18 of server 1. The question dictionary DB181 stores the question dictionary.

[0047] The dictionary generation unit 101 stores candidate question sentences C containing a more general keyword K that is related to the entire related text W or its constituent elements, in the question sentence dictionary DB 181, associating the keyword K with the candidate question sentence C and the related text W, so that the candidate question sentence C is extracted when the keyword K is entered into the search window B. This generates a dictionary of question phrases that allows users performing FAQ searches to easily find the information they need.

[0048] The keyword reception unit 102 receives the keyword K entered into the search window B. Specifically, for example, in the example shown in Figure 1 above, when keywords K such as "cannot connect" or "internet" are entered into the search box B, the keyword reception unit 102 receives them.

[0049] The question candidate extraction unit 103 extracts one or more question candidate C from the question dictionary DB 181 that contain at least part of the keyword K entered in the search window B.

[0050] The question candidate presentation unit 104 presents the user with one or more question candidate C extracted by the question candidate extraction unit 103. Specifically, the question candidate presentation unit 104 presents one or more question candidate C extracted by the question candidate extraction unit 103 to the user by executing control to display them on the user terminal 2.

[0051] The question text confirmation unit 105 confirms the selected question text from among the one or more question text candidates C presented to the user by the question text candidate presentation unit 104, as the question text from the user.

[0052] The related text presentation unit 106 extracts related texts W associated with the candidate question C, which has been determined by the question confirmation unit 105 as a question from the user, and presents them to the user. Specifically, the related text presentation unit 106 extracts the text from the FAQ screen as related text W associated with the confirmed question candidate C, and presents it to the user by executing control to display it on the user terminal 2. As a result, the user can quickly resolve the problem by referring to the text on the displayed FAQ screen. Furthermore, the related document presentation unit 106 further extracts information about the persons concerned with the previously extracted related document W and the organizations to which those persons belong, and presents this information to the user as related document W. Furthermore, the related document presentation unit 106 presents related pages and detail pages to the user.

[0053] The history management unit 107 manages the keyword K received by the keyword reception unit 102, the question candidate C selected by the user from one or more question candidate C presented to the user by the question candidate presentation unit 104, and the related texts W presented to the user by the related texts presentation unit 106 in response to that selection, as history information. Furthermore, the history management unit 107 extracts keywords K that were received by the keyword reception unit 102 but were not selected by the user, and related texts W were not presented by the related text presentation unit 106.

[0054] The dictionary improvement unit 108 improves the question dictionary, including the correspondence between the keyword K included in the question candidate C and the question candidate C, if the question candidate is not confirmed by the question confirmation unit 105 after the question candidate is presented to the user by the question candidate presentation unit 104 as history information by the history management unit 107.

[0055] As described above, the information processing device in Figure 5 has the functional configuration shown in Figure 6, which allows users of this FAQ search site to easily find the text on their desired FAQ screen. Furthermore, by identifying products and services that users desire but are not currently offered, it can contribute to the creation of new products and services that users want.

[0056] The following further explains why users of this FAQ search site will be able to easily find the text on the desired FAQ screen, unlike conventional FAQ search sites. Figure 7 shows an example of a search on a conventional FAQ search site. In other words, the conventional approach involved the user entering a string of characters, which was then accepted. Pages (articles) containing that string and their corresponding FAQs and answers were then presented to the user as matches. The user then had to select the article they were looking for from this list. However, typically, a large number of articles containing the string (search keyword) entered by the user will be found in the search results.

[0057] As a result, there were many irrelevant answers that made it difficult to choose a suitable one. In other words, the correct answer (the article that should be presented to the user) might exist somewhere in the list, but the user themselves didn't know which article to view. Therefore, the user had to check the list from top to bottom (from the beginning).

[0058] As a result, sometimes nothing was listed. In other words, articles containing the string entered by the user were listed. Therefore, if the string entered by the user was not a term used in an article, nothing was listed, and the user had to reconsider and enter a different string.

[0059] Assuming the traditional approach of listing articles containing strings entered by the user, the following countermeasures can be considered. In other words, one approach is to prepare a large number of articles. Specifically, for example, by creating multiple versions of articles with different wording or by reframing the problem in a different way, it becomes possible to ensure that the search results are found regardless of what string of text the user enters. However, this approach has a drawback: while it may generate hits, it can also lead to irrelevant articles being included in the search results. Furthermore, for example, multiple similar articles (answers) may appear in the search results, leading to confusion. Additionally, maintainability may be compromised.

[0060] Alternatively, instead of creating a list based on whether the article contains text entered by the user, one could employ AI (Artificial Intelligence) to evaluate the relevance between the text entered by the user and the articles that should be presented to the user. If machine learning-based AI is used, the chances of finding no relevant articles will decrease. However, the problem of the list of relevant articles containing many articles that the user did not intend to read remains unresolved. Furthermore, if AI is used, articles deemed highly relevant by the AI ​​will be presented as results, but these may sometimes produce answers that pose compliance or business problems, and it is difficult to completely eliminate them. Furthermore, when AI is adopted, it is vulnerable to changes in the service or business environment. In other words, because AI learns from past search results, it becomes difficult to respond quickly when there is an increase in specific inquiries, or when the service, business environment, or the service itself changes. Furthermore, if AI is adopted, a large amount of data will be required for individual tuning of each search target (e.g., service). Cleansing this large amount of data will also require significant engineering resources, leading to high costs.

[0061] Thus, conventional approaches and methods using AI, etc., are nothing more than evaluating the relationship between the string entered by the user and the article (including simply evaluating whether or not it is included), and have the aforementioned problems. In contrast, the following explains whether users of this FAQ search site will be able to easily find the desired FAQ and its answer, unlike the conventional FAQ search sites mentioned above.

[0062] Figure 8 illustrates the concept of intent deployment in this service. In the example in Figure 8, the target text T1 that should be presented to the user is an article containing an FAQ titled "Information on how to get a refund" and its answers. Here, an article (or purpose statement, etc.) titled "Information on Refund Methods" should be presented to users of online shopping sites, etc., who are experiencing problems due to receiving a defective product, receiving a product different from what they ordered, or who, for any reason, want their money back.

[0063] Therefore, in this service, the technical writer expands the target text, etc. T1, into an expression of intent (intermediate) T2 (in the example in Figure 8, it is expanded by three times). In other words, the technical writer develops the expression to take into account the user's intentions. Specifically, in Figure 8, for example, the technical writer develops the expression assuming a user who intends to get a response by conveying the fact that "there is a defect in the product." For example, the technical writer develops the expression assuming a user who intends to get a response by conveying the fact that "the wrong product was delivered." For example, the technical writer develops the expression assuming a user who intends to get a response by conveying the desire (solution) of "I want a refund." In this way, technical writers develop the article from the target text (T1) to the intent expression (intermediate) (T2), considering the target user's intentions for which the article should be presented.

[0064] Furthermore, Server 1 further expands the intent expression (intermediate) T2. Specifically, Server 1 expands the intent expression (intermediate) T2 into intent expression T3 using a predetermined algorithm (expanded 16 times in the example in Figure 8). Specifically, as shown in Figure 8, for example, the intended expression (intermediate) T2, "There is a problem," is expanded to "The product is broken," "The product was defective," etc. That is, for example, Server 1 expands the expression by adding a subject, replacing words with synonyms, or changing combinations of sentence endings and auxiliary verbs to create a sentence with the same meaning. The generated (expanded) intent expression T3 is stored in the aforementioned question dictionary DB181 as a candidate question C. Similarly, each of the multiple objective sentences T1 goes through the stages of intent expression (intermediate) T2 and intent expression T3, and the resulting expanded versions are stored in the aforementioned question dictionary DB181 as candidate question sentences C.

[0065] As a result, users will be able to perform the following types of searches: Figure 9 shows an example of user guidance achieved through the intended deployment of this service. In other words, as shown in Figure 9, for example, users can perform searches by selecting an intention such as "fuguai" (faulty), "wrong," or "refund," and entering it as a word (keyword K) that comes to mind. In other words, multiple questions (expressions of intent) are generated for a given article (target text, etc. T1), allowing users to arrive at that article through various phrasing. Furthermore, as shown in Figures 8 and 9, this service presents the user with appropriate question candidates C even if the keyword K entered by the user is different (for example, if it is different from "defective product" and "fuguai", "different" and "different", or "give me my money back" and "refund"). This will be explained later using Figures 11 and 12.

[0066] The above explains one of the differences between this service and conventional FAQ search sites: intent development. The following describes other features of this service that improve user convenience.

[0067] Figure 10 shows an example of user guidance achieved through intent prediction in this service. As shown in Figure 10, the server 1 of this service has a function that predicts the user's intent based on just one character entered by the user. This allows the user to check the candidate question sentence C without having to enter the entire question sentence (keyword K). In other words, this service has a predictive function that predicts what keyword K the user will ultimately intend to enter, based on the characters entered at that point while the user is typing keyword K. The prediction function is adjusted for each service that this service searches (for example, services targeted by the FAQ search site). That is, for example, in the case of an FAQ search site for a service that uses the term "points," if a user enters "po," it will be predicted that the keyword K "points" will be entered. For example, in the case of an FAQ search site for a service that uses the term "portfolio," if a user types "po," it is predicted that the keyword K "portfolio" will be entered.

[0068] Even when the prediction function is active, as mentioned above, the user is presented with candidate question sentences C that include the keyword K. In other words, normally, when a user enters the keyword K "points," even a natural person cannot understand the user's intentions regarding what they want to do with the points. However, with this service, the user is presented with candidate question sentences C that include the keyword K, so the user can always arrive at an answer by selecting the candidate question sentence C that matches their intention.

[0069] The following provides further explanation regarding the feature where the user is presented with candidate question sentences C that include the keyword K entered by the user. Figure 11 shows an example of the suggested question wording presented in this service. As shown in Figure 11, when a keyword K is entered by the user, candidate question sentences C containing that keyword K are presented to the user. The user can then confirm the target text T1 by selecting the candidate question sentence C that best suits their intention from the presented candidates. Server 1 associates definitions, explanations, synonyms, paraphrases, etc., of terms and stores and manages them as a definition database. Specifically, for example, in the example in Figure 11, "automatic braking system" and "ASV" are associated and stored in the definition database. Then, when Server 1 searches for candidate question sentences C based on the keyword K entered by the user, it uses the definition database to search for candidate question sentences C and presents them to the user. This results in a hit for intent expansion T3, "What is an automatic braking system?" In this case, if the user searches using the keyword K "ASV", the user will be presented with a candidate question C in which "automatic braking system" in intent development T3, "What is an automatic braking system?", has been replaced with the word "ASV". As a result, the user is presented with candidate question C that is independent of the user's search level, because the user's words are "always" included in the question. In other words, the system (server 1) can present candidate question C according to the user's level of vocabulary. From the user's perspective, this makes it possible for them to search using their own words and makes it easier to choose from multiple candidate question options. In other words, by simply entering keyword K, the user is presented with candidate question sentences C expressed using the keyword K they entered, making it easy to select from multiple candidate question sentences.

[0070] Figure 12 shows an example of how a user selects a candidate question in this service. As shown in Figure 12, the user is presented with candidate question sentences C that include the keyword K entered by the user. As a result, as explained using Figure 11, users can search using their own words. The user then selects a candidate question C that contains the keyword K they entered, and as a result, reaches the target text T1 corresponding to that question. The above further explains why users of this FAQ search site, unlike conventional FAQ search sites, can easily find the text on the desired FAQ screen.

[0071] As explained above, the target text T1 is described as an FAQ and its answer text (article). However, as stated above, this service can be applied to specific words, phrases, or sentences presented to the user in order to achieve the user's predetermined objective.

[0072] In other words, for example, the objective text T1 may be a description of the product (service). Specifically, for example, suppose a user has the intention of using a financial service, specifically expressing concern about their child's education expenses. In this case, the user enters a keyword such as "child" into the input field displayed on the financial service website along with a message like "What would you like to ask?". As a result, the user is presented with candidate question C such as, "I want to prepare funds for my child's education," and "Up to what age is [service name] for children?" Here, "[service name]" is the name of a service for preparing funds for a child's education, in other words, a so-called educational insurance service. Next, by selecting the candidate question C, "I want to prepare funds for my child's education," which the user had not yet verbalized and entered as keyword K, they will be able to view the product description page for the educational insurance product "〇〇".

[0073] Furthermore, the aforementioned keyword K and candidate question C will also function appropriately for users who intend to use the XX service and whose intention is to "find out to what age range of children are eligible." Furthermore, as mentioned above, if product descriptions are used in the purpose statement (T1) of this service, it is possible to have it function as a dedicated product description page rather than as part of the FAQ search site. Thus, this service can be applied not only to FAQs, but also to specific words, phrases, or sentences presented to users in order to help them achieve their intended objectives. Furthermore, in the above-described embodiment, it was assumed that the user would input keyword K in the form of a question to express their intention. However, what is presented to the user is not limited to a question; it is sufficient if it is a guiding statement that the user desires (intends to) express the user's intention.

[0074] The functional configuration of Server 1, which presents candidate guidance sentences and target sentences (T1), will be explained using Figure 13. Figure 13 is a functional block diagram showing an example of the functional configuration of the server in Figure 5, specifically the configuration for executing the process of allowing the user to search for and present target documents, etc.

[0075] As shown in Figure 6, when Server 1 executes the guidance text dictionary generation process, the dictionary generation unit 111 functions in CPU 11. When Server 1 executes the target text extraction process, the keyword reception unit 112, the guidance text candidate extraction unit 113, the guidance text candidate presentation unit 114, the guidance text selection reception unit 115, and the target text presentation unit 116 function in CPU 11. When Server 1 executes the guidance text dictionary improvement process, the history management unit 117 and the dictionary improvement unit 118 function in CPU 11.

[0076] Furthermore, a guidance message dictionary DB182 is provided in one area of ​​the storage unit 18 of server 1. The guidance message dictionary is stored in the guidance message dictionary DB182.

[0077] The dictionary generation unit 111 stores candidate guide sentences C containing a more general keyword K that is related to the entirety or components of the target text T1 in the guide sentence dictionary DB 182, associating keyword K, candidate guide sentence C, and the target text T1 so that they can be extracted when keyword K is entered into the search window B. This generates a dictionary of guiding sentences that allows users performing searches on target texts, etc. (T1), to easily find the desired information.

[0078] The keyword reception unit 112 receives the keyword K entered into the search window B. Specifically, for example, in the case of the aforementioned educational insurance, when keywords K such as "child" and "educational funds" are entered into the search box B, the keyword reception unit 112 receives them.

[0079] The guidance sentence candidate extraction unit 113 extracts one or more guidance sentence candidates C from the guidance sentence dictionary DB 181 that contain at least part of the keyword K entered in the search window B.

[0080] The guidance sentence candidate presentation unit 114 presents the user with one or more guidance sentence candidates C extracted by the guidance sentence candidate extraction unit 113. Specifically, the guidance sentence candidate presentation unit 114 presents one or more guidance sentence candidates C extracted by the guidance sentence candidate extraction unit 113 to the user by executing control to display them on the user terminal 2.

[0081] When one guidance sentence candidate C is selected from among the one or more guidance sentence candidates C presented to the user by the guidance sentence candidate presentation unit 114, the guidance sentence selection receiving unit 115 confirms this as the guidance sentence from the user.

[0082] The target text presentation unit 116 extracts the target text T1, which is associated with the candidate guidance text C determined by the guidance text selection reception unit 115 as a guidance text from the user, and presents it to the user. Specifically, the target text presentation unit 116 extracts the text of the FAQ screen as the target text T1 associated with the confirmed guide sentence candidate C, and presents it to the user by executing control to display it on the user terminal 2. As a result, the user can quickly resolve the problem by referring to the text of the displayed FAQ screen. Furthermore, the target document presentation unit 116 further extracts information about the parties involved in the previously extracted target document T1 and the organizations to which those parties belong, and presents this information to the user as target document T1.

[0083] The history management unit 117 manages the keyword K received by the keyword reception unit 112, the guided sentence candidate C selected by the user from among one or more guided sentence candidates C presented to the user by the guided sentence candidate presentation unit 114, and the target text T1 presented to the user by the target text presentation unit 116 in response to that selection, as history information. Furthermore, the history management unit 117 extracts keywords K that were received by the keyword reception unit 112 but were not selected by the user, and therefore were not presented by the target text presentation unit 116.

[0084] The dictionary improvement unit 118 improves the guidance sentence dictionary, including the correspondence between keyword K included in guidance sentence candidate C and guidance sentence candidate C, if guidance sentence candidate C has been presented to the user by the guidance sentence candidate presentation unit 114 as history information by the history management unit 117, but has not been confirmed by the guidance sentence selection reception unit 115.

[0085] This allows the user to select a candidate guide sentence C that corresponds to their intention from among the candidate guide sentences C that contain the keyword K they entered. As a result, the user is presented with an appropriate target sentence T1 that reflects their intention.

[0086] The above describes an example of the functional configuration of Server 1, in which this service is applied not only to FAQs but also to purpose-driven documents including product descriptions. The following summarizes the features of this service, which is implemented by Server 1 as described above.

[0087] The search function using intent development in this service can be considered a search method that is well-suited to the behavioral characteristics of users. Specifically, users are often unfamiliar with the products (services) they are searching for. Therefore, they often do not understand the technical terms typically used in FAQs and product descriptions. In other words, they cannot recall keywords for searching that utilize such technical terms. Furthermore, even if they do enter keywords, it is difficult for them to select the correct article (one that matches their intent) from a list of articles that simply contain those keywords. Furthermore, the reality is that many users give up if they cannot reach their intended target text (T1) after several attempts (an average of about three). As mentioned above, this service can solve these problems.

[0088] Furthermore, the search function using intent development in this service can be described as a mechanism that searches for predicted leading sentences (questions) rather than searching for the purpose or other texts T1 that would serve as answers based on the keywords K entered by the user. Specifically, even if users are presented with the answers they are looking for (Target text, etc. T1), it is difficult for them to choose the appropriate one. In this service, the user is presented with the question they have thought of themselves (leading sentence), making it easier for them to make a choice. This will improve the user's rate of achieving the correct answer.

[0089] Although one embodiment of the present invention has been described above, the present invention is not limited to the embodiments described above, and any modifications, improvements, etc. that can achieve the objectives of the present invention are considered to be included in the present invention.

[0090] For example, in the embodiment described above, only one text is displayed on the FAQ screen as related text, etc. W, but two or more texts may be displayed.

[0091] Furthermore, the examples are not limited to those mentioned above. For example, suppose that incorrect payment processing occurred for three users who used online payment. In this case, each of the three users enters the keywords K, such as "malfunction," "give me my money back," and "defective," into the search box B of this FAQ site, and each selects a suggested question C that is displayed. Then, the same FAQ screen text that each of the three users desires is presented to each of the three users.

[0092] Furthermore, in the above embodiment, for example, when a word is entered, candidate question sentences containing a word related to a predetermined word, phrase, or sentence that is more common are extracted. However, the system is not limited to this. When a word is entered, candidate question sentences containing a word related to a predetermined word, phrase, or sentence that is less common or of similar commonness are also extracted. Specifically, for example, when the word "delete" is entered into the search box, sentences containing the word "delete" may be extracted.

[0093] Furthermore, the system configuration shown in Figure 4 and the hardware configuration of Server 1 shown in Figure 5 are merely illustrative examples for achieving the objectives of the present invention and are not particularly limited.

[0094] Furthermore, the functional block diagram shown in Figure 6 is merely illustrative and not particularly limiting. In other words, it is sufficient if the information processing system in Figure 4 has the functionality to execute the above-mentioned question dictionary generation process and related text extraction process as a whole, and the functional blocks and databases used to realize this functionality are not particularly limited to the example in Figure 5.

[0095] Furthermore, the location of the functional blocks and database is not limited to Figure 6, but can be any location. In the example in Figure 6, the question dictionary generation process, related text extraction process, and question dictionary improvement process are controlled by the CPU 11 of Server 1 in Figure 5, which constitutes the information processing system in Figure 4, but the system is not limited to this configuration. For example, at least a portion of the functional blocks and database located on Server 1 may be provided on User Terminal 2 or another information processing device (not shown).

[0096] Furthermore, the series of processes described above can be executed by hardware or by software. Furthermore, a single functional block may consist of hardware alone, software alone, or a combination of both.

[0097] When a series of processes are executed by software, the programs that make up that software are installed on a computer or other device from a network or storage medium. The computer may be a computer that is built into dedicated hardware. Furthermore, a computer can be any computer capable of performing various functions by installing various programs, such as a server, a general-purpose smartphone, or a personal computer.

[0098] Such recording media containing programs may consist not only of removable media (not shown) distributed separately from the main unit of the device to provide the program to the user, but also of recording media provided to the user in a state where they are pre-installed in the main unit of the device.

[0099] In this specification, the step of describing a program to be recorded on a recording medium includes not only processes that are performed chronologically in that order, but also processes that are not necessarily performed chronologically, but are executed in parallel or individually.

[0100] In summary, the information processing device to which the present invention applies only needs to have the following configuration, and can take various forms. In other words, the information processing apparatus to which the present invention is applied is: The system includes a dictionary generation means (for example, the dictionary generation unit 101 in Figure 6) that stores in a question dictionary a question sentence candidate (for example, the question sentence candidate C in Figure 1, for example, "The product was defective") which contains a more general word related to a predetermined word, phrase, or sentence (for example, the entire text or its components in the FAQ screen as related text W in Figure 1, specifically, for example, "Information on how to get a refund" in Figure 7), by associating the word, the question sentence candidate, and information containing the predetermined word, phrase, or sentence.

[0101] In other words, when a word is entered, the word, the candidate question, and the information containing the predetermined word, phrase, or sentence are associated and stored as a question dictionary, so that a more common candidate question containing that word related to the predetermined word, phrase, or sentence is extracted. This generates a dictionary of question phrases that allows users performing FAQ searches to easily find the information they need.

[0102] Furthermore, the information processing apparatus to which the present invention is applied is: The system includes a dictionary generation means that stores in a question dictionary a word that is related to a predetermined word, phrase, or sentence, and contains a word that is less common or of similar commonness, so that when the word is entered, the candidate question sentence is extracted, associating the word, the candidate question sentence, and information containing the predetermined word, phrase, or sentence.

[0103] In other words, when a word is entered, the word, the candidate question, and the information containing the predetermined word, phrase, or sentence are associated and stored as a question dictionary, so that a candidate question containing a word related to the predetermined word, phrase, or sentence, which is less common or of similar commonness, is extracted. This generates a dictionary of question phrases that allows users performing FAQ searches to easily find the information they need.

[0104] Furthermore, the information processing apparatus to which the present invention is applied is: A receiving means (for example, the keyword receiving unit 102 in Figure 5) that receives a predetermined word, phrase, or word related to a sentence (for example, keyword K mentioned above) entered by the user, Extraction means (for example, the question sentence candidate extraction unit 103 in Figure 5) for extracting one or more question sentence candidates that contain at least part of the aforementioned word, A first presentation means (for example, the question candidate presentation unit 104 in Figure 5) presents the one or more question candidate sentences extracted by the extraction means to the user, A second presentation means (for example, the related text presentation unit 106 in Figure 5) extracts a predetermined word, phrase, or sentence corresponding to the question sentence candidate selected by the user from among the one or more question sentence candidates presented to the user by the first presentation means, and presents it to the user. Any information processing device equipped with this feature will suffice.

[0105] In other words, when a word, phrase, or sentence related to a predetermined word entered by the user is received, one or more candidate question sentences containing at least part of that word are extracted. The extracted one or more candidate question sentences are presented to the user. From the one or more candidate question sentences presented to the user, the predetermined word, phrase, or sentence corresponding to the candidate question sentence selected by the user is extracted and presented to the user. This will allow users performing FAQ searches to easily find the information they are looking for.

[0106] Furthermore, the second presentation means is Further information regarding the persons concerned with the previously extracted predetermined word, phrase, or sentence can be extracted and presented to the user.

[0107] In other words, information about the individuals involved with the previously extracted predetermined words, phrases, or sentences, and the organizations to which those individuals belong, is extracted and presented to the user. This will make it easier for users performing FAQ searches to find the information they are looking for.

[0108] Furthermore, the second presentation means is Further information regarding the organization to which the person concerned belongs, relating to the previously extracted predetermined word, phrase, or sentence, can be extracted and presented to the user.

[0109] In other words, information about the organizations to which the individuals related to the previously extracted predetermined words, phrases, or sentences belong is extracted and presented to the user. This will make it easier for users performing FAQ searches to find the information they are looking for.

[0110] Furthermore, the system includes a management means (for example, a history management unit 107 in Figure 5) that manages as history information a word related to the predetermined word, phrase, or sentence received by the reception means, a question sentence candidate selected by the user from among the one or more question sentence candidates presented to the user by the first presentation means, and the predetermined word, phrase, or sentence presented to the user by the second presentation means in response to that selection. The aforementioned management means further, Words that have been received by the reception means but were not selected by the user, or words that were not presented by the second presentation means, can be extracted.

[0111] In other words, the words related to the accepted predetermined word, phrase, or sentence, the question sentence candidate selected by the user from one or more question sentence candidates presented to the user, and the predetermined word, phrase, or sentence presented to the user as a result of that selection are managed as historical information. Then, words that were accepted but not selected by the user, or words for which no predetermined word, phrase, or sentence was presented, are extracted. This allows us to extract information to improve and enhance this FAQ search site. Furthermore, it identifies products and services that users desire but are not currently offered, thus contributing to the creation of new products and services that users want.

[0112] Furthermore, the information processing apparatus to which the present invention is applied is: A dictionary generation means (for example, the dictionary generation unit 111 in Figure 13) stores in a guided sentence dictionary a predetermined word, phrase, or sentence presented to the user in order to achieve a predetermined purpose, as the target sentence, etc., and a guided sentence candidate that the user intends to reach the target sentence, etc., which includes a more general word related to the target sentence, etc. (corresponding to the thought-of word (keyword K) in Figure 7) or a similar word, so that when the word is entered, the guided sentence candidate is extracted, associating the word, the guided sentence candidate, and the target sentence, etc. Having that will suffice.

[0113] Furthermore, a dictionary generation means stores in a guided sentence dictionary a predetermined word, phrase, or sentence presented to the user to achieve a predetermined objective, as the target sentence, etc., and a guided sentence candidate that the user intends to reach the target sentence, etc., which is related to the target sentence, etc., but is less common or of similar commonness (the word that came to mind in Figure 7) or a similar word, in association with the word, the guided sentence candidate, and the target sentence, etc., so that the guided sentence candidate is extracted when the word is entered. It can be equipped with.

[0114] Furthermore, the candidate inductive sentences can be candidates for questions, and the dictionary of inductive sentences can be a dictionary of questions.

[0115] Furthermore, the information processing apparatus to which the present invention is applied is: A receiving means (for example, the keyword receiving unit 112 in Figure 13) that receives words related to a predetermined word, phrase, or sentence, which is presented to the user in order to achieve a predetermined purpose, as the purpose sentence, etc., Extraction means (for example, the guidance sentence candidate extraction unit 113 in Figure 13) extracts one or more guidance sentence candidates that contain at least part of the word or a similar word from a guidance sentence dictionary in which multiple guidance sentence candidates intended for the user to reach the target text are pre-registered and associated with the target text, etc. A first presentation means (for example, the presentation unit 114 in Figure 13) presents to the user the one or more guide sentence candidates extracted by the extraction means or guide sentence candidates processed based thereon (for example, a guide sentence candidate "The product was defective" extracted using the keyword "bad product" as a casual expression for a defective product, which is then processed to "The product was a bad product"), A second presentation means (for example, the target text presentation unit 116 in Figure 13) extracts from the guidance text dictionary the target text, etc., associated with the guidance text candidate itself or its source, selected by the user from among the one or more guidance text candidates presented to the user by the first presentation means, and presents it to the user. Having these will suffice.

[0116] Furthermore, the aforementioned candidate inductive sentences are candidate question sentences, and the aforementioned dictionary of inductive sentences is a dictionary of question sentences. It can be done this way.

[0117] The second presentation means is, Further information regarding the parties involved in the aforementioned target document, etc., which was extracted earlier, is then extracted and presented to the user. It is possible.

[0118] Furthermore, the second presentation means is Further extract information about the organizations to which the individuals involved in the aforementioned target documents belong, and present it to the user. It is possible.

[0119] The system further includes a management means (for example, a history management unit 117 in Figure 13) that manages the word received by the reception means, the question sentence candidate selected by the user from among the one or more question sentence candidates presented to the user by the first presentation means, and the predetermined purpose sentence etc. presented to the user by the second presentation means in response to the selection, as history information. The aforementioned management means further, It is possible to extract words that have been received by the reception means but were not selected by the user as candidates for the guidance sentence, or words for which the target sentence, etc., was not presented by the second presentation means. [Explanation of symbols]

[0120] 1...Server, 2...User terminal, 11...CPU, 12...ROM, 13...RAM, 14...Bus, 15...Input / Output interface, 16...Input unit, 17...Output unit, 18...Storage unit, 19...Communication unit, 20...Drive, 40...Removable media, 101...Dictionary generation unit, 102...Keyword reception unit, 103...Question sentence candidate extraction unit, 104...Question sentence candidate presentation unit, 105...Question sentence confirmation unit, 106...Related text presentation Department, 107...History Management Department, 108...Dictionary Improvement Department, 181...Question Text Dictionary DB, 111...Dictionary Generation Department, 112...Keyword etc. Reception Department, 113...Guidance Text Candidate Extraction Department, 114...Guidance Text Candidate Presentation Department, 105...Guidance Text Selection Reception Department, 116...Target Text etc. Presentation Department, 117...History Management Department, 118...Dictionary Improvement Department, 182...Question Text Dictionary DB, B...Search Window, K...Keyword, C...Question Text Candidate, W...Related Text etc., NW...Network

Claims

1. A dictionary generating means for storing in a guidance sentence dictionary a word, a plurality of types of guidance sentence candidates, and a target string, which is presented to a user in order to achieve the user's predetermined purpose, in association with the word, a plurality of types of guidance sentence candidates, and the target string, so that a guidance sentence candidate intended by the user to reach the target string is extracted when the word is input. An information processing device comprising:

2. The guidance sentence candidate is a question sentence candidate, and the guidance sentence dictionary is a question sentence dictionary. The information processing device according to claim 1 .

3. An information processing method executed by an information processing device, a dictionary generating step of storing a target string, which is a character string consisting of a predetermined word, phrase, or sentence presented to a user in order to achieve a predetermined purpose of the user, in a guidance sentence dictionary by associating the word, a plurality of types of guidance sentence candidates, and the target string so that a guidance sentence candidate intended by the user to reach the target string is extracted when the word is input; An information processing method comprising:

4. The guidance sentence candidate is a question sentence candidate, and the guidance sentence dictionary is a question sentence dictionary. The information processing method according to claim 3.

5. On the computer, a dictionary generating step of storing a target string, which is a character string consisting of a predetermined word, phrase, or sentence presented to a user in order to achieve a predetermined purpose of the user, in a guidance sentence dictionary by associating the word, a plurality of types of guidance sentence candidates, and the target string so that a guidance sentence candidate intended by the user to reach the target string is extracted when the word is input; A program that executes control processing including:

6. The guidance sentence candidate is a question sentence candidate, and the guidance sentence dictionary is a question sentence dictionary. The program according to claim 5.

Images