JP7759987B2 - Information processing device, information processing method, and information processing program - Google Patents

Description

The present invention relates to an information processing device, an information processing method, and an information processing program.

Technologies related to prompt engineering are known. For example, there is a technology that generates prompts using template information for creating prompts to be input into a generation AI and instructions entered by a user using a terminal device.

Patent No. 7370118

However, the above-mentioned conventional technology simply generates prompts using template information for creating prompts to be input into the generation AI and instructions entered by the user using a terminal device, and therefore does not necessarily provide support for selection at each level when the user selects data contained in the lowest level from among data in a hierarchical structure.

The purpose of this application is to support selection at each level, allowing users to select data contained in the lowest level from hierarchically structured data.

The information processing device according to the present application includes: a reception unit that receives answer information indicating a user's answer to a question for identifying desired data included in hierarchically structured data including a data hierarchy including, as elements, each of a plurality of data items including desired data desired by a user; a category hierarchy including, as elements, each of a plurality of categories into which the plurality of data items are classified; and a higher category hierarchy including, as elements, each of a plurality of higher categories into which the plurality of categories are classified; an acquisition unit that acquires hierarchy information regarding a plurality of elements included in one hierarchy in the hierarchical structure; a generation unit that inputs the answer information and a prompt including the hierarchy information into a generative model and causes the generative model to generate candidate information regarding candidate elements among the plurality of elements that are candidates for elements corresponding to the answer information; and an output control unit that outputs the candidate information regarding the candidate elements.

According to one aspect of the embodiment, it is possible to assist the user in making selections at each level to select data included in the lowest level from among data in a hierarchical structure.

FIG. 1 is a diagram for explaining the review of the functional layout common to a plurality of systems. FIG. 2 is a diagram showing an outline of the data structure of TMF information according to the embodiment. FIG. 3 is a diagram illustrating an example of the first layer of TMF information according to the embodiment. FIG. 4 is a diagram illustrating an example of the second layer of TMF information according to the embodiment. FIG. 5 is a diagram illustrating an example of the third layer of TMF information according to the embodiment. FIG. 6 is a diagram illustrating an example of the fourth layer of TMF information according to the embodiment. FIG. 7 is a diagram illustrating an example of the fifth layer of TMF information according to the embodiment. FIG. 8 is a diagram illustrating an example of the configuration of an information processing system according to the embodiment. FIG. 9 is a diagram illustrating an example of the configuration of an information processing apparatus according to the embodiment. FIG. 10 is a diagram illustrating an example of information stored in the TMF information storage unit according to the embodiment. FIG. 11 is a diagram illustrating an example of information stored in the TMF information storage unit according to the embodiment. FIG. 12 is a diagram illustrating an example of information stored in the TMF information storage unit according to the embodiment. FIG. 13 is a diagram illustrating an example of information stored in the TMF information storage unit according to the embodiment. FIG. 14 is a diagram illustrating an example of information stored in the TMF information storage unit according to the embodiment. FIG. 15 is a diagram illustrating an example of a screen of the terminal device according to the embodiment. FIG. 16 is a diagram illustrating an example of a prompt according to the embodiment. FIG. 17 is a diagram illustrating an example of a prompt according to the embodiment. FIG. 18 is a diagram illustrating an example of a screen of the terminal device according to the embodiment. FIG. 19 is a diagram illustrating an example of a screen of the terminal device according to the embodiment. FIG. 20 is a diagram illustrating an example of a prompt according to the embodiment. FIG. 21 is a diagram illustrating an example of a screen of the terminal device according to the embodiment. FIG. 22 is a diagram illustrating an example of a screen of the terminal device according to the embodiment. FIG. 23 is a diagram illustrating an example of a prompt according to the embodiment. FIG. 24 is a diagram illustrating an example of a screen of the terminal device according to the embodiment. FIG. 25 is a diagram illustrating an example of a prompt according to the embodiment. FIG. 26 is a diagram illustrating an example of a screen of the terminal device according to the embodiment. FIG. 27 is a flowchart showing the procedure of information processing by the information processing device according to the embodiment. FIG. 28 is a hardware configuration diagram illustrating an example of a computer that realizes the functions of the information processing device.

Below, detailed explanations will be given of the information processing device, information processing method, and information processing program (hereinafter referred to as "embodiments") according to the present application, with reference to the drawings. Note that the information processing device, information processing method, and information processing program according to the present application are not limited to these embodiments. Furthermore, identical components in the following embodiments will be designated by the same reference numerals, and duplicate explanations will be omitted.

(Embodiment)
1. Introduction
Conventionally, overlapping functions have been developed across multiple different systems. Figure 1 is a diagram illustrating the review of the allocation of functions common to multiple systems. In Figure 1, the XX system and the ◇◇ system each have an application registration function and an email transmission function. In other words, the XX system and the ◇◇ system share the application registration function and the email transmission function. Therefore, efforts are being made to consolidate the functions common to the XX system and the ◇◇ system into a single function. For example, efforts are being made to consolidate functions common to multiple systems into functions defined by TM Forum, a standardization organization in the telecommunications industry. Functions defined by TM Forum are called "functions." Hereinafter, information about functions defined by TM Forum will be referred to as "TMF information." For example, when developing a new system or adding a new function to an existing system, users investigate whether the corresponding function (i.e., function) is available in the TMF information.

Here, each function (i.e., function) in the TMF information is classified into multiple categories. For example, each function in the TMF information is classified into a major category, a medium category within each major category, and a small category within each medium category. First, the user selects a major category from among the multiple major categories to which the desired function is likely to be classified. Next, the user selects a medium category from among the multiple medium categories within the selected major category to which the desired function is likely to be classified. Next, the user selects a small category from among the multiple small categories within the selected medium category to which the desired function is likely to be classified. In other words, TMF information is hierarchically structured data that includes a higher-level category hierarchy containing multiple major categories, a category hierarchy containing multiple medium categories, and a data hierarchy containing multiple small categories (e.g., individual functions). Therefore, the user must traverse the multiple levels in the hierarchical structure, starting from the highest level, to finally reach the desired function. In this way, in the past, users had to become familiar with the contents of the hierarchical structure and then manually search from the highest level to the lowest level to find the desired function. Furthermore, in the past, users had to manually make each of the multiple selections required to arrive at the desired function. Hereinafter, data indicating the function desired by the user may be referred to as "desired data." As a result, the functions selected by each user may differ (i.e., personal). Furthermore, because TMF information contains a huge amount of information, it takes a great deal of effort for users to understand the TMF information and arrive at the desired data.

Figure 2 is a diagram showing an overview of the data structure of TMF information according to an embodiment. To facilitate understanding, Figure 2 simplifies the structure of the TMF information and illustrates a case in which the TMF information is hierarchically structured data including three levels. In Figure 2, the TMF information is denoted as DT1. As shown in Figure 2, the TMF information is hierarchically structured data including a data level H3, a category level H2, and an upper category level H1. Data level H3 is a level that includes, as elements, multiple pieces of data, including desired data desired by the user. In Figure 2, each of the multiple pieces of data included in data level H3 is denoted as D1 to D8. Furthermore, the desired data desired by the user is denoted as D5. Furthermore, category level H2 is a level that includes, as elements, multiple categories that classify multiple pieces of data. In Figure 2, each of the multiple categories included in category level H2 is denoted as C1 to C4. Category C1 is a category that classifies data D1 and D2. Category C2 is a category that classifies data D3 and D4. Category C3 is a category that classifies data D5 and D6. Category C4 is a category that classifies data D7 and D8. Furthermore, upper category hierarchy H1 is a hierarchy that includes, as elements, multiple upper categories that classify multiple categories. In Figure 2, each of the multiple upper categories included in upper category hierarchy H1 is indicated by UC1 and UC2. UC1 is a higher category that classifies categories C1 and C2. UC2 is a higher category that classifies categories C3 and C4.

As shown in Figure 2, the desired data D5 desired by the user from among the TMF information is included in data hierarchy H3, which is the lowest hierarchy in the hierarchical structure. Therefore, the user cannot reach the desired data D5 unless they select the appropriate element from among the multiple elements included in each hierarchy, starting from the top-level upper category hierarchy H1. Conventionally, each user manually made multiple selections to reach the desired data. As a result, the elements selected by each user may differ, and the desired data that each user ultimately arrives at varies. Furthermore, because TMF information contains a huge amount of information, it takes a great deal of effort for users to understand the TMF information. Therefore, it takes a great deal of effort for users to understand the TMF information and then reach the desired data.

In response to this, the information processing device 100 according to the embodiment extracts candidate elements for the user to select for each level and presents them to the user so that the user can select data included in the lowest level from among hierarchically structured data. This allows the information processing device 100 to assist the user in making selections at each level to select data included in the lowest level from among hierarchically structured data. Furthermore, by extracting candidate elements for the user to select for each level and presenting them to the user, the information processing device 100 can reduce the possibility that elements selected at each level will differ significantly between users. Furthermore, because the information processing device 100 can reduce the possibility that elements selected at each level will differ significantly between users, it can also reduce the variance in the data ultimately selected by users. Furthermore, the information processing device 100 can reduce the effort required for users to understand TMF information and arrive at desired data.

Specifically, the information processing device 100 first receives answer information indicating the user's answer to a question included in the TMF information to identify the desired data. Furthermore, upon receiving the answer information, the information processing device 100 acquires first hierarchical information related to the upper category hierarchy H1, which is the highest hierarchy in the hierarchical structure. Specifically, the information processing device 100 acquires, as the first hierarchical information, information related to the upper categories UC1 and UC2 included in the upper category hierarchy H1. Furthermore, based on the answer information and the first hierarchical information, the information processing device 100 selects the upper category UC2, which is the upper category corresponding to the answer information, from the upper categories UC1 and UC2 included in the upper category hierarchy H1.

Furthermore, when the information processing device 100 selects the upper category UC2, it acquires second hierarchical information related to category hierarchy H2, which is one hierarchy below the upper category hierarchy H1. Specifically, the information processing device 100 acquires, as the second hierarchical information, information related to categories C3 and C4, which are multiple child elements of the selected upper category UC2 as a parent element. Furthermore, based on the answer information and the second hierarchical information, the information processing device 100 generates candidate information related to candidate elements among categories C3 and C4, which are candidates for elements corresponding to the answer information. For example, the information processing device 100 outputs, as candidate information, the probability (e.g., 90%) that category C3 is the element corresponding to the answer information and the probability (e.g., 10%) that category C4 is the element corresponding to the answer information. Furthermore, the information processing device 100 outputs the candidate information and accepts selection information related to a selected element, which is a candidate element selected by the user from the two candidate elements. In FIG. 2, the information processing device 100 accepts selection information related to category C3, which is the selected element selected by the user.

Furthermore, when the information processing device 100 receives the selection information, it acquires third hierarchical information related to the data hierarchy H3, which is one hierarchy below the category hierarchy H2. Specifically, the information processing device 100 acquires, as the third hierarchical information, information related to data D5 and D6, which are child elements of category C3, the selection element, as the parent element. Furthermore, based on the answer information and the third hierarchical information, the information processing device 100 generates candidate information related to candidate elements among data D5 and D6, which are candidates for the element corresponding to the answer information. For example, the information processing device 100 outputs, as candidate information, the probability (e.g., 80%) that data D5 is the element corresponding to the answer information and the probability (e.g., 20%) that data D6 is the element corresponding to the answer information. Furthermore, the information processing device 100 outputs the candidate information and receives selection information related to the selection element, which is the candidate element selected by the user from the two candidate elements. In FIG. 2, the information processing device 100 receives selection information related to data D5, the selection element selected by the user. Furthermore, when the information processing device 100 receives selection information related to data D5, it outputs detailed information about data D5.

Note that Figure 2 simplifies the structure of the TMF information, and describes a case where the TMF information is hierarchically structured data containing three layers. Below, actual TMF information is described. Actual TMF information is hierarchically structured data containing five layers. Below, actual TMF information is simply referred to as "TMF information."

Figure 3 is a diagram showing an example of the first hierarchical level of TMF information according to an embodiment. The first hierarchical level of TMF information is the highest hierarchical level in the hierarchical structure of TMF information. In Figure 3, the first hierarchical level is indicated by H10. For example, the first hierarchical level H10 corresponds to the upper category hierarchical level H1 described in Figure 2. The first hierarchical level H10 also includes, as elements, information indicating each of the four business flow groups of TMF information. In Figure 3, elements 11 to 14 indicate each of the four business flow groups of TMF information. The four business flow groups of TMF information also correspond to the four major categories of TMF information. For example, elements 11 to 14 indicate each of the four business flow groups of TMF information, corresponding to the upper categories UC1 and UC2 described in Figure 2. The four business flow groups in the TMF information are called "Customer Centric Processes" (corresponding to element 11 in Figure 3), "Network Centric Processes" (corresponding to element 12 in Figure 3), "Product Centric Processes" (corresponding to element 13 in Figure 3), and "Engaged Party Flows" (corresponding to element 14 in Figure 3). Here, a business flow group is information that includes each of multiple business flows as an element. Also, a business flow is information that includes each of multiple tasks included in a flow diagram that shows the flow of business operations as an element. Also, in Figure 3, information processing device 100 selects element 11, which corresponds to "Customer Centric Processes," from elements 11 to 14.

Figure 4 is a diagram showing an example of the second hierarchical level of TMF information according to an embodiment. The second hierarchical level of TMF information is the second hierarchical level counting from the top hierarchical level in the hierarchical structure of the TMF information. In other words, the second hierarchical level of TMF information is the level one level below the first hierarchical level in the hierarchical structure of the TMF information. In Figure 4, the second hierarchical level is indicated by H20. For example, the second hierarchical level H20 corresponds to the category hierarchical level H2 described in Figure 2. The second hierarchical level H20 also includes, as elements, information indicating each of the multiple business flows included in the business flow group of the TMF information. More specifically, the second hierarchical level H20 includes multiple child elements having each element included in the first hierarchical level H10 as a parent element. The second hierarchical level H20 includes multiple child elements having each of elements 11 to 14 included in the first hierarchical level H10 as a parent element. Figure 4 shows multiple child elements 211 to 215 having element 11 as a parent element. For example, the "Customer Centric processes" selected in Figure 3 (corresponding to element 11 in Figure 3) includes five child elements, such as a "Request-to-Answer" (corresponding to element 211 in Figure 4), an "Order-to-Payment" (corresponding to element 212 in Figure 4), a "Usage-to-Payment" (corresponding to element 213 in Figure 4), a "Request-to-Change" (corresponding to element 214 in Figure 4), and a "Termination-to-Confirmation" (corresponding to element 215 in Figure 4). Also, in Figure 4, the group including child elements 211 to 215 of element 11 is indicated by 21. Although not shown, the second hierarchical level H20 includes, in addition to group 21, group 22 of multiple child elements with element 12 as its parent element, group 23 of multiple child elements with element 13 as its parent element, and group 24 of multiple child elements with element 14 as its parent element. In Figure 4, the user selects element 212, which corresponds to the "Order-to-Payment Process," from elements 211-215.

Figure 5 is a diagram showing an example of the third hierarchical level of TMF information according to an embodiment. The third hierarchical level of TMF information is the third hierarchical level counting from the top hierarchical level in the hierarchical structure of the TMF information. In other words, the third hierarchical level of TMF information is the level one level below the second hierarchical level in the hierarchical structure of the TMF information. The third hierarchical level of TMF information is also the third hierarchical level counting from the bottom hierarchical level in the hierarchical structure of the TMF information. In Figure 5, the third hierarchical level is indicated by H30. For example, the third hierarchical level H30 corresponds to the category hierarchical level H2 described in Figure 2. Note that, from the perspective of the third hierarchical level H30, the second hierarchical level H20 corresponds to the upper category hierarchical level H1 described in Figure 2. The third hierarchical level H30 includes, as elements, information indicating each of the multiple tasks included in the business flow of the TMF information. More specifically, the third hierarchical level H30 includes multiple child elements with each element included in the second hierarchical level H20 as a parent element. FIG. 5 shows multiple child elements 311 to 315 with element 212 as the parent element. For example, the "Order-to-Payment Process (Order-to-Payment)" selected in FIG. 4 (corresponding to element 212 in FIG. 4 ) includes five child elements, such as "Customer Inventory Management" (corresponding to element 311 in FIG. 5 ). Also, in FIG. 5 , a group including child elements 311 to 315 of element 212 is indicated by 31. Although not shown, the third hierarchical level H30 includes, in addition to group 31, multiple child elements with each element included in group 21 (elements 211, 213 to 215) as its parent element. Furthermore, the third hierarchical level H30 includes, in addition to multiple child elements with each element included in group 21 as its parent element, multiple child elements with each element included in group 22 as its parent element, multiple child elements with each element included in group 23 as its parent element, and multiple child elements with each element included in group 24 as its parent element. Also, in Figure 5, the user selects element 311, which corresponds to "Customer Inventory Management," from elements 311 to 315.

Figure 6 is a diagram showing an example of the fourth layer of TMF information according to an embodiment. The fourth layer of TMF information is the fourth layer counting from the top layer in the hierarchical structure of TMF information. In other words, the fourth layer of TMF information is the layer immediately below the third layer in the hierarchical structure of TMF information. The fourth layer of TMF information is also the second layer counting from the bottom layer in the hierarchical structure of TMF information. In Figure 6, the fourth layer is indicated by H40. For example, the fourth layer H40 corresponds to the category layer H2 described in Figure 2. Note that, from the perspective of the fourth layer H40, the third layer H30 corresponds to the upper category layer H1 described in Figure 2. The fourth layer H40 also includes, as elements, information indicating the classification of functions (e.g., by application) for realizing each task of the TMF information. More specifically, the fourth layer H40 includes multiple child elements with each element included in the third layer 30 as a parent element. FIG. 6 shows multiple child elements 411-413 with element 311 as the parent element. For example, "Customer Inventory Management" (corresponding to element 311 in FIG. 5) selected in FIG. 5 includes three child elements: "Billing Account Management" (corresponding to element 411 in FIG. 6), "Invoice Issuance" (corresponding to element 412 in FIG. 6), and "Implementation of Risk Mitigation Measures" (corresponding to element 413 in FIG. 6). Also, in FIG. 6, the group including child elements 411-413 of element 311 is indicated by 41. Although not shown, the fourth hierarchical level H40 includes, in addition to group 41, multiple child elements with each element included in group 31 (elements 312-315) as its parent element. Furthermore, the fourth hierarchical level H40 includes multiple child elements with each element included in group 31 as its parent element, as well as multiple child elements with each element included in the third hierarchical level 30 as its parent element. Also, in FIG. 6, the user selects element 411 corresponding to "Billing Account Management" from among elements 411-413.

Figure 7 is a diagram showing an example of the fifth layer of TMF information according to an embodiment. The fifth layer of TMF information is the fifth layer counting from the top layer in the hierarchical structure of TMF information. In other words, the fifth layer of TMF information is the layer immediately below the fourth layer in the hierarchical structure of TMF information. The fifth layer of TMF information is also the lowest layer in the hierarchical structure of TMF information. In Figure 7, the fifth layer is indicated by H50. For example, the fifth layer H50 corresponds to the data layer H3 described in Figure 2. The fifth layer H50 also includes, as elements, information indicating functions (specifically, applications) for realizing each task of the TMF information. More specifically, the fifth layer H50 includes multiple child elements with each element included in the fourth layer 40 as a parent element. Figure 7 shows multiple child elements 511 to 516 with element 411 as a parent element. For example, "Billing Account Management" (corresponding to element 411 in FIG. 6 ) selected in FIG. 6 includes six child elements, such as "Billing Account Creation" (corresponding to element 511 in FIG. 7 ), "Billing Account Structure Configuration" (corresponding to element 512 in FIG. 7 ), "Billing Account Information Configuration" (corresponding to element 513 in FIG. 7 ), "Customer Billing Display" (corresponding to element 514 in FIG. 7 ), "Customer Billing Cost Center Analysis" (corresponding to element 515 in FIG. 7 ), and "Customer Billing Hierarchy Management" (corresponding to element 516 in FIG. 7 ). Also, in FIG. 7 , a group including child elements 511 to 516 of element 411 is indicated by 51. Although not shown, the fifth hierarchical level H50 includes, in addition to group 51, multiple child elements whose parent elements are the elements included in group 41 (elements 412 to 413). Furthermore, the fifth hierarchical level H50 includes, in addition to multiple child elements whose parent elements are the elements included in group 41, multiple child elements whose parent elements are the elements included in the fourth hierarchical level 40. Also, in Figure 7, the user selects element 514, which corresponds to "Customer Billing Display," from elements 511 to 516.

2. Configuration of the information processing system
An example of the configuration of an information processing system 1 according to an embodiment will be described with reference to FIG. 8 . FIG. 8 is a diagram showing an example of the configuration of the information processing system 1 according to an embodiment. As shown in FIG. 8 , the information processing system 1 includes a terminal device 10 and an information processing device 100. The terminal device 10 and the information processing device 100 are connected to each other via a network N so as to be able to communicate with each other via a wired or wireless connection. The network N is, for example, a wide area network (WAN) such as the Internet. Note that the information processing system 1 shown in FIG. 8 may include a plurality of terminal devices 10 and a plurality of information processing devices 100.

The terminal device 10 is an information processing device used by a user. The terminal device 10 may be realized, for example, as a smartphone, tablet terminal, notebook PC (Personal Computer), desktop PC, mobile phone, or PDA (Personal Digital Assistant). The terminal device 10 also displays information received from the information processing device 100 or the like using a web browser or application.

The information processing device 100 is an information processing device that performs information processing according to the embodiment, and is realized, for example, by a server device or a cloud system. In order for a user to select data included in the lowest hierarchical level from the TMF information, the information processing device 100 extracts candidate elements for the user to select for each hierarchical level and presents them to the user.

3. Configuration of information processing device
An example of the configuration of the information processing device 100 according to the embodiment will be described with reference to Fig. 9. Fig. 9 is a diagram showing an example of the configuration of the information processing device 100 according to the embodiment. The information processing device 100 includes a communication unit 110, a storage unit 120, and a control unit 130.

(Communication unit 110)
The communication unit 110 is realized by a network interface card (NIC), an antenna, etc. The communication unit 110 is connected to various networks by wire or wirelessly, and transmits and receives information to and from the terminal device 10, for example.

(Storage unit 120)
The storage unit 120 is realized by, for example, a semiconductor memory element such as a random access memory (RAM) or a flash memory, or a storage device such as a hard disk or an optical disk. Specifically, the storage unit 120 stores various types of data. For example, the storage unit 120 stores various types of information received by the reception unit 131. The storage unit 120 also stores various types of information generated by the generation unit 133.

The storage unit 120 also stores various programs. For example, the storage unit 120 stores an information processing program according to the embodiment. The storage unit 120 also stores information related to generative models. For example, the storage unit 120 stores information related to large-scale language models (LLMs).

The storage unit 120 also has a TMF information storage unit 121. The TMF information storage unit 121 stores various information related to TMF information. Here, an example of the information stored in the TMF information storage unit 121 will be described using Figures 10 to 14. Figures 10 to 14 are diagrams showing an example of the information stored in the TMF information storage unit 121.

For example, the TMF information storage unit 121 stores tables 121-1 to 121-5 shown in Figures 10 to 14. Tables 121-1 to 121-5 have fields such as "Hierarchy ID," "Parent Element ID," "Child Element ID," "Name," and "Description." "Hierarchy ID" indicates identification information for identifying a hierarchy in the hierarchical structure of the TMF information. "Parent Element ID" indicates identification information for identifying the parent element of each element included in the hierarchy identified by the hierarchy ID. "Child Element ID" indicates identification information for identifying each element included in the hierarchy identified by the hierarchy ID. "Name" indicates the name of each element included in the hierarchy identified by the hierarchy ID. "Description" indicates a description of each element included in the hierarchy identified by the hierarchy ID.

Table 121-1 shown in Figure 10 stores information related to the first hierarchical level of the TMF information described in Figure 3. For example, the hierarchical level ID "#1" shown in Figure 10 is identification information for identifying the first hierarchical level of the TMF information. Furthermore, the parent element ID "#10" is identification information for identifying the parent element of each element included in the first hierarchical level of the TMF information. Specifically, the parent element of each element included in the first hierarchical level of the TMF information is the root node RT (see Figure 2) in the hierarchical structure of the TMF information. Furthermore, each of the child element IDs "#11" to "#14" is identification information for identifying each of elements 11 to 14 (see Figure 3) included in the first hierarchical level of the TMF information. Furthermore, the name "Customer Centric processes" indicates the name of element 11. Furthermore, the name "Network Centric processes" indicates the name of element 12. Furthermore, the name "Product Centric processes" indicates the name of element 13. Additionally, the name "Engaged Party Flows" indicates the name of element 14.

Table 121-2 shown in Figure 11 stores information related to the second hierarchical level of the TMF information described in Figure 4. For example, the hierarchical level ID "#2" shown in Figure 11 is identification information for identifying the second hierarchical level of the TMF information. Furthermore, each of the parent element IDs "#11" to "#14" is identification information for identifying the parent element of each element included in the second hierarchical level of the TMF information. Specifically, the parent elements of each element included in the second hierarchical level of the TMF information are elements 11 to 14 included in the first hierarchical level of the TMF information (see Figure 3). Furthermore, each of the child element IDs "#211" to "#215" is identification information for identifying each of elements 211 to 215 included in the second hierarchical level of the TMF information (see Figure 4). Furthermore, the name "Request-to-Answer" indicates the name of element 211. Furthermore, the name "Order-to-Payment" indicates the name of element 212. Also, the name "Termination-to-Confirmation" indicates the name of element 215.

Table 121-3 shown in FIG. 12 stores information related to the third hierarchical level of the TMF information described in FIG. 5. For example, the hierarchical level ID "#3" shown in FIG. 12 is identification information for identifying the third hierarchical level of the TMF information. Furthermore, each of the parent element IDs "#211" to "#213" is identification information for identifying the parent element of each element included in the third hierarchical level of the TMF information. Specifically, the parent element of each element included in the third hierarchical level of the TMF information is each element included in the second hierarchical level of the TMF information (e.g., elements 211 to 215 (see FIG. 4)). Furthermore, each of the child element IDs "#311" to "#315" is identification information for identifying each of elements 311 to 315 (see FIG. 5) included in the third hierarchical level of the TMF information. Furthermore, the name "Customer Inventory Management" indicates the name of element 311.

Table 121-4 shown in FIG. 13 stores information related to the fourth hierarchical level of the TMF information described in FIG. 6. For example, the hierarchical level ID "#4" shown in FIG. 13 is identification information for identifying the fourth hierarchical level of the TMF information. Furthermore, each of the parent element IDs "#311" to "#312" is identification information for identifying the parent element of each element included in the fourth hierarchical level of the TMF information. Specifically, the parent element of each element included in the fourth hierarchical level of the TMF information is each element included in the third hierarchical level of the TMF information (e.g., element 311 (see FIG. 5)). Furthermore, each of the child element IDs "#411" to "#413" is identification information for identifying each of elements 411 to 413 (see FIG. 6) included in the fourth hierarchical level of the TMF information. Furthermore, the name "Billing Account Management" indicates the name of element 411. Furthermore, the name "Invoice Issuance" indicates the name of element 412. Furthermore, the name "Implementation of Risk Mitigation Measures" indicates the name of element 413.

Table 121-5 shown in FIG. 13 stores information related to the fifth hierarchical level of the TMF information described in FIG. 7. For example, the hierarchical level ID "#5" shown in FIG. 13 is identification information for identifying the fifth hierarchical level of the TMF information. Furthermore, each of the parent element IDs "#411" to "#412" is identification information for identifying the parent element of each element included in the fifth hierarchical level of the TMF information. Specifically, the parent element of each element included in the fifth hierarchical level of the TMF information is each element included in the fourth hierarchical level of the TMF information (e.g., elements 411 to 413 (see FIG. 6)). Furthermore, each of the child element IDs "#511" to "#516" is identification information for identifying each of elements 511 to 516 (see FIG. 7) included in the fifth hierarchical level of the TMF information. Furthermore, the name "Billing Account Creation" indicates the name of element 511. Furthermore, the name "Customer Billing Hierarchical Management" indicates the name of element 516.

(Control unit 130)
The control unit 130 is a controller, and is realized, for example, by a CPU (Central Processing Unit) or an MPU (Micro Processing Unit) using RAM as a work area to execute various programs stored in a storage device inside the information processing device 100. The control unit 130 is also a controller, and is realized, for example, by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

The control unit 130 has a reception unit 131, an acquisition unit 132, a generation unit 133, and an output control unit 134 as functional units, and may realize or execute the information processing functions described below. Note that the internal configuration of the control unit 130 is not limited to the configuration shown in FIG. 9, and may be any other configuration that performs the information processing described below. Furthermore, each functional unit indicates a function of the control unit 130, and does not necessarily have to be physically distinct.

(Reception unit 131)
The receiving unit 131 receives various information. Specifically, the receiving unit 131 receives various information from the terminal device 10. More specifically, the terminal device 10 displays, on a screen, question information indicating a question for identifying desired data included in hierarchically structured data including a data hierarchy including, as elements, each of a plurality of data items including desired data desired by the user, a category hierarchy including, as elements, each of a plurality of categories that classify the plurality of data items, and a higher category hierarchy including, as elements, each of a plurality of higher categories that classify the plurality of categories. For example, the hierarchically structured data may be TMF information. The terminal device 10 also displays, on a screen, an input form that allows the user to input, in text format, answer information indicating the user's answer to the question along with the question information. The terminal device 10 also transmits, as the answer information, text input by the user into the input form to the information processing device 100. The receiving unit 131 also receives answer information from the terminal device 10. For example, the receiving unit 131 receives text input by the user as the answer information. Furthermore, when the receiving unit 131 receives answer information, the receiving unit 131 outputs the answer information to the generating unit 133 .

In this way, the receiving unit 131 receives answer information indicating the user's answer to a question for identifying desired data contained in hierarchically structured data that includes a data hierarchy that includes, as elements, each of a plurality of data items including the desired data desired by the user, a category hierarchy that includes, as elements, each of a plurality of categories into which the plurality of data items are classified, and a higher-level category hierarchy that includes, as elements, each of a plurality of higher-level categories into which the plurality of categories are classified. Note that, below, the "hierarchical structure including a data hierarchy, a category hierarchy, and a higher-level category hierarchy" may be abbreviated as "hierarchical structure." Also, below, a case will be described in which the data in the hierarchical structure is TMF information.

The terminal device 10 also displays on the screen candidate information related to candidate elements that are candidates for elements corresponding to the answer information, among the multiple elements included in one level of the hierarchical structure. For example, the terminal device 10 displays on the screen a form (e.g., radio buttons) that allows each of the multiple candidate elements to be selected, along with candidate information related to the multiple candidate elements. The terminal device 10 also transmits selection information related to a selected element that is a candidate element selected by the user from the multiple candidate elements to the information processing device 100. The receiving unit 131 also receives selection information related to a selected element that is a candidate element selected by the user from the multiple candidate elements. For example, the receiving unit 131 receives information that can identify the selected element (e.g., a child element ID of the selected element) as the selection information. When the receiving unit 131 receives the selection information, it outputs the selection information to the acquisition unit 132.

(Acquisition unit 132)
The acquisition unit 132 acquires various information. Specifically, the acquisition unit 132 acquires selection information from the reception unit 131. For example, the acquisition unit 132 acquires information that can identify the selected element (e.g., the element ID of the selected element) as the selection information. Furthermore, the acquisition unit 132 acquires, as hierarchical information, child element information regarding multiple child elements having the selected element as a parent element based on the selection information. For example, when the acquisition unit 132 acquires the element ID of the selected element, the acquisition unit 132 references the TMF information storage unit 121 and acquires child element information regarding each of multiple child elements identified by each of multiple child element IDs having the element ID of the selected element as a parent element ID. For example, the acquisition unit 132 acquires, as child element information, information indicating the child element ID, name, and description of each of the multiple child elements. In this way, the acquisition unit 132 acquires hierarchical information regarding multiple elements included in one layer in the hierarchical structure. Furthermore, when the acquisition unit 132 acquires hierarchical information, the acquisition unit 132 outputs the hierarchical information to the generation unit 133.

(Generation unit 133)
The generation unit 133 generates various types of information. Specifically, the generation unit 133 acquires answer information from the receiving unit 131. Furthermore, the generation unit 133 acquires hierarchical information from the acquisition unit 132. When the generation unit 133 acquires the answer information and the hierarchical information, the generation unit 133 generates a prompt including the answer information and the hierarchical information. More specifically, the generation unit 133 inputs the answer information and the hierarchical information into a pre-prepared template, respectively, to generate a prompt including the answer information and the hierarchical information. For example, the generation unit 133 generates a prompt that instructs the generative model to calculate the similarity between the answer information and the hierarchical information. For example, the generation unit 133 generates a prompt that instructs the generative model to calculate the similarity between the answer information and information about each of a plurality of elements included in a hierarchical level corresponding to the hierarchical information. Furthermore, the generation unit 133 generates a prompt that instructs the generative model to select a predetermined number of candidate elements from among a plurality of elements in descending order of similarity, based on the similarity between the answer information and information about each of the plurality of elements. The generation unit 133 also generates a prompt that instructs the generative model to generate candidate information regarding a candidate element selected based on the similarity. When the generation unit 133 generates a prompt including answer information and hierarchical information, the generation unit 133 inputs the prompt including the answer information and hierarchical information to the generative model, causing the generative model to generate candidate information regarding candidate elements that are candidates for elements corresponding to the answer information among multiple elements included in a single hierarchical structure. For example, the generation unit 133 causes the generative model to calculate the similarity between the answer information and the hierarchical information, and causes the generative model to generate candidate information regarding candidate elements selected from the multiple elements based on the similarity. For example, the generation unit 133 causes the generative model to calculate the similarity between the answer information and information regarding each of multiple elements included in a single hierarchical structure corresponding to the hierarchical information, and causes the generative model to generate candidate information regarding candidate elements selected from the multiple elements based on the similarity. For example, the generation unit 133 inputs a prompt including child element information to the generative model, causing the generative model to generate candidate information regarding candidate elements among the multiple child elements. For example, the generation unit 133 causes a generative model to calculate a similarity between information about each of a plurality of child elements corresponding to the child element information and the answer information, and causes the generative model to generate candidate information about a candidate element selected from the plurality of child elements based on the similarity. Here, the generative model is a machine learning model that generates and outputs information according to input information. For example, the generative model is a large language model (LLM). For example, the generative model is a generative pre-trained transformer (GPT) model.

For example, the generation unit 133 inputs a prompt to the generative model instructing it to output information that can identify the candidate elements (e.g., the name of the candidate element, the element ID of the candidate element, etc.). Then, the generation unit 133 causes the generative model to generate, as candidate information, information that indicates the information that can identify the candidate elements (e.g., the name of the candidate element, the element ID of the candidate element, etc.). The generation unit 133 also inputs a prompt to the generative model instructing it to output a description of the candidate element. Then, the generation unit 133 causes the generative model to generate, as candidate information, information that indicates the description of the candidate element. The generation unit 133 also inputs a prompt to the generative model instructing it to output candidate information regarding a candidate element selected based on the similarity between the answer information and the hierarchical information. For example, the generation unit 133 calculates the similarity between the answer information and information regarding each of multiple elements included in a layer corresponding to the hierarchical information, and inputs a prompt to the generative model instructing it to output candidate information regarding a candidate element selected from the multiple elements based on the calculated similarity. For example, the generation unit 133 selects a predetermined number of candidate elements from the plurality of elements in descending order of similarity based on the degree of similarity between information about each of the plurality of elements and the answer information, and inputs a prompt to the generation model instructing the generation model to output candidate information about the selected predetermined number of candidate elements. The generation unit 133 also inputs a prompt to the generation model instructing the generation model to output information about each of the plurality of elements included in a layer corresponding to the hierarchical information and the degree of similarity between the answer information. For example, the generation unit 133 inputs a prompt to the generation model instructing the generation model to convert the degree of similarity between the answer information and information about each of the plurality of elements included in a layer corresponding to the hierarchical information into a numerical value between 0 and 100 and output the converted value as a probability. In this way, the generation unit 133 inputs a prompt including the answer information and the hierarchical information to the generation model, and causes the generation model to generate candidate information about candidate elements from the plurality of elements that are candidates for the element corresponding to the answer information. When the generation unit 133 generates candidate information, it outputs the candidate information to the output control unit 134.

(Output control unit 134)
The output control unit 134 outputs various information. Specifically, the output control unit 134 acquires candidate information from the generation unit 133. The output control unit 134 also outputs candidate information related to candidate elements. For example, the output control unit 134 outputs candidate information related to multiple candidate elements. For example, when the output control unit 134 acquires candidate information, it transmits the candidate information to the terminal device 10. The terminal device 10 receives candidate information related to multiple candidate elements. When the terminal device 10 receives candidate information related to multiple candidate elements, it displays the candidate information related to the multiple candidate elements on the screen. For example, the terminal device 10 displays, on the screen, the candidate information related to the multiple candidate elements as well as a form (e.g., radio buttons) that allows each of the multiple candidate elements to be selected.

Next, an example of information processing according to the embodiment will be described in detail using Figures 15 to 26. Figure 15 is a diagram illustrating an example of a screen of the terminal device 10 according to the embodiment. Screen G1 shown in Figure 15 is a screen that displays question information indicating a question for identifying the user's desired data contained in the TMF information. Specifically, screen G1 is a screen that displays question information indicating a question for identifying the user's desired business flow group from four business flow groups in the TMF information and identifying the user's desired business flow from multiple business flows contained in the identified business flow group. For example, the information processing device 100 may transmit content corresponding to screen G1 to the terminal device 10 in response to a request from the terminal device 10. For example, the terminal device 10 receives content corresponding to screen G1 from the information processing device 100. Furthermore, when the terminal device 10 receives content corresponding to screen G1, it displays screen G1. Screen G1 includes a form F11 in which the development objective can be entered in text format and a form F12 in which a project summary can be entered in text format. In FIG. 15 , the user inputs answer information A11, which is the text "We are providing smartphone setup support to standardize and unify initial setup and some services due to the increase in online purchases. In this development, we are creating a subscription plan instead of the previous one-time payment system, and providing an option for unlimited smartphone setup support on a monthly basis." into form F11. The user also inputs answer information A12, which is the text "1. Check the application information based on the application received from the customer and register the monthly smartphone setup support option. 2. Manage the link between customer information and optional services. 3. Activate the optional service." into form F12. Screen G1 also includes a button B1 displaying the words "Identify Business Flow." The user presses button B1 when completing the input operations for answer information A11 and A12. In response to the user's operation of pressing button B1, the terminal device 10 transmits answer information A11 and A12 to the information processing device 100. More specifically, terminal device 10 transmits hierarchical ID "#1" that identifies the first hierarchical layer of the TMF information to information processing device 100, along with answer information A11 and A12. Furthermore, reception unit 131 receives hierarchical ID "#1" that identifies the first hierarchical layer H10 of the TMF information from terminal device 10, along with answer information A11 and A12. When reception unit 131 receives hierarchical ID "#1" that identifies the first hierarchical layer H10 of the TMF information, along with answer information A11 and A12, it outputs answer information A11 and A12 to generation unit 133. Furthermore, reception unit 131 outputs hierarchical ID "#1" to acquisition unit 132.

Figure 16 is a diagram showing an example of a prompt according to the embodiment. Prompt P11 shown in Figure 16 is used to identify the business flow group desired by the user from the four business flow groups of the TMF information. Specifically, the generation unit 133 acquires answer information A11 and A12 from the reception unit 131. Furthermore, the acquisition unit 132 acquires hierarchical ID "#1" from the reception unit 131. When the acquisition unit 132 acquires hierarchical ID "#1", it acquires first hierarchical information regarding the first hierarchical level H10 corresponding to hierarchical level ID "#1" based on the hierarchical level ID "#1". For example, the acquisition unit 132 acquires, as the first hierarchical level information, information HD11 to HD14 regarding each of the four elements 11 to 14 included in the first hierarchical level H10 of the TMF information. More specifically, the acquisition unit 132 references the TMF information storage unit 121 and acquires information HD11 to HD14 relating to each of the four elements 11 to 14 stored in table 121-1 corresponding to hierarchical ID "#1." For example, the acquisition unit 132 acquires information HD11 to HD14 relating to each of the four elements 11 to 14, such as the element ID, name, and description that identify each of the four elements 11 to 14. In this way, the acquisition unit 132 acquires first hierarchical information relating to the first hierarchical layer H10 of the TMF information. Furthermore, when the acquisition unit 132 acquires the first hierarchical information, it outputs the first hierarchical information to the generation unit 133.

The generation unit 133 also acquires first hierarchical information from the acquisition unit 132. When the generation unit 133 acquires answer information A11 and A12 and the first hierarchical information, it generates a prompt P11 that includes answer information A11 and A12 and the first hierarchical information. More specifically, the generation unit 133 inputs answer information A11 and A12 and the first hierarchical information into a template prepared in advance for generating the prompt P11, thereby generating the prompt P11 that includes answer information A11 and A12 and the first hierarchical information. In FIG. 16, the first hierarchical information is indicated by HD11 to HD14. The first hierarchical information HD11 to HD14 is information such as element IDs, names, and descriptions that identify each of the four elements 11 to 14. When the generation unit 133 generates the prompt P11, it inputs the prompt P11 into the generative model. The generation unit 133 also inputs the prompt P11 to the generative model, causing the generative model to generate first candidate information regarding a first candidate element, which is a candidate for an element corresponding to the answer information A11 and A12, among the four elements 11 to 14 included in the first hierarchical level H10 of the TMF information. For example, the generation unit 133 causes the generative model to calculate the similarity between each of the pieces of information HD11 to HD14 regarding the four elements 11 to 14 corresponding to the first hierarchical level information and the answer information A11 and A12. The generation unit 133 also causes the generative model to select, from among the four elements 11 to 14, the element corresponding to the information having the highest similarity to the answer information as the first candidate element. The generation unit 133 also causes the generative model to generate first candidate information regarding the selected first candidate element. For example, the generative model generates first candidate information regarding element 11, which is the first candidate element. For example, the generative model generates first candidate information that includes the element ID, name, and description of element 11, which is the first candidate element, as well as the reason why element 11 is a group of business flows desired by the user. In this way, the generation unit 133 generates first candidate information regarding element 11, which is the first candidate element. Furthermore, when the generation unit 133 generates the first candidate information, it outputs the first candidate information to the acquisition unit 132 and the output control unit 134.

Figure 17 is a diagram showing an example of a prompt according to the embodiment. Prompt P12 shown in Figure 17 is used to identify a business flow desired by a user from among multiple business flows included in the business flow group identified in Figure 16. Specifically, when generating first candidate information, the generation unit 133 generates prompt P12. More specifically, the acquisition unit 132 acquires the first candidate information from the generation unit 133. When acquiring the first candidate information, the acquisition unit 132 acquires second hierarchical information based on the first candidate information. For example, the acquisition unit 132 acquires, as the second hierarchical information, child element information HD211 to HD215 related to each of multiple child elements 211 to 215 having element 11, which is the first candidate element, as a parent element. For example, the acquisition unit 132 acquires, as the first candidate information, element ID "#11" that identifies element 11, which is the first candidate element. Furthermore, when the acquisition unit 132 acquires the element ID "#11" of element 11, it references the TMF information storage unit 121 and acquires child element information HD211-HD215 related to each of the multiple child elements 211-215 identified by the multiple child element IDs "#211"-"#215" whose parent element ID is the element ID "#11" of element 11, the first candidate element. For example, the acquisition unit 132 acquires, as child element information HD211-HD215, information indicating the child element ID, name, and description of each of the multiple child elements 211-215. In this way, the acquisition unit 132 acquires second hierarchical information related to the second hierarchical layer H20 of the TMF information. Furthermore, when the acquisition unit 132 acquires second hierarchical information, it outputs the second hierarchical information to the generation unit 133.

Furthermore, when the generation unit 133 acquires answer information A11 and A12 and the second hierarchical information, it generates a prompt P12 that includes answer information A11 and A12 and the second hierarchical information. More specifically, the generation unit 133 inputs answer information A11 and A12 and the second hierarchical information into a template prepared in advance for generating prompt P12, thereby generating prompt P12 that includes answer information A11 and A12 and the second hierarchical information. In FIG. 17, the second hierarchical information is indicated by HD211 to HD215. The second hierarchical information HD211 to HD215 is information that indicates the child element ID, name, and description of each of the multiple child elements 211 to 215. Furthermore, when the generation unit 133 generates prompt P12, it inputs prompt P12 into the generation model. The generation unit 133 also inputs the prompt P12 into the generative model, causing the generative model to generate second candidate information regarding second candidate elements, which are candidates for elements corresponding to the answer information A11 and A12, among the multiple elements 211-215 included in the second hierarchical level H20 of the TMF information. For example, the generation unit 133 causes the generative model to calculate the similarity between each of the pieces of information HD211-HD215 regarding the multiple elements 211-215 corresponding to the second hierarchical level information and the answer information A11 and A12. The generation unit 133 also causes the generative model to select, from the multiple elements 211-215, three elements with the highest similarity to the answer information as second candidate elements. The generation unit 133 also causes the generative model to generate second candidate information regarding the selected second candidate elements. For example, the generative model generates second candidate information regarding the second candidate elements 211, 212, and 214. For example, the generative model generates second candidate information including information such as the element ID, name, and description of each of the second candidate elements 211, 212, and 214, as well as information on why each of the elements 211, 212, and 214 is part of the business flow desired by the user. The generative model also generates information on each of the second candidate elements 211, 212, and 214, and information indicating the degree of similarity between the information A11 and A12. In this way, the generation unit 133 generates second candidate information on the second candidate elements 211, 212, and 214. When the generation unit 133 generates second candidate information, it outputs the second candidate information to the output control unit 134.

Figure 18 is a diagram showing an example of a screen of the terminal device 10 according to the embodiment. Screen G2 shown in Figure 18 is a screen displayed after the user presses button B1 on screen G1 shown in Figure 15. Specifically, the output control unit 134 acquires first candidate information and second candidate information from the generation unit 133. For example, when the output control unit 134 acquires the first candidate information and the second candidate information, it generates content corresponding to screen G2 based on the first candidate information and the second candidate information. When the output control unit 134 generates content corresponding to screen G2, it may transmit the content corresponding to screen G2 to the terminal device 10. For example, the terminal device 10 receives content corresponding to screen G2 from the information processing device 100. Furthermore, when the terminal device 10 receives content corresponding to screen G2, it displays screen G2. Screen G2 includes first candidate information CD2 and second candidate information CD21 to CD23. The first candidate information CD2 is information related to element 11, which is the first candidate element. For example, the first candidate information CD2 includes the name of element 11, which is the first candidate element, and information indicating why element 11 is a business flow group desired by the user. Each of the second candidate information CD21-CD23 includes the name of each of elements 211, 212, and 214, which are second candidate elements, and the reason why each of elements 211, 212, and 214 is a business flow desired by the user. Each of the second candidate information CD21-CD23 also includes information about each of elements 211, 212, and 214, which are second candidate elements, and information indicating the degree of similarity between each of the elements and answer information A11 and A12 (probability shown in FIG. 18). Screen G2 also includes radio buttons R21-R23 that allow selection of each of elements 211, 212, and 214, which are second candidate elements. In FIG. 18, the user selects radio button R21 corresponding to element 212. Screen G2 also includes a button B2 displaying the words "Go to STEP 2." When the user completes the selection operation of radio button R21, the user presses button B2. In response to the user's operation of pressing button B2, terminal device 10 transmits to information processing device 100 first selection information regarding element 212, which is the second candidate element (also referred to as the first selected element) selected by the user from elements 211, 212, and 214, which are second candidate elements. Furthermore, reception unit 131 receives the first selection information from terminal device 10. For example, reception unit 131 receives element ID "#212" of element 212, the first selected element, as the first selection information. Upon receiving the first selection information, reception unit 131 outputs the first selection information to acquisition unit 132. For example, reception unit 131 outputs element ID "#212" of element 212, the first selected element, to acquisition unit 132.

19 is a diagram showing an example of a screen of the terminal device 10 according to the embodiment. Screen G3 shown in FIG. 19 is a screen displayed after the user presses button B2 on screen G2 shown in FIG. 18. For example, when the reception unit 131 receives first selection information, the output control unit 134 may output content corresponding to screen G3 to the terminal device 10. Furthermore, the terminal device 10 may receive content corresponding to screen G3 from the information processing device 100. Furthermore, when the terminal device 10 receives content corresponding to screen G3, it displays screen G3. Screen G3 includes form F3 in which details of the business can be entered in text format. In FIG. 19, the user enters answer information A3, which is text stating, "Manage customer subscription information in a database based on application information received from the customer," into form F3. Furthermore, screen G3 includes button B3 displaying the words "ETOM analysis." When the user has completed entering answer information A3, he or she presses button B3. In response to the user pressing button B3, terminal device 10 transmits answer information A3 to information processing device 100. Furthermore, reception unit 131 receives answer information A3 from terminal device 10. Furthermore, reception unit 131 outputs answer information A3 to generation unit 133.

Figure 20 is a diagram showing an example of a prompt according to the embodiment. Prompt P13 shown in Figure 20 is used to identify a desired task from among multiple tasks included in the business flow identified in Figure 19. Note that "ETOM" in Figures 19 and 20 is a term referring to a task included in a business flow. Specifically, the acquisition unit 132 acquires first selection information from the reception unit 131. For example, the acquisition unit 132 acquires, as the first selection information, the element ID "#212" of element 212, which is the first selected element. Furthermore, when acquiring the first selection information, the acquisition unit 132 acquires third hierarchical information based on the first selection information. For example, the acquisition unit 132 acquires, as the third hierarchical information, child element information HD311 to HD315 related to each of multiple child elements 311 to 315 having element 212, which is the first selected element, as a parent element. For example, when the acquisition unit 132 acquires the element ID "#212" of the first selected element, element 212, it references the TMF information storage unit 121 and acquires child element information HD311-HD315 related to each of the multiple child elements 311-315 identified by each of the multiple child element IDs whose parent element ID is the element ID "#212" of the first selected element, element 212. For example, the acquisition unit 132 acquires, as child element information HD311-HD315, information indicating the child element ID, name, and description of each of the multiple child elements 311-315. In this way, the acquisition unit 132 acquires third hierarchical information related to the third hierarchical layer H30 of the TMF information. Furthermore, when the acquisition unit 132 acquires the third hierarchical information, it outputs the third hierarchical information to the generation unit 133.

The generation unit 133 also acquires answer information A3 from the reception unit 131. The generation unit 133 also acquires third hierarchical information from the acquisition unit 132. When the generation unit 133 acquires answer information A3 and the third hierarchical information, it generates a prompt P13 including answer information A3 and the third hierarchical information. More specifically, the generation unit 133 inputs answer information A3 and the third hierarchical information into a template prepared in advance for generating prompt P13, thereby generating prompt P13 including answer information A3 and the third hierarchical information. In FIG. 20, the third hierarchical information is indicated by HD311 to HD315. HD311 to HD315, which are the third hierarchical information, are information indicating the child element ID, name, and description of each of the multiple child elements 311 to 315. When the generation unit 133 generates prompt P13, it inputs prompt P13 into the generation model. The generation unit 133 also inputs the prompt P13 into the generative model, causing the generative model to generate third candidate information regarding third candidate elements, which are candidates for elements corresponding to the answer information A3, among the multiple elements 311-315 included in the third hierarchical level H30 of the TMF information. For example, the generation unit 133 causes the generative model to calculate the similarity between each of the pieces of information HD311-HD315 regarding the multiple elements 311-315 corresponding to the third hierarchical level information and the answer information A3. The generation unit 133 also causes the generative model to select, from the multiple elements 311-315, three elements with the highest similarity to the answer information as third candidate elements. The generation unit 133 also causes the generative model to generate third candidate information regarding the selected third candidate elements. For example, the generative model generates third candidate information regarding the elements 311-313, which are the third candidate elements. For example, the generative model generates third candidate information that includes information such as the element ID, name, and description of each of the third candidate elements 311-313, as well as information on why each of the elements 311-313 is a task desired by the user. The generative model also generates information on each of the third candidate elements 311-313 and information indicating the degree of similarity with answer information A3. In this way, the generation unit 133 generates third candidate information related to the third candidate elements 311-313. Furthermore, when the generation unit 133 generates the third candidate information, it outputs the third candidate information to the output control unit 134.

Figure 21 is a diagram showing an example of a screen of the terminal device 10 according to the embodiment. Screen G4 shown in Figure 21 is a screen displayed after the user presses button B3 on screen G3 shown in Figure 19. Specifically, the output control unit 134 acquires third candidate information from the generation unit 133. For example, when the output control unit 134 acquires the third candidate information, it generates content corresponding to screen G4 based on the third candidate information. When the output control unit 134 generates content corresponding to screen G4, it may transmit the content corresponding to screen G4 to the terminal device 10. For example, the terminal device 10 receives content corresponding to screen G4 from the information processing device 100. Furthermore, when the terminal device 10 receives content corresponding to screen G4, it displays screen G4. Screen G4 includes third candidate information CD41 to CD43. Each of the third candidate information CD41 to CD43 is information relating to each of the third candidate elements 312, 313, and 311. For example, each of the third candidate information CD41 to CD43 includes the names of the third candidate elements 312, 313, and 311, as well as the reasons why each of the elements 312, 313, and 311 is the user's desired task. Each of the third candidate information CD41 to CD43 also includes information about the third candidate elements 312, 313, and 311, as well as information indicating the degree of similarity with the answer information A3 (the probability shown in FIG. 21). Screen G2 also includes radio buttons R41 to R43, each of which allows the user to select the third candidate elements 312, 313, and 311. In FIG. 21, the user selects radio button R43 corresponding to element 311. Screen G4 also includes button B4, on which the words "Go to STEP 3" are displayed. After completing the selection of radio button R43, the user presses button B4. In response to the user pressing button B4, the terminal device 10 transmits to the information processing device 100 second selection information related to element 311, which is the third candidate element (also referred to as the second selected element) selected by the user from elements 311 to 313, which are the third candidate elements. The reception unit 131 also receives the second selection information from the terminal device 10. For example, the reception unit 131 receives the element ID "#311" of element 311, the second selected element, as the second selection information. When the reception unit 131 receives the second selection information, it outputs the second selection information to the acquisition unit 132. For example, the reception unit 131 outputs the element ID "#311" of element 311, the second selected element, to the acquisition unit 132.

22 is a diagram showing an example of a screen of the terminal device 10 according to the embodiment. Screen G5 shown in FIG. 22 is a screen displayed after the user presses button B4 on screen G4 shown in FIG. 21. For example, when the reception unit 131 receives second selection information, the output control unit 134 may output content corresponding to screen G5 to the terminal device 10. Furthermore, when the terminal device 10 receives content corresponding to screen G5, it displays screen G5. Screen G5 includes form F5, which allows the user to input functions (e.g., application units, etc.) for implementing each business in text format. In FIG. 22, the user inputs answer information A5, which is the text "Manage customer subscription information in a database based on application information received from the customer," into form F5. Screen G5 also includes button B53 displaying the words "Specify Aggregate Function." Screen G5 also includes button B52 for the user to add a function and button B51 for the user to delete a function. For example, when the user has completed the input operation of answer information A5, the user presses button B53. In response to the user pressing button B53, the terminal device 10 transmits answer information A5 to the information processing device 100. The reception unit 131 also receives answer information A5 from the terminal device 10. The reception unit 131 also outputs answer information A5 to the generation unit 133.

Figure 23 is a diagram showing an example of a prompt according to the embodiment. Prompt P14 shown in Figure 23 is used to identify the category of function desired by the user from the category of functions for realizing the business identified in Figure 21. Note that "Aggregate Function" shown in Figures 22 to 24 is a term referring to a category of function. Also, "Function" shown in Figures 22, 25 to 26 is a term referring to a function. Specifically, the acquisition unit 132 acquires second selection information from the reception unit 131. For example, the acquisition unit 132 acquires, as the second selection information, the element ID "#311" of element 311, which is the second selection element. Furthermore, when the acquisition unit 132 acquires the second selection information, it acquires fourth hierarchical information based on the second selection information. For example, the acquisition unit 132 acquires, as the fourth hierarchical information, child element information HD411-HD413 relating to each of the multiple child elements 411-413 whose parent element is element 311, the second selected element. For example, when the acquisition unit 132 acquires the element ID "#311" of element 311, the second selected element, the acquisition unit 132 references the TMF information storage unit 121 and acquires child element information HD411-HD413 relating to each of the multiple child elements 411-413 identified by each of the multiple child element IDs whose parent element ID is the element ID "#311" of element 311, the second selected element. For example, the acquisition unit 132 acquires, as the child element information HD411-HD413, information indicating the child element ID, name, and description of each of the multiple child elements 411-413. In this way, the acquisition unit 132 acquires fourth hierarchical information relating to the fourth hierarchical level H40 of the TMF information. Furthermore, when the acquisition unit 132 acquires the fourth hierarchical information, it outputs the fourth hierarchical information to the generation unit 133.

The generation unit 133 also acquires answer information A5 from the reception unit 131. The generation unit 133 also acquires fourth hierarchical information from the acquisition unit 132. When the generation unit 133 acquires answer information A5 and the fourth hierarchical information, it generates a prompt P14 including answer information A5 and the fourth hierarchical information. More specifically, the generation unit 133 inputs answer information A5 and the fourth hierarchical information into a template prepared in advance for generating prompt P14, thereby generating prompt P14 including answer information A5 and the fourth hierarchical information. In FIG. 23, the fourth hierarchical information is indicated by HD411 to HD413. HD411 to HD413, which are the fourth hierarchical information, are information indicating the child element ID, name, and description of each of the multiple child elements 411 to 413. When the generation unit 133 generates prompt P14, it inputs prompt P14 into the generation model. The generation unit 133 also inputs the prompt P14 into the generative model, causing the generative model to generate fourth candidate information regarding a fourth candidate element, which is a candidate for the element corresponding to the answer information A5, among the multiple elements 411-413 included in the fourth hierarchical layer 40 of the TMF information. For example, the generation unit 133 causes the generative model to calculate the similarity between each of the pieces of information HD411-HD413 regarding the multiple elements 411-413 corresponding to the fourth hierarchical layer information and the answer information A5. The generation unit 133 also causes the generative model to select, from the multiple elements 411-413, three elements with the highest similarity to the answer information as fourth candidate elements. The generation unit 133 also causes the generative model to generate fourth candidate information regarding the selected fourth candidate element. For example, the generative model generates fourth candidate information regarding the elements 411-413, which are the fourth candidate elements. For example, the generative model generates fourth candidate information that includes information such as the element ID, name, and description of each of the fourth candidate elements 411-413, as well as information on why each of the elements 411-413 is in the category of function desired by the user. The generative model also generates information on each of the fourth candidate elements 411-413 and information indicating the degree of similarity with answer information A5. In this way, the generation unit 133 generates fourth candidate information on the fourth candidate elements 411-413. When the generation unit 133 generates the fourth candidate information, it outputs the fourth candidate information to the output control unit 134.

Figure 24 is a diagram showing an example of a screen of the terminal device 10 according to the embodiment. Screen G6 shown in Figure 24 is a screen displayed after the user presses button B53 on screen G5 shown in Figure 22. Specifically, the output control unit 134 acquires fourth candidate information from the generation unit 133. For example, when the output control unit 134 acquires the fourth candidate information, it generates content corresponding to screen G6 based on the fourth candidate information. When the output control unit 134 generates content corresponding to screen G6, it may transmit the content corresponding to screen G6 to the terminal device 10. For example, the terminal device 10 receives content corresponding to screen G6 from the information processing device 100. Furthermore, when the terminal device 10 receives content corresponding to screen G6, it displays screen G6. Screen G6 includes fourth candidate information CD61 to CD63. Each of the fourth candidate information CD61 to CD63 is information relating to each of elements 411 to 413, which are the fourth candidate elements. For example, each of the fourth candidate information CD61-CD63 includes the name of each of the fourth candidate elements 411-413, as well as the reason why each of the elements 411-413 is a category of function desired by the user. Each of the fourth candidate information CD61-CD63 also includes information about each of the fourth candidate elements 411-413 and information indicating the degree of similarity with the answer information A5 (probability shown in FIG. 24). Screen G6 also includes radio buttons R61-R63 that allow selection of each of the fourth candidate elements 411-413. In FIG. 24, the user selects radio button R61 corresponding to element 411. Screen G6 also includes button B6 labeled "Function Identification." After completing the selection of radio button R61, the user presses button B6. In response to the user pressing button B6, the terminal device 10 transmits to the information processing device 100 third selection information related to element 411, which is the fourth candidate element (also referred to as the third selected element) selected by the user from elements 411 to 413, which are fourth candidate elements. The reception unit 131 also receives the third selection information from the terminal device 10. For example, the reception unit 131 receives the element ID "#411" of element 411, the third selected element, as the third selection information. When the reception unit 131 receives the third selection information, it outputs the third selection information to the acquisition unit 132. For example, the reception unit 131 outputs the element ID "#411" of element 411, the third selected element, to the acquisition unit 132.

Figure 25 is a diagram showing an example of a prompt according to the embodiment. Prompt P15 shown in Figure 25 is used to identify a function desired by the user from the function classifications identified in Figure 24. Specifically, the acquisition unit 132 acquires third selection information from the reception unit 131. For example, the acquisition unit 132 acquires, as the third selection information, the element ID "#411" of element 411, which is the third selection element. Furthermore, when acquiring the third selection information, the acquisition unit 132 acquires fifth hierarchical information based on the third selection information. For example, the acquisition unit 132 acquires, as the fifth hierarchical information, child element information HD511 to HD516 relating to multiple child elements 511 to 516 having element 411, which is the third selection element, as a parent element. For example, when the acquisition unit 132 acquires the element ID "#411" of the third selected element, element 411, it references the TMF information storage unit 121 and acquires child element information HD511-HD516 related to each of the multiple child elements 511-516 identified by each of the multiple child element IDs whose parent element ID is the element ID "#411" of the third selected element, element 411. For example, the acquisition unit 132 acquires, as child element information HD511-HD516, information indicating the child element ID, name, and description of each of the multiple child elements 511-516. In this way, the acquisition unit 132 acquires fifth hierarchical information related to the fifth hierarchical layer H50 of the TMF information. Furthermore, when the acquisition unit 132 acquires the fifth hierarchical information, it outputs the fifth hierarchical information to the generation unit 133.

The generation unit 133 also acquires fifth hierarchical information from the acquisition unit 132. When the generation unit 133 acquires answer information A5 and the fifth hierarchical information, it generates a prompt P15 that includes answer information A5 and the fifth hierarchical information. More specifically, the generation unit 133 inputs answer information A5 and the fifth hierarchical information into a template prepared in advance for generating prompt P15, thereby generating prompt P15 that includes answer information A5 and the fifth hierarchical information. In FIG. 25, the fifth hierarchical information is indicated by HD511 to HD516. HD511 to HD516, which are the fifth hierarchical information, are information that indicates the child element ID, name, and description of each of the multiple child elements 511 to 516. When the generation unit 133 generates prompt P15, it inputs prompt P15 into the generation model. The generation unit 133 also inputs the prompt P15 into the generative model, causing the generative model to generate fifth candidate information regarding a fifth candidate element, which is a candidate for the element corresponding to the answer information A5, among the plurality of elements 511-516 included in the fifth hierarchical layer 50 of the TMF information. For example, the generation unit 133 causes the generative model to calculate the similarity between each of the pieces of information HD511-HD516 regarding the plurality of elements 511-516 corresponding to the fifth hierarchical layer information and the answer information A5. The generation unit 133 also causes the generative model to select, from the plurality of elements 511-516, three elements having the highest similarity to the answer information as fifth candidate elements. The generation unit 133 also causes the generative model to generate fifth candidate information regarding the selected fifth candidate element. For example, the generative model generates fifth candidate information regarding the fifth candidate elements, elements 514, 516, and 511. For example, the generative model generates fifth candidate information including information such as the element ID, name, and description of each of the fifth candidate elements 514, 516, and 511, as well as information on why each of elements 514, 516, and 511 is a function desired by the user. The generative model also generates information on each of the fifth candidate elements 514, 516, and 511, and information indicating the degree of similarity with answer information A5. In this way, the generation unit 133 generates fifth candidate information related to the fifth candidate elements 514, 516, and 511. Furthermore, when the generation unit 133 generates the fifth candidate information, it outputs the fifth candidate information to the output control unit 134.

Figure 26 is a diagram showing an example of a screen of the terminal device 10 according to the embodiment. Screen G7 shown in Figure 26 is a screen displayed after the user presses button B6 on screen G6 shown in Figure 24. Specifically, the output control unit 134 acquires fifth candidate information from the generation unit 133. For example, when the output control unit 134 acquires the fifth candidate information, it generates content corresponding to screen G7 based on the fifth candidate information. When the output control unit 134 generates content corresponding to screen G7, it may transmit the content corresponding to screen G7 to the terminal device 10. For example, the terminal device 10 receives content corresponding to screen G7 from the information processing device 100. Furthermore, when the terminal device 10 receives content corresponding to screen G7, it displays screen G7. Screen G7 includes fifth candidate information CD71 to CD73. Each of the fifth candidate information CD71 to CD73 is information relating to each of elements 411 to 413, which are fifth candidate elements. For example, each of the fifth candidate information CD71 to CD73 includes the name of each of the fifth candidate elements 514, 516, and 511, as well as the reason why each of the elements 514, 516, and 511 is a function desired by the user. Each of the fifth candidate information CD71 to CD73 also includes information about each of the fifth candidate elements 514, 516, and 511, and information indicating the degree of similarity with the answer information A5. Screen G7 also includes radio buttons R71 to R73 that allow selection of each of the fifth candidate elements 514, 516, and 511. In FIG. 26 , the user selects radio button R71 corresponding to element 514. In response to the user's operation of selecting radio button R71, the terminal device 10 transmits to the information processing device 100 fourth selection information regarding element 514, which is a fifth candidate element (also referred to as a fourth selected element) selected by the user from elements 514, 516, and 511, which are fifth candidate elements. The receiving unit 131 also receives the fourth selection information from the terminal device 10. For example, the receiving unit 131 receives the element ID "#514" of element 514, which is the fourth selected element, as the fourth selection information. When the receiving unit 131 receives the fourth selection information, the output control unit 134 may refer to the TMF information storage unit 121 to acquire information indicating the name and description of element 514, which is the fourth selected element. The output control unit 134 may also output the information indicating the name and description of element 514, which is the fourth selected element, to the terminal device 10. The terminal device 10 may display the information indicating the name and description of element 514, which is the fourth selected element, on the screen.

4. Processing Procedure
FIG. 27 is a flowchart illustrating the procedure of information processing by the information processing device 100 according to the embodiment. In FIG. 27 , the accepting unit 131 accepts answer information indicating a user's answer to a question for identifying desired data included in hierarchically structured data including a data hierarchy including, as elements, each of a plurality of data items including desired data desired by the user, a category hierarchy including, as elements, each of a plurality of categories that classify the plurality of data items, and a higher category hierarchy including, as elements, each of a plurality of higher categories that classify the plurality of categories (step S101). The acquiring unit 132 acquires hierarchy information regarding a plurality of elements included in one hierarchy of the hierarchical structure (step S102). The generating unit 133 generates a prompt including the answer information and the hierarchy information (step S103). The generating unit 133 inputs the prompt including the answer information and the hierarchy information into a generative model, causing the generative model to generate candidate information regarding candidate elements that are candidates for the element corresponding to the answer information among the plurality of elements (step S104). The output controlling unit 134 outputs the candidate information regarding the candidate elements (step S105).

5. Effects
As described above, the information processing device 100 according to the embodiment includes a receiving unit 131, an acquiring unit 132, a generating unit 133, and an output control unit 134. The receiving unit 131 accepts answer information indicating a user's answer to a question for identifying desired data included in hierarchically structured data including a data hierarchy including, as elements, each of a plurality of data items including desired data desired by the user, a category hierarchy including, as elements, each of a plurality of categories that classify the plurality of data items, and a higher category hierarchy including, as elements, each of a plurality of higher categories that classify the plurality of categories. The acquiring unit 132 acquires hierarchy information regarding a plurality of elements included in one hierarchy in the hierarchical structure. The generating unit 133 inputs a prompt including the answer information and the hierarchy information to a generative model, and causes the generative model to generate candidate information regarding candidate elements that are candidates for elements corresponding to the answer information among the plurality of elements. The output control unit 134 outputs the candidate information regarding the candidate elements.

In this way, the information processing device 100 extracts candidate elements for the user to select for each level and presents them to the user so that the user can select data included in the lowest level from hierarchically structured data. This allows the information processing device 100 to support the user's selection at each level to select data included in the lowest level from hierarchically structured data. Furthermore, by extracting candidate elements for the user to select for each level and presenting them to the user, the information processing device 100 can reduce the possibility that the elements selected at each level will differ significantly between users. Furthermore, because the information processing device 100 can reduce the possibility that the elements selected at each level will differ significantly between users, the variance in the data ultimately selected by users can also be reduced. Furthermore, because the information processing device 100 can support the user's selection at each level to select data included in the lowest level from hierarchically structured data, it can contribute to the achievement of Sustainable Development Goal 9 (SDGs), "Build resilient infrastructure, promote inclusive and sustainable development, and build resilient infrastructure."

The output control unit 134 also outputs candidate information related to multiple candidate elements. The receiving unit 131 further receives selection information related to a selected element, which is a candidate element selected by a user from the multiple candidate elements. The acquiring unit 132 acquires, as hierarchical information, child element information related to multiple child elements having the selected element as a parent element, based on the selection information. The generating unit 133 inputs a prompt including the child element information into the generative model, and causes the generative model to generate candidate information related to the candidate element from among the multiple child elements.

This allows the information processing device 100 to appropriately extract candidate elements to be selected by the user based on the user's selection results at one level of hierarchically structured data. Furthermore, because the information processing device 100 can appropriately extract candidate elements to be selected by the user, it can appropriately support the user's selection at levels below the one level.

The generation unit 133 also causes the generative model to calculate the similarity between the answer information and the hierarchical information, and causes the generative model to generate candidate information regarding candidate elements selected from multiple elements based on the similarity.

As a result, the information processing device 100 can extract candidate elements selected based on the similarity between the answer information and hierarchical information and present them to the user, thereby assisting the user in making an appropriate selection.

The generative model is a large-scale language model (LLM).

This allows the information processing device 100 to generate appropriate candidate information based on the user's response information.

6. Hardware Configuration
The information processing device 100 according to the embodiment described above is realized by, for example, a computer 1000 configured as shown in Fig. 28. Fig. 28 is a hardware configuration diagram showing an example of a computer that realizes the functions of the information processing device 100. The computer 1000 includes a CPU 1100, a RAM 1200, a ROM 1300, a HDD 1400, a communication interface (I/F) 1500, an input/output interface (I/F) 1600, and a media interface (I/F) 1700.

The CPU 1100 operates based on programs stored in the ROM 1300 or HDD 1400, and controls each component. The ROM 1300 stores a boot program executed by the CPU 1100 when the computer 1000 starts up, as well as programs that depend on the computer 1000's hardware.

HDD 1400 stores programs executed by CPU 1100 and data used by such programs. Communication interface 1500 receives data from other devices via a specified communication network and sends it to CPU 1100, and transmits data generated by CPU 1100 to other devices via a specified communication network.

The CPU 1100 controls output devices such as displays and printers, and input devices such as keyboards and mice, via the input/output interface 1600. The CPU 1100 acquires data from input devices via the input/output interface 1600. The CPU 1100 also outputs generated data to output devices via the input/output interface 1600.

Media interface 1700 reads programs or data stored on recording medium 1800 and provides them to CPU 1100 via RAM 1200. CPU 1100 loads the programs from recording medium 1800 onto RAM 1200 via media interface 1700 and executes the loaded programs. Recording medium 1800 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or a PD (Phase Change Rewritable Disc), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory.

For example, when the computer 1000 functions as the information processing device 100 according to the embodiment, the CPU 1100 of the computer 1000 executes programs loaded onto the RAM 1200 to realize the functions of the control unit 130. The CPU 1100 of the computer 1000 reads and executes these programs from the recording medium 1800, but as another example, the CPU 1100 may obtain these programs from another device via a specified communications network.

The above describes in detail some of the embodiments of the present application based on the drawings, but these are merely examples, and the present invention can be implemented in other forms that incorporate various modifications and improvements based on the knowledge of those skilled in the art, including the aspects described in the Disclosure of the Invention section.

[7. Other]
Furthermore, among the processes described in the above embodiments and modifications, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically using known methods. In addition, the information including the processing procedures, specific names, various data, and parameters shown in the above documents and drawings can be changed as desired unless otherwise specified. For example, the various information shown in each drawing is not limited to the information shown in the drawings.

Furthermore, the components of each device shown in the figure are functional concepts and do not necessarily have to be physically configured as shown. In other words, the specific form of distribution and integration of each device is not limited to that shown, and all or part of them can be functionally or physically distributed and integrated in any unit depending on various loads, usage conditions, etc.

Furthermore, the above-described embodiments and variations can be combined as appropriate to the extent that the processing content is not contradictory.

REFERENCE SIGNS LIST 1 Information processing system 10 Terminal device 100 Information processing device 110 Communication unit 120 Storage unit 121 TMF information storage unit 130 Control unit 131 Reception unit 132 Acquisition unit 133 Generation unit 134 Output control unit

Claims (6)

a receiving unit that receives answer information indicating an answer from the user to a question for identifying the desired data included in hierarchically structured data including a data hierarchy including, as elements, each of a plurality of data including the desired data desired by the user, a category hierarchy including, as elements, each of a plurality of categories into which the plurality of data are classified, and a higher category hierarchy including, as elements, each of a plurality of higher categories into which the plurality of categories are classified;
an acquisition unit that acquires hierarchical information regarding a plurality of elements included in one hierarchical level in the hierarchical structure;
a generation unit that inputs a prompt including the answer information and the hierarchical information into a generative model and causes the generative model to generate candidate information regarding the plurality of candidate elements, the candidate elements being candidates for an element corresponding to the answer information, among the plurality of elements, the candidate information including a degree of similarity between the answer information and each of the plurality of candidate elements ;
an output control unit that outputs the candidate information regarding the plurality of candidate elements, the candidate information including a similarity between each of the plurality of candidate elements and the answer information ;
An information processing device comprising: The output control unit
outputting the candidate information for a plurality of the candidate elements;
The reception unit
further receiving selection information regarding a selected element that is the candidate element selected by the user from among the plurality of candidate elements;
The acquisition unit
acquiring, as the hierarchical information, child element information relating to a plurality of child elements having the selected element as a parent element based on the selection information;
The generation unit
inputting the prompt including the child element information into the generative model, and causing the generative model to generate the candidate information regarding the candidate element from among the plurality of child elements;
The information processing device according to claim 1 . The generation unit
causing the generative model to calculate a similarity between the answer information and the hierarchical information, and causing the generative model to generate the candidate information regarding the candidate element selected from the plurality of elements based on the similarity;
The information processing device according to claim 1 . The generative model is a large language model (LLM).
The information processing device according to claim 1 . An information processing method realized by a program executed by an information processing device,
a receiving step of receiving answer information indicating an answer of the user to a question for identifying the desired data included in hierarchically structured data including a data hierarchy including, as elements, each of a plurality of data including the desired data desired by the user, a category hierarchy including, as elements, each of a plurality of categories into which the plurality of data are classified, and a higher category hierarchy including, as elements, each of a plurality of higher categories into which the plurality of categories are classified;
an acquisition step of acquiring hierarchical information regarding a plurality of elements included in one hierarchical level in the hierarchical structure;
a generation step of inputting a prompt including the answer information and the hierarchical information into a generative model, and causing the generative model to generate candidate information regarding the plurality of candidate elements, the candidate elements being candidates for the element corresponding to the answer information, among the plurality of elements, the candidate information including a degree of similarity between the answer information and each of the plurality of candidate elements ;
an output control step of outputting the candidate information regarding the plurality of candidate elements, the candidate information including a similarity between each of the plurality of candidate elements and the answer information ;
An information processing method including: a receiving step of receiving answer information indicating an answer of the user to a question for identifying the desired data included in hierarchically structured data including a data hierarchy including, as elements, each of a plurality of data including the desired data desired by the user, a category hierarchy including, as elements, each of a plurality of categories into which the plurality of data are classified, and a higher category hierarchy including, as elements, each of a plurality of higher categories into which the plurality of categories are classified;
an acquisition step of acquiring hierarchical information regarding a plurality of elements included in one hierarchical level in the hierarchical structure;
a generation step of inputting a prompt including the answer information and the hierarchical information into a generative model, and causing the generative model to generate candidate information regarding the plurality of candidate elements, the candidate elements being candidates for the element corresponding to the answer information, among the plurality of elements, the candidate information including a degree of similarity between the answer information and each of the plurality of candidate elements ;
an output control step of outputting the candidate information regarding the plurality of candidate elements, the candidate information including a similarity between each of the plurality of candidate elements and the answer information ;
An information processing program that causes a computer to execute the above.