Information Processing Apparatus, Information Processing Method, and Information Processing Program

Abstract

Search using a learned AI model infers an answer to a question called a prompt and outputs the inference result as a search result. Therefore, there is a problem that individual search conditions cannot be modified or strict search conditions cannot be ensured, unlike search by a search engine that obtains search results by entering keywords in the search box for each item. 【Solution means】 In order to solve the above problems, there is provided an information processing apparatus or the like having a search instruction setting unit that sets a search instruction in natural language, and a search condition generation unit that inputs the set search instruction into a learned AI model to generate corresponding individual search conditions.

Description

Technical Field

[0001] The present invention relates to an information processing apparatus for retrieving information.

Background Art

[0002] In recent years, the use of large language models (hereinafter sometimes abbreviated as LLM) has been spreading. An LLM can execute various natural language processing tasks such as text classification, information extraction, text summarization, text generation, and question answering. For example, Patent Document 1 discloses a technique for extracting predetermined information from various printed materials such as advertising flyers using an LLM.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] As in the technique of Patent Document 1, by giving a prompt instructing the object to be extracted to the LLM, predetermined information is answered as an extraction result. The prompt is in natural language. For example, by giving a sentence such as "Please show the 1 / 10000 topographic map of Setagaya Ward." to the LLM, a topographic map meeting the conditions is returned.

[0005] The search using the LLM as described above infers the answer to the question of the prompt and outputs the inference result as a search result. Therefore, there is a problem that it is impossible to modify the search conditions or ensure strict search conditions like the search by a search engine that obtains search results by inputting keywords in a search box for each item.

Means for Solving the Problems

[0006] Therefore, in the present invention, in order to solve the above problems, the following information processing apparatuses and the like are provided. That is, an information processing apparatus having a search instruction setting unit that sets a search instruction in natural language, and a search condition generation unit that inputs the set search instruction into a learned AI model to generate corresponding individual search conditions is provided.

[0007] In addition to the above features, an information processing apparatus further having a search condition setting unit that sets the generated search conditions on a UI screen is provided.

[0008] In addition to the above features, an information processing apparatus further having a corresponding location display unit that displays the corresponding location in the set search instruction for the set individual search conditions is provided.

[0009] In addition to the above features, an information processing apparatus further having a cursor movement unit that moves the cursor to the area on the UI screen where the search condition corresponding to the corresponding location is set by an instruction operation on the corresponding location to be displayed is provided.

[0010] In addition, an information processing method executed by an information processing apparatus, the method having a search instruction setting step of setting a search instruction in natural language, and a search condition generation step of inputting the set search instruction into a learned AI model to generate corresponding individual search conditions is provided.

[0011] In addition, an information processing program for causing an information processing apparatus to execute a search instruction setting step of setting a search instruction in natural language and a search condition setting step of inputting the set search instruction into a learned AI model to generate corresponding individual search conditions is provided.

Advantages of the Invention

[0012] According to the present invention, it is possible to provide an information processing apparatus that enables search instructions in natural language while allowing modification of search conditions and ensuring strict search conditions.

Brief Description of the Drawings

[0013]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Mode for Carrying Out the Invention

[0014] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. It should be noted that the present invention should not be limited to these embodiments, and can be implemented in various modes without departing from the gist thereof.

[0015] <Example> <Overview> The present invention is characterized in that, when searching for information for various sources, corresponding individual search conditions are generated from a search instruction set in natural language. Thereby, it is possible to modify the generated search conditions and the like.

[0016] Hereinafter, the functions and processing flow of the information processing apparatus, as well as the details of the hardware, will be described. Note that the functional blocks of the present system described below can be realized as a combination of hardware and software. Specifically, for those using a computer, it includes hardware components such as a CPU (Central Processing Unit), main memory, bus, or secondary storage device (hard disk drive, non-volatile memory, storage media such as CDs and DVDs, and their reading drives), input devices used for information input, printing devices, display devices, and other external peripheral devices, as well as interfaces for those external peripheral devices, communication interfaces, driver programs for controlling those hardware, other application programs, and user interface applications. Then, through the arithmetic processing of the CPU according to the program expanded on the main memory, data input from input devices or other interfaces and held in the memory or hard disk is processed and stored, or instructions for controlling the above-mentioned various hardware and software are generated. Alternatively, the functional blocks of the present system may be realized by dedicated hardware.

[0017] In addition, this invention can be realized not only as a system but also as a method. Also, a part of such an invention can be configured as software. Furthermore, the program used to cause a computer to execute such software, and the recording medium on which the program is fixed are naturally included in the technical scope of this invention (the same applies throughout this specification).

[0018] <Functional Configuration> FIG. 1 is a block diagram showing an example of the functional configuration of the information processing apparatus according to this embodiment. As shown in FIG. 1, the information processing apparatus 100 includes a search instruction setting unit 101, a search condition generation unit 102, a search condition setting unit 103, a corresponding location display unit 104, and a cursor movement unit 105.

[0019] In this embodiment, the pre-trained AI model is a machine learning model that has been pre-trained to handle general tasks. Although large language models are well-known as representative ones, they may be different ones such as small language models or multi-modal language models that can handle images and other formats.

[0020] <Search instruction setting unit> The search instruction setting unit 101 has a function of setting a search instruction in natural language. A search instruction in natural language is a linguistic expression that identifies the object to be searched or requests to present the object, assuming the spoken language commonly used by people. This is assumed for inputting a prompt (dialogue-type instruction or question) to the pre-trained AI model when using the pre-trained AI model for search.

[0021] The setting of the search instruction may be performed, for example, by the user inputting a sentence indicating the search target in an input box, or by voice input based on the user's utterance. Also, the information processing apparatus may present a predetermined search instruction to the user and configure it so that the user modifies the presented search instruction and then sets it as the search instruction.

[0022] <Search condition generation unit> The search condition generation unit 102 has a function of inputting the set search instruction into the pre-trained AI model and generating corresponding individual search conditions. A search condition is a condition specified when performing a search. For example, when searching for a restaurant, conditions specifying search items such as "location or station name", "genre", and "scene" become search conditions.

[0023] For example, when the set search instruction is something like "Tell me a restaurant where I can have an Italian banquet near Shibuya Station.", by inputting this search instruction into the learned AI model, the individual search conditions included in the search instruction are inferred to generate search conditions such as "Location or station name = Shibuya Station", "Genre = Italian cuisine", and "Scene = Banquet". Also, by inputting a search instruction like "Show me the 1:10,000 topographic map of Setagaya Ward." into the learned AI model, search conditions such as "Region = Setagaya Ward", "Scale = 1:10,000", and "Type = Topographic map" are generated.

[0024] In addition, the generation of search conditions may not only be directly generated from the set search instruction, but also by selecting appropriate search conditions from the search conditions preset in the source (database, server, etc.) to be searched as the generation result. For example, in a database of bridges, when search items such as "Bridge type classification", "Bridge form", "Road surface form", "Slab form", and "Erection method" are preset, the corresponding search items are selected from the set search instruction to generate the search conditions.

[0025] <Search condition setting section> The search condition setting unit 103 has a function of setting the generated search conditions on the UI screen. FIG. 2 is a conceptual diagram showing an example of displaying a search instruction and search conditions on the UI screen. FIG. 2(a) shows the situation when a search instruction is input. As shown in the figure, on the UI (User Interface) screen 201, there is a field 202 for inputting a search instruction, and the text "What map are you looking for?" is displayed as a guide.

[0026] And at this point, a search instruction "Please show me the 1:10,000 topographic map of Setagaya Ward." is entered in this field. Also, a search condition field 203 for entering search conditions is displayed on the left side of the UI screen. In this example, search condition fields for each of the preset search items of "Region", "Type", "Scale", and "Display Items" are displayed. Further below, a clear button 204 for clearing the search conditions and a search button 205 for searching based on the set search conditions are displayed.

[0027] When the input of the search instruction is confirmed, as shown in Fig. 2(b), the input of each search condition field is automatically performed, and search conditions of "Setagaya Ward" for "Region", "Topographic Map" for "Type", "1 / 10,000" for "Scale", and "Not Specified" for "Display Items" are displayed in 203. This is because the learned AI model generates search conditions corresponding to the preset search items from the input search instruction "Please show me the 1:10,000 topographic map of Setagaya Ward.", and the search condition setting section displays those search conditions on the UI screen. Also, the user can delete or modify the displayed search conditions.

[0028] By entering the search instruction in this way, when the individual search conditions are displayed on the UI screen, it becomes clear how the search instruction is reflected in the search conditions, and the search conditions can be modified, etc. Also, by setting the original search instruction again, the search conditions can be generated again.

[0029] <Relevant Location Display Section> The relevant location display section 104 has a function of displaying the relevant location in the input search instruction for each of the set individual search conditions. Thereby, the correspondence between each search condition and the relevant location in the search instruction becomes clear.

[0030] Figure 3 is a conceptual diagram for explaining the relevant location display section and the cursor movement section. As shown in Fig. 3(a), in the input field 302 for the search instruction on the UI screen 301, a search instruction "Please show the 1:10,000 topographic map of Setagaya Ward." is displayed. And each of the words "Setagaya Ward", "1:10,000", and "topographic map" is enclosed in square brackets and underlined. These words are search conditions for "region", "type", and "scale" respectively. By displaying in this way, it can be shown which word in the search instruction in natural language is used as a search condition.

[0031] Furthermore, it can also be configured to clearly show which word in the search instruction is used for which search condition. As shown in Fig. 3(b), among the search instruction "Please show the 1:10,000 topographic map of Setagaya Ward." displayed in the input field 302 for the search instruction on the UI screen 301, by pointing to the word "Setagaya Ward" with the cursor 304, both the word "Setagaya Ward" and the word "Setagaya Ward" of the "region" search condition in the corresponding search condition field 305 are highlighted with black and white inversion. Thereby, the correspondence between each search condition and the relevant location in the search instruction becomes clear. Note that various highlighting displays can be made, not limited to black and white inversion.

[0032] Conversely, by indicating any search condition in the search condition field 303 with a cursor or the like, it can also be configured such that the relevant location in the search instruction is highlighted. Furthermore, voice input may also be used.

[0033] <Cursor movement section> The cursor movement section 105 has a function of moving the cursor to the area on the UI screen where the search condition corresponding to the relevant location is set by an instruction operation on the displayed relevant location.

[0034] For example, as shown in FIG. 3(b), when the cursor 304 points to the word "Setagaya Ward" in the search instruction, by performing an operation such as a selection or execution operation like a left click, the cursor 304 is moved to the "Region" field of the search condition field 303. The cursor can be moved while maintaining the arrow-shaped mouse cursor mode as shown in the figure, or the mode can be changed to a text cursor and placed at the end of the word "Setagaya" within the "Region" field, and then moved so that text input becomes possible immediately.

[0035] By configuring it so that such cursor movement is possible, the back-and-forth between the search condition and the words in the search instruction used for that search condition can be easily and quickly performed, and the review and modification of the search condition can be performed quickly and reliably.

[0036] <Hardware Configuration> FIG. 4 is a conceptual diagram showing a configuration example of the hardware that realizes the information processing apparatus according to the embodiment. As shown in the figure, the information processing apparatus 400 includes a CPU 401 that performs various arithmetic processes, a RAM 402 that is a volatile recording medium, a storage 403 such as a flash memory or an HDD that is a non-volatile storage medium, a communication interface 404, and an input / output interface 405. The RAM 402 reads out a program for causing the CPU 401 to perform various arithmetic processes and provides a work area (working area) for that program. Also, a plurality of addresses are assigned to the RAM 402, and the programs executed by the CPU 401 can exchange data with each other by specifying and accessing those addresses and perform processing (the same applies throughout this specification).

[0037] Here, each function of the search instruction setting unit 101, the search condition generation unit 102, and the search condition setting unit 103 of the information processing apparatus 100 in FIG. 1 is mainly realized by the CPU 401 and the RAM 402 in FIG. 4. Further, the functions of the corresponding location display unit 104 and the cursor movement unit 105 are mainly realized by the CPU 401, the RAM 402, and the input / output interface 405 in FIG. 4. Further, when the search condition generation unit 102 uses a learned AI model existing outside, each function is realized by exchanging signals and information with each other via the communication interface 404 and the input / output interface 405.

[0038] <Flow of processing> FIG. 5 is a flowchart briefly showing an example of the flow of processing of the information processing apparatus according to the first embodiment. First, a search instruction in natural language is set (S501: search instruction setting step). Then, the set search instruction is input to a learned AI model to generate corresponding individual search conditions (S502: search condition generation step). Then, the generated search conditions are set on the UI screen (S503: search condition setting step). Then, for the set individual search conditions, the corresponding location in the set search instruction is displayed (S504: corresponding location display step). Then, by an instruction operation on the displayed corresponding location, the cursor is moved to the area on the UI screen where the search condition corresponding to the corresponding location is set (S505: cursor movement step).

[0039] <Effect> According to the information processing apparatus of the present embodiment, it is possible to correct each individual search condition generated from a search instruction in natural language.

Description of reference numerals

[0040] 100, 400: Information processing apparatus 101: Search instruction setting unit 102: Search condition generation unit 103: Search condition setting unit 104: Corresponding location display unit 105: Cursor movement unit 401: CPU 402: RAM 403: Storage 404: Communication Interface 405: Input / Output Interface

Claims

1. A search instruction setting unit that sets search instructions in a natural language input by a user; A search condition generation unit that inputs the set search instructions into a trained AI model to generate corresponding individual search conditions; a search condition setting unit that displays the generated search conditions on a UI screen; a corresponding part display unit that displays a corresponding part in the set search instruction for each of the displayed search conditions on the UI screen; a cursor moving unit that moves a cursor to an area on the UI screen where a search condition corresponding to a relevant portion is displayed, in response to an instruction operation on the displayed relevant portion; An information processing device having the above configuration.

2. An information processing method executed by an information processing device, a search instruction setting step of setting a search instruction in a natural language input by a user; A search condition generation step of inputting the set search instructions into a trained AI model to generate corresponding individual search conditions; a search condition setting step of displaying the generated search conditions on a UI screen; a corresponding part display step of displaying a corresponding part in the set search instruction for each of the displayed search conditions on the UI screen; a cursor moving step of moving a cursor to an area on the UI screen where a search condition corresponding to the relevant portion is displayed, in response to an instruction operation on the displayed relevant portion; An information processing method comprising the steps of:

3. A search instruction setting step of setting a search instruction in a natural language input by a user; A search condition generation step of inputting the set search instructions into a trained AI model to generate corresponding individual search conditions; a search condition setting step of displaying the generated search conditions on a UI screen; a corresponding part display step of displaying a corresponding part in the set search instruction for each of the displayed search conditions on the UI screen; a cursor moving step of moving a cursor to an area on the UI screen where a search condition corresponding to the relevant portion is displayed, in response to an instruction operation on the displayed relevant portion; An information processing program that causes an information processing device to execute the above.

Images