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

Abstract

Provided is, for example, an information processing device comprising: an input value reception unit that receives an input value for each item; a prompt setting unit that sets, for each item, a prompt for generating a similar value of the received input value for the item; and a similar value generation unit that inputs the received input value and the prompt set for the item to a trained AI model and generates a similar value of the received input value.

Description

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

[0001] The present invention relates to an information processing apparatus, an information processing method, and an information processing program that assist in inputting characters, numerical values, and the like.

[0002] When inputting keywords to search for desired information, not only when the input word is an official name, but also when its abbreviation is input, the intended search is performed. This is because, as disclosed in Japanese Patent Application Laid-Open No. 2021-189694 and the like, a thesaurus is prepared in advance and configured to accept synonyms of the input word as input keywords by referring to this thesaurus. By providing such a thesaurus, input assistance is realized.

[0003] However, it is unrealistic to register all synonyms in the dictionary, and it is impossible to respond to the occurrence of new synonyms that were not anticipated (such as a newly established facility name and its abbreviation), so it is insufficient as input assistance.

[0004] Therefore, in order to solve the above problems, the present invention provides the following information processing apparatus and the like. That is, an input value reception unit that receives an input value for each item, a prompt setting unit that sets, for each item, a prompt for generating a similar value of the input value received in the item, and a similar value generation unit that inputs the received input value and the set prompt for that item into a learned AI model to generate a similar value of the received input value. An information processing apparatus having the above is provided.

[0005] In addition to the above features, the present invention provides an information processing apparatus in which the similar value generation unit further inputs synonyms of the received input value, which are referred to from a thesaurus, into the learned AI model to generate similar values.

[0006] In addition to the above features, the present invention provides an information processing apparatus in which the input value is an electronic object other than a symbol.

[0007] In addition to the features described above, the prompt setting unit sets prompts that are stored for each item that accepts input values, providing an information processing device.

[0008] Furthermore, in addition to the above features, the present invention provides an information processing device that further includes a processing unit that performs predetermined processing on the generated similar value and its original input value as the setting value for that item.

[0009] In addition to the features described above, the processing unit also provides an information processing device that performs a search using the set value as a search condition.

[0010] Furthermore, the present invention provides an information processing method executed by an information processing device, comprising: an input value reception step of receiving input values ​​for each item; a prompt setting step of setting a prompt for each item to generate a similar value to the input value received for that item; and a similar value generation step of inputting the received input value and the prompt set for that item to a trained AI model to generate a similar value to the received input value.

[0011] Furthermore, the present invention provides an information processing program that causes an information processing device to execute the following steps: an input value reception step that receives input values ​​for each item; a prompt setting step that sets a prompt for each item to generate similar values ​​to the input values ​​received for that item; and a similar value generation step that inputs the received input values ​​and the prompts set for that item into a trained AI model to generate similar values ​​to the received input values.

[0012] The present invention provides an information processing device that offers superior input assistance compared to the prior art.

[0013] A block diagram showing an example of the functional configuration of the information processing device of the embodiment. A conceptual diagram illustrating a database using the embodiment. A conceptual diagram showing the generated similar values ​​in the database illustrated in Figure 2. A conceptual diagram showing the results of a search process in the database illustrated in Figure 2. A conceptual diagram showing an example of the hardware configuration for realizing the information processing device of the embodiment. A flowchart showing a simplified example of the processing flow of the information processing device of the embodiment.

[0014] Embodiments of the present invention will be described below with reference to the accompanying drawings. However, the present invention is not limited in any way to these embodiments, and can be implemented in various ways without departing from its essence.

[0015] <Examples> <Summary> The present invention sets a prompt for each item that accepts input such as characters to generate a value similar to the received input value, and the synonyms generated by that prompt are also used in processing after being received as input value. This eliminates the need to have a dictionary. Furthermore, for example, if the received input value is used for a search, the intended search can be performed without search omissions or irrelevant search results.

[0016] The following describes the functions and processing flow of the information processing device, as well as the hardware components. The functional blocks of this system described below can be implemented as a combination of hardware and software. Specifically, if a computer is used, these may include hardware components such as a CPU (Central Processing Unit), main memory, a bus, or secondary storage devices (hard disk drives, non-volatile memory, storage media such as CDs and DVDs, and their readers), input devices used for information input, printing equipment, display devices, and other external peripheral devices, as well as interfaces for these external peripheral devices, communication interfaces, driver programs and other application programs for controlling the hardware, and user interface applications. The CPU's arithmetic processing, based on programs deployed in main memory, processes and stores data input from input devices and other interfaces, and stored in memory and on the hard disk, and generates instructions for controlling the aforementioned hardware and software. Alternatively, the functional blocks of this system may be implemented using dedicated hardware.

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

[0018] <Functional Configuration> Figure 1 is a block diagram showing an example of the functional configuration of the information processing device of the embodiment. As shown in Figure 1, the information processing device 100 has an input value receiving unit 101, a prompt setting unit 102, a similar value generation unit 103, and a processing unit 104.

[0019] In this embodiment, the trained AI model is a machine learning model that has been pre-trained and is capable of handling general-purpose tasks. A well-known example is the Language Language Model (LLM), but it may also be a different model, such as a small-scale language model or a multimodal language model that can handle images and other formats.

[0020] <Input Value Reception Unit> The input value reception unit 101 has the function of receiving input values ​​for each item. An item is each individual item that indicates the content of something. For example, if the items are for searching for literature, there may be various items such as "field," "author name," "publication year," and "publisher." If the items are for searching for similar images, they may be items such as "target image," "target name," and "similar range." Database columns and variables in programs are also items.

[0021] The input values ​​for each item are the content of the items mentioned above. For example, in a literature search, the input values ​​for the "Field" field would be the various field names such as "Humanities" or "Nutrition." In the case of building a building database, the input value for the "Exterior" field would be the uploaded image of the building's exterior.

[0022] Thus, input values ​​include not only symbols (letters, numbers), but also various electronic objects that can be created, processed, and stored in a digital environment, such as images, diagrams, graphs, BIM (Building Information Modeling), 2D CAD (Computer Aided Design), 3D CAD, 3D models, and audio.

[0023] Input values ​​can be accepted in various ways. For example, users can input text, upload images, specify strings written in a Word file, specify cells in an Excel file, accept pre-registered input values ​​based on some trigger, or accept values ​​extracted from a pre-registered file based on some trigger. More specifically, a user click could be used as a trigger to extract the "building name" from a pre-registered drawing and use that as the input value. In addition to user input on a UI (User Interface) screen, variables passed via API (Application Programming Interface) or other scripts can also be accepted as input values.

[0024] Figure 2 is a conceptual diagram illustrating a database using this embodiment. In the illustrated "High-rise Building Database" 201, the search items 202 include "Name," "Nearest Station," "Height," and "Completion Date," and accept text and numerical input values. When the search button 204 is pressed, a search process is performed based on the accepted search conditions, and the search results 205 are displayed. In addition, in the "Image" item 203, an image can be uploaded as input value, and a process is performed to search for buildings that correspond to the buildings included (pictured) in that image. Furthermore, input text and images may be used in a single search process.

[0025] <Prompt Setting Unit> The prompt setting unit 102 has the function of setting a prompt for each of the above items to generate a similar value to the input value received in the above item. A similar value is a similar value, but like the input value, it includes not only similar characters or numbers but also similar objects such as the images mentioned above.

[0026] Similar values ​​are those that are identical or similar in meaning, appearance, etc., and include synonyms and synonyms that differ in their form of expression (e.g., Gregorian calendar vs. Japanese calendar, different units). Abbreviations of official names and the official names of abbreviations are also considered similar values. Furthermore, in the case of images, BIM, and 3D models, those in which some of the included components have been replaced with others may also be considered similar values. For example, a building that is the same but whose flooring material has been replaced with a different material may be considered a similar value.

[0027] For example, the prompt for generating similar values ​​in the "Name" input field of the "High-Rise Building Database" shown in Figure 2 is set as follows: "I will now provide you with the name of a building. This name may be an abbreviated version, so in that case, please guess the full, unabbreviated name of the building and provide multiple answers. Conversely, if it is possible to create an abbreviation by abbreviating the name, please guess the abbreviation and provide multiple answers."

[0028] Furthermore, the prompt for generating similar values ​​in the "Image" input field in Figure 2 is set to: "I will now provide you with an image of a building. Please generate or search for images that are similar in shape to the building, rather than in color or surrounding objects, and provide multiple responses."

[0029] The prompts for each of the above items can be entered and set by the user, or they can be pre-configured. Furthermore, the system can be configured to accept and set modifications to already configured prompts.

[0030] <Similar Value Generation Unit> The similar value generation unit 103 has the function of inputting the received input value and the prompt set for that item into the trained AI model and generating a similar value to the received input value.

[0031] Figure 3 is a conceptual diagram showing the similar values ​​generated in the database exemplified in Figure 2. As shown in the diagram, when "Maru Building" is entered in the "Name" field 301, the answer including the similar values ​​generated using "Maru Building" as the input value is shown in the callout 302. Here, by inputting "Maru Building" along with the prompt exemplified in the prompt setting section into the trained AI model, the following answer is returned: "'Maru Building' is an abbreviation for a famous building in Marunouchi, Tokyo. The official name is "Marunouchi Building". However, other buildings with similar names are also possible, so here are some guesses below. Guessions without abbreviation 1. Marunouchi Building 2. Marunouchi Building 3. Marunouchi Center Building 4. Marunouchi Tower Building 5. Marunouchi Plaza Building Examples after abbreviation 1. Maru Building 2. Marunaka Building 3. Marusen Building 4. Maru Tower" This shows similar building names.

[0032] The generated similar values ​​can be displayed in various ways, not just using the example callout. Furthermore, the generated similar values ​​may be displayed as exemplified, or they may be processed without being displayed. Alternatively, the system may be configured to display the generated similar values ​​after processing.

[0033] <Processing Unit> The processing unit 104 has the function of performing predetermined processing on the generated similar value and the original input value as the setting value for that item.

[0034] Figure 4 is a conceptual diagram of the search process performed in the database exemplified in Figure 2. In this "high-rise building database" 401, in the example of each field 402 for entering search conditions, the following is entered for each item 402: "Name" is "International Tower Building", "Nearest Station" is "Shinjuku", "Height" is "80-100m", and "Completion Date" is "July 2003".

[0035] Here, the similar values ​​generated from the input values ​​for each item, for example, "International Tower Building," which are similar to "International Tower," "International Building," and "International Tower Building," along with the original input value "International Tower Building," are used as the setting values ​​for the "Name" item. For other items, the input value and similar values ​​are used as the setting values ​​for each item. Then, by accepting an operation input such as clicking the search button 403, a search process is performed using each setting value as the search condition, and the building image 404 and property description 405 are displayed as the search results.

[0036] Alternatively, the system may take an image of a toilet illustration as input, generate similar toilet illustrations as input values, perform image matching based on these generated similar illustrations, and count the number of toilets in the drawing.

[0037] Furthermore, in some cases, the processing unit generates similar values ​​using the provided data as input in an intermediate process in which the user is not directly involved, and then uses these generated similar values ​​for subsequent processing. It is also possible to configure the system to display the generated similar values ​​after processing to show the reason for the processing result. Additionally, if multiple similar values ​​are generated, the system can be configured to allow the user to select and discard similar values ​​to determine the setting.

[0038] <Hardware Configuration> Figure 5 is a conceptual diagram showing an example of the hardware configuration for realizing the information processing device of the embodiment. As shown in the figure, the information processing device 500 has a CPU 501 that performs various calculations, a RAM 502 which is a volatile recording medium, a storage 503 which is a non-volatile storage medium such as flash memory or HDD, a communication interface 504, and an input / output interface 505. The RAM 502 reads programs that perform various calculations for the CPU 501 to execute and provides a work area (work area) for those programs. In addition, multiple addresses are assigned to the RAM 502, and programs executed by the CPU 501 can exchange data with each other and perform processing by identifying and accessing these addresses.

[0039] Here, the functions of the information processing device 100 in Figure 1—the input value receiving unit 101, the prompt setting unit 102, the similar value generation unit 103, and the processing unit 104—are mainly realized by the CPU 501, RAM 502, and input / output interface 505 in Figure 5. Furthermore, when using an externally existing pre-trained AI model, each function is realized by exchanging signals and information with each other via the communication interface 504 and the input / output interface 505.

[0040] <Processing Flow> Figure 6 is a simplified flowchart showing an example of the processing flow of the information processing device of the embodiment. First, input values ​​are received for each item (S601: Input Value Reception Step). Then, a prompt for generating a similar value to the input value received for that item is set for each item (S602: Prompt Setting Step). Then, the received input value and the prompt set for that item are input to the trained AI model to generate a similar value to the received input value (S603: Similar Value Generation Step). Then, the generated similar value and its original input value are used as the set value for that item and a predetermined process is performed (S604: Processing Step).

[0041] <Effects> According to this embodiment, an information processing device can be provided that offers superior input support compared to the conventional technology.

[0042] 100, 500: Information processing unit 101: Input value receiving unit 102: Prompt setting unit 103: Similar value generation unit 104: Processing unit 501: CPU 502: RAM 503: Storage 504: Communication interface 505: Input / output interface

Claims

1. An information processing device comprising: an input value receiving unit that receives input values ​​for each item; a prompt setting unit that sets a prompt for each item for generating a similar value to the input value received for that item; and a similar value generation unit that inputs the received input value and the prompt set for that item to a trained AI model to generate a similar value to the received input value.

2. The information processing apparatus according to claim 1, wherein the similarity value generation unit refers to a thesaurus and further inputs synonyms of the received input value into the trained AI model to generate similarity values.

3. The information processing apparatus according to claim 1, wherein the input value is an electronic object other than a symbol.

4. The information processing apparatus according to claim 1, wherein the prompt set by the prompt setting unit sets a prompt that has been stored for each item that accepts an input value.

5. The information processing apparatus according to claim 1, further comprising a processing unit that performs predetermined processing on the generated similar value and the original input value as a set value for that item.

6. The information processing apparatus according to claim 5, wherein the processing unit performs a search using the set value as a search condition.

7. An information processing method executed by an information processing device, comprising: an input value reception step of receiving input values ​​for each item; a prompt setting step of setting a prompt for each item to generate similar values ​​to the input values ​​received for that item; and a similar value generation step of inputting the received input values ​​and the prompts set for that item to a trained AI model to generate similar values ​​to the received input values.

8. An information processing program that causes an information processing device to execute: an input value reception step that receives input values ​​for each item; a prompt setting step that sets a prompt for each item to generate similar values ​​to the input values ​​received for that item; and a similar value generation step that inputs the received input values ​​and the prompts set for that item to a trained AI model to generate similar values ​​to the received input values.

Images