Information processing system, information processing method, and program

Abstract

To solve the problem that a file name cannot be attached to a document image when there is no description coinciding with a naming rule in the document image.SOLUTION: An information processing server 103 acquires a character string constituting a sample document image 114 and a file name of the sample document image 114 (S501), acquires a character string constituting a document image 113 (S502 and S503), generates an instruction statement 700 (Fig. 7) for causing a large-scale language model 116 to output a file name for the document image 113, on the basis of the character string constituting the sample document image 114 and the file name of the sample document image 114 (S504 and S505), transmits the instruction statement 700 to the large-scale language model 116 (S506), receives a response 720 from the large-scale language model 116 (S507), and presents the file name extracted from the response (S508).SELECTED DRAWING: Figure 5

Description

[Technical Field]

[0001] The present invention relates to a process for assigning a file name to a document image. [Background technology]

[0002] Conventionally, there is a technique for automatically assigning a file name to file data. File data refers to photographs, illustrations, document images, videos, audio, etc., and it is necessary to assign unique file names to all of these files in the same space. Unique file names are generally assigned automatically based on property information such as the camera's unique value or the date and time of storage, or numbers such as (2) or rev2.

[0003] Additionally, in recent years, the use of machine learning models and large-scale language models (LLMs) has been attracting attention as a method for extracting information from text information such as documents. Large-scale language models, in particular, can extract information using vague instructions. Large-scale language models are language models constructed using large datasets and deep learning technology. Large-scale language models are capable of human-like conversations and can perform various processes using natural language with high accuracy.

[0004] Patent Document 1 discloses a technology for assigning file names by inferring a naming rule from the file names of document images present in a folder. According to Patent Document 1, a user can assign file names simply by placing a document image with a file name in a folder. In other words, the user does not need to specify a naming rule for file names in advance. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Patent No. 6955434 Summary of the Invention [Problem to be solved by the invention]

[0006] However, in the technology of Patent Document 1, although the user does not need to specify a naming rule in advance, because the naming rule is used, if no description matching the naming rule is found in the document image, a file name cannot be assigned. For example, if the items that can be included in the naming rule are "title," "document number," "company name," and "issue date," an item such as "delivery note order number" that exists in the document image cannot be included in the file name. As described above, the conventional technology has a problem in that it is not possible to give a file name if the file name item does not match the character string in the document image.

[0007] The present invention has been made to solve the above-mentioned problems, and aims to provide a mechanism that enables flexible file naming without relying on whether the file name items match the character strings in the document image. [Means for solving the problem]

[0008] The present invention comprises a first acquisition means for acquiring a character string constituting a first document image and a file name for the first document image, a second acquisition means for acquiring a character string constituting a second document image, a generation means for generating an instruction statement for causing a large-scale language model to output a file name for the second document image based on the first document image and the file name for the first document image, a transmission means for transmitting the generated instruction statement to the large-scale language model, a reception means for receiving a response from the large-scale language model to the transmitted instruction statement, and a presentation means for presenting the file name obtained from the received response. [Effects of the Invention]

[0009] According to the present invention, file names can be assigned flexibly regardless of whether the file name items match the character strings in the document image. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram showing an example of the configuration of an information processing system according to an embodiment of the present invention. [Figure 2] FIG. 2 is a diagram showing an example of the configuration of an information processing device and an information processing server. [Figure 3] FIG. 2 is a diagram showing an example of a usage sequence of the information processing system according to the first embodiment. [Figure 4] 10 is a flowchart showing an example of a process for acquiring data of a sample document. [Figure 5] 6 is a flowchart showing an example of a process for proposing a file name of a document image in the first embodiment. [Figure 6A] 10A and 10B are diagrams showing specific examples of a sample document image and a document image to be input. [Figure 6B] 10A and 10B are diagrams showing specific examples of a sample document image and a document image to be input. [Figure 7] 3A and 3B are diagrams showing specific examples of directives and responses to be input to a large-scale language model in the first embodiment. [Figure 8] 4A and 4B are diagrams showing an example of a setting screen and an example of a screen that proposes a file name for an input document image in the first embodiment. [Figure 9] 10 is a flowchart showing an example of a process for proposing a file name of a document image in the second embodiment. [Figure 10] 10A and 10B are diagrams showing specific examples of directives and responses to be input to a large-scale language model in the second embodiment. [Figure 11] 10A and 10B are diagrams showing an example of a setting screen and an example of a screen that proposes a file name for an input document image in the second embodiment. [Figure 12A] FIG. 11 is a diagram showing an example of a usage sequence of an information processing system according to the third embodiment. [Figure 12B] FIG. 11 is a diagram showing an example of a usage sequence of an information processing system according to the third embodiment. [Figure 13] 11 is a flowchart showing an example of a process for proposing a file name of a document image in the third embodiment. [Figure 14]13A and 13B are diagrams showing an example of a screen that proposes a file name for an input document image and an example of a sample document setting screen when an exception is applied in the third embodiment. [Figure 15A] FIG. 13 is a diagram showing an example of a usage sequence of an information processing system according to the fourth embodiment. [Figure 15B] FIG. 13 is a diagram showing an example of a usage sequence of an information processing system according to the fourth embodiment. [Figure 16] 13 is a flowchart showing an example of a process for proposing a file name of a document image in the fourth embodiment. [Figure 17] 13A and 13B are diagrams showing specific examples of directives and responses to be input to a large-scale language model in the fourth embodiment. [Figure 18A] 13A and 13B are diagrams showing an example of a setting screen, an example of a screen that proposes a file name for an input document image, and an example of a sample document setting screen when an exception is applied, according to the fourth embodiment. [Figure 18B] 13A and 13B are diagrams showing an example of a setting screen, an example of a screen that proposes a file name for an input document image, and an example of a sample document setting screen when an exception is applied, according to the fourth embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. Note that the components described in the present embodiments are merely examples and are not intended to limit the scope of the present invention.

[0012] [First embodiment] <Information Processing System> FIG. 1 is a diagram showing an example of the configuration of an information processing system according to an embodiment of the present invention. As shown in FIG. 1, an information processing system 100 includes, for example, an information processing device 101, an information processing server 103, and an external information processing server 105, which are connected to each other via a network 104.

[0013] In the information processing system 100, the information processing device 101, the information processing server 103, and the external information processing server 105 may be configured to be connected to the network 104 in multiple connections rather than in a single connection. For example, the information processing server 103 may be configured to be composed of a first server device having high-speed computing resources and a second server device having large-capacity storage, and the like, which are connected to each other via the network 104.

[0014] The information processing device 101 is realized by an MFP (Multi-Function Peripheral) or the like having multiple functions such as printing, scanning, and faxing, and functions as an image acquisition unit 151 and a display unit 158 ​​.

[0015] The image acquisition unit 151 optically reads, for example, an original 111 printed on a recording medium such as paper, performs predetermined scan image processing to generate a document image 113, and transmits the document image 113 to the information processing server 103. The image acquisition unit 151 also receives, for example, FAX data 112 transmitted from a FAX transmitter (not shown), performs predetermined FAX image processing to generate a document image 113, and transmits the document image 113 to the information processing server 103.

[0016] The information processing device 101 may be configured to be realized by a PC (Personal Computer), a tablet terminal, a smartphone, or the like, in addition to the MFP having the above-mentioned scan and fax functions. Specifically, for example, a document image 113 such as a PDF or JPEG image generated using a document creation application running on a PC serving as the information processing device 101 may be transmitted to the information processing server 103. The image acquisition unit 151 also transmits a sample document image 114, which is used as a sample for naming files, to the information processing server 103 together with the assigned file name.

[0017] The display unit 158 ​​displays on a display the information received from the information processing server 103. The display unit 158 ​​presents to the user, for example, a sample document setting screen 800 as shown in Fig. 8(a) and a file name suggestion screen 820 as shown in Fig. 8(b), which will be described later.

[0018] The information processing server 103 functions as a document image analysis unit 154, an instruction sentence generation unit 155, an output conversion unit 156, and a storage unit 157. The information processing server 103 has a role of receiving the document image 113, the sample document image 114, and their file names as input, and transmitting the processed results to the information processing device 101 via the external information processing server 105. This will be described in detail below.

[0019] First, the document image analysis unit 154 performs OCR processing on the document image 113 received from the information processing device 101 and the sample document image 114, and obtains multiple character strings as the character recognition results obtained by the OCR processing.

[0020] Next, the instruction sentence generation unit 155 applies the character strings contained in the document image 113 and the sample document image 114, and the file name of the sample document image 114 to a prepared instruction sentence template, thereby generating an instruction sentence (also called a "prompt") to be input to the large-scale language model 116. Here, the prepared instruction sentence template may be one prepared as a preset by an engineer or the like, one prepared by the user, or one that has been modified or added to by the system or the user. Thereafter, the information processing server 103 transmits the generated instruction sentence to the external information processing server 105 via the network 104. Furthermore, the information processing server 103 receives a response result to the instruction sentence from the external information processing server 105.

[0021] The output conversion unit 156 converts the above-mentioned response result received from the external information processing server 105 into information to be displayed to the user. The storage unit 157 stores the results of conversion by the output conversion unit 156. Note that the information converted from the above-mentioned answer results stored in the storage unit 157 is transmitted to the information processing device 101 by the function of the information processing server 103, and is displayed on the display of the information processing device 101 by the display unit 158 ​​and presented to the user.

[0022] The network 104 is realized by a LAN, a WAN, or the like, and is a communication unit that connects the information processing device 101, the information processing server 103, and the external information processing server 105 to one another, and transmits and receives data between the devices.

[0023] The external information processing server 105 functions as a large-scale language model 116. The large-scale language model 116 is accessed by an API (Application Programming Interface) via the network 104, and outputs a response result to an instruction received from the information processing server 103. The external information processing server 105 may be configured to exist on a different system of the same vendor, or may be configured to exist on a system of an external vendor. Note that the large-scale language model 116 may be configured to exist within the information processing server 103, or may be configured such that some functions or devices exist within the information processing server 103.

[0024] <Device configuration> FIG. 2 is a diagram showing an example of the configuration of the information processing device 101 and the information processing server 103 for realizing the information processing system 100. As shown in FIG. FIG. 2A is a diagram illustrating an example of the configuration of the information processing device 101. As shown in FIG. 2(a), the information processing device 101 has a CPU 201, a ROM 202, a RAM 204, a printer device 205, a scanner device 206, a document transport device 207, a storage 208, an input device 209, a display device 210, an external interface 211, etc., which are connected to each other via a data bus 203.

[0025] The CPU 201 is a control unit for controlling the overall operation of the information processing device 101. The CPU 201 executes a startup program stored in a ROM 202 to start up the system of the information processing device 101, and executes a control program stored in a storage 208 to realize various functions of the information processing device 101, such as printing, scanning, and faxing, as well as the image acquisition unit 151 and display unit 158 ​​shown in FIG.

[0026] The ROM 202 is realized by a nonvolatile memory, and is a storage unit for storing a boot program for booting the information processing device 101 and the like. The data bus 203 is a communication unit for transmitting and receiving data between devices that make up the information processing device 101 .

[0027] The RAM 204 is realized by a volatile memory, and is a storage unit used as a work memory when the CPU 201 executes a control program. The printer device 205 is an image output device, and is a processing unit for printing and outputting a document image inside the information processing device 101 onto a storage medium such as paper.

[0028] The scanner device 206 is an image input device and a processing unit for optically reading a storage medium such as paper on which characters, diagrams, and the like are printed, and acquiring the result as a document image. The document transport device 207 is realized by an ADF (Auto Document Feeder) or the like, and is a processing unit for detecting documents placed on a document table and transporting the detected documents one by one to the scanner device 206.

[0029] The storage 208 is realized by a hard disk drive (HDD), a solid state drive (SSD), an embedded multimedia card (eMMC), or the like, and is a storage unit for storing the control programs and document images described above.

[0030] The input device 209 is realized by a touch panel, hard keys, or the like, and is a processing unit for receiving operation inputs from a user who uses the information processing device 101 . The display device 210 is realized by a liquid crystal display or the like, and is a display unit for displaying and outputting the setting screen of the information processing device 101 to the user.

[0031] The external interface 211 connects the information processing device 101 to the network 104, and is an interface unit for receiving fax data from a fax transmitter (not shown) and transmitting document images to the information processing server 103.

[0032] FIG. 2B is a diagram illustrating an example of the configuration of the information processing server 103. As shown in FIG. 2( b ), the information processing server 103 includes a CPU 261 , a ROM 262 , a RAM 264 , a storage 265 , an input device 266 , a display device 267 , an external interface 268 , and the like, which are connected to one another via a data bus 263 .

[0033] The CPU 261 is a control unit for controlling the overall operation of the information processing server 103. The CPU 261 starts up the system of the information processing server 103 by executing a boot program stored in the ROM 262, and realizes the various functions shown in Fig. 1 by executing an information processing program stored in the storage 265, and performs information processing such as character recognition (OCR) and file name suggestion.

[0034] The ROM 262 is realized by a nonvolatile memory, and is a storage unit for storing a boot program that starts up the information processing server 103. The data bus 263 is a communication unit for transmitting and receiving data between devices that make up the information processing server 103 . The RAM 264 is realized by a volatile memory, and is a storage unit used as a work memory when the CPU 261 executes an information processing program.

[0035] The storage 265 is realized by an HDD, SSD, eMMC, etc., and is a memory unit for storing the aforementioned information processing program, document image 113, sample document image 114, instruction sentences to be input to the large-scale language model 116, instruction sentence templates, and answer results, etc.

[0036] The input device 266 is realized by a mouse, keyboard, etc., and is a processing unit for accepting operational input to the information processing server 103 from a user who uses the information processing server 103 or an engineer who controls the information processing server 103. The display device 267 is realized by an LCD display or the like, and is a display unit for displaying and outputting the setting screen of the information processing server 103 to a user who uses the information processing server 103 or an engineer who controls the information processing server 103.

[0037] The external interface 268 connects the information processing server 103 and the network 104. The external interface 268 is an interface unit for transmitting an instruction statement to the external information processing server 105, receiving a response to the instruction statement from the external information processing server 105, receiving the document image 113, the sample document image 114, etc. from the information processing device 101, and transmitting information to the information processing device 101.

[0038] <Usage sequence> For example, the information system 100 inputs a document image such as that shown in Fig. 6B(c) described later, proposes a file name using the large-scale language model 116, and presents a file name proposal screen such as that shown in Fig. 8(b) described later to the user on the display unit 158. The usage sequence until the information processing system 100 obtains the result will be described with reference to Fig. 3.

[0039] FIG. 3 is a diagram showing an example of a usage sequence of the information processing system 100. As shown in FIG. 3(a) is a diagram illustrating the flow of advance preparation for proposing a file name using the large-scale language model 116 for a sample document image 114 acquired by the information processing device 101 in accordance with an instruction from a user. The detailed processing flow of FIG. 3(a) will be described later with reference to FIG. 4.

[0040] First, when a user specifies and inputs a sample document image 114 to which a file name has been assigned, which serves as a sample for file naming, to the information processing device 101, the information processing device 101 accepts this input (S301). Note that this input is made from a sample document setting screen, such as that shown in FIG. 8(a) to be described later, displayed on the display unit 158. In response to accepting this input, the information processing device 101 transmits the sample document image 114 and its file name to the information processing server 103 (S302). The information processing server 103 receives the sample document image 114 and its file name. Note that the setting of the sample document in the above steps S301 and S302 may be made not only by the information processing device 101 but also from a web browser of a user terminal (not shown) or the input device 266 of the information processing server 103.

[0041] Next, the information processing server 103 executes OCR processing on the sample document image 114 received in S302, and acquires a group of character strings that are the OCR results (S303). Thereafter, the information processing server 103 stores the group of character strings acquired in S303 and the file name received in S302 as a pair of data in the storage unit 157 (S304).

[0042] 3(b) is a diagram illustrating a flow of proposing a file name for a document image 113 acquired by the information processing device 101 in accordance with an instruction from a user, using the large-scale language model 116. The detailed processing flow of FIG. 3(b) will be described later with reference to FIG.

[0043] First, when a user of the information processing system 100 places an original 111 on the information processing device 101 and issues a scan instruction to the information processing system 100, the information processing device 101 accepts this scan instruction (S311). In response to accepting this scan instruction, the information processing device 101 transmits a document image 113 obtained by scanning the original 111 to the information processing server 103 (S312). The information processing server 103 receives the document image 113.

[0044] Next, the information processing server 103 executes OCR processing on the document image 113 received in S312, and acquires a group of character strings that are the OCR results (S313). Next, the information processing server 103 generates an instruction statement for suggesting a file name for the document image received in S312 above (S314) using the group of character strings acquired in S313 above and the group of character strings and file name saved in S304 above. For example, an instruction statement such as that shown in Fig. 7(a) described later may be generated. Next, the information processing server 103 transmits the instruction generated in S314 to the external information processing server 105 (S315).

[0045] Next, the external information processing server 105 receives the instruction statement transmitted in S315 above, and causes the large-scale language model 116 to generate a response to the received instruction statement. The response result to the instruction statement is acquired by the information processing server 103 (S316). For example, the received response result is as shown in FIG. 7(b) described below. Finally, the information processing server 103 presents the answer result acquired in S316 to the user (S317). For example, this answer result is received from the information processing server 103 by the information processing device 101 and presented on the display unit 158 ​​as a file name suggestion screen shown in Fig. 8(b) to be described later.

[0046] <Getting sample document data> The detailed processing flow of FIG. 3(a) will be described below with reference to FIG. 4 is a flowchart showing an example of a process for acquiring sample document data. Note that an execution program for each step shown in FIG. 4 is stored in the ROM 262, RAM 264, or storage 265 of the information processing server 103, and is executed by the CPU 261 of the information processing server 103.

[0047] First, in S401, the CPU 261 acquires the sample document image 114 input by the user in S301 of Fig. 3 and transmitted to the information processing server 103 in S302. The sample document image 114 is a document image to which a file name has been assigned. Specific examples of the sample document image 114 and the file name will be described below with reference to Figs. 6A(a) and 6B(b). Hereinafter, Figs. 6A and 6B will be collectively referred to as "Fig. 6."

[0048] The document image 600 shown in FIG. 6( a ) is an example of the sample document image 114 , and is an invoice having a title 601 , an issue date 602 , and an invoice number 603 . The file name 610 shown in FIG. 6(b) is an example of a file name for the document image 600. The file name 610 is formed by connecting the title 601, invoice number 603, and issue date 602 written in the document image 600 with underscores. Note that the file name 610 may be partially composed of a character string such as "Issuance Date 20151030," which combines the item name "Issuance Date" and the value of the issue date 602, or may be connected with a character other than an underscore, including an empty character. In the following description, it is assumed that the document image 600 shown in FIG. 6(a) is acquired as the sample document image 114 in S401 of FIG. 4.

[0049] Next, in S402, the CPU 261 executes block selection (BS) processing and OCR processing on the sample document image 114 acquired in S401, and acquires a group of character strings included in the sample document image 114.

[0050] Here, block selection (BS) processing refers to a process of selecting block areas so as to divide document image data into object units, and determining the attributes of each block area. Specifically, this process determines the attributes of, for example, text, photographs, diagrams, etc., and divides the image into block areas with different attributes. This can be achieved using known area determination technology. The character string data obtained as the OCR result may be, for example, word-delimited character strings arranged at intervals or separated by ruled lines within a document image, read out consecutively in a predetermined read order based on the arrangement information.The character string data obtained may also be, for example, word-delimited character strings divided into sentences constituting the document image using a morphological analysis technique, read out consecutively in a predetermined read order based on the arrangement information.

[0051] Next, in S403, the CPU 261 acquires the file name already assigned to the sample document image 114 acquired in S401 (the file name transmitted together with the sample document image 114 in S302 of FIG. 3).

[0052] Finally, in S404, the CPU 261 associates the group of character strings acquired in S402 and the file name acquired in S403 as paired data, and saves them in the storage 265. Note that the CPU 261 realizes the association as paired data in a dictionary format (e.g., JSON (JavaScript Object Notation)) in which, for example, the group of character strings acquired in S402 is associated with an "OCR result" key, and the file name acquired in S403 is associated with a "file name" key. At this time, the CPU 261 may associate the group of character strings acquired in S402 with the file name acquired in S403 as a key, or may realize this using an array or object in which a key is not set. Note that in the following description, the data in which the group of character strings acquired in S402 and the file name acquired in S403 are associated as paired data will be referred to as "sample document data." After the process of S404, the CPU 261 ends the process of this flowchart.

[0053] <File name suggestion process> The detailed processing flow of FIG. 3(b) will be described below with reference to FIG. Fig. 5 is a flowchart showing an example of a process for proposing a file name for a document image in the first embodiment. Note that an execution program for each step shown in Fig. 5 is stored in the ROM 262, RAM 264, or storage 265 of the information processing server 103, and is executed by the CPU 261 of the information processing server 103.

[0054] First, in S501, the CPU 261 acquires, from the storage 265, the sample document data for the sample document image 114 saved in S404 of FIG.

[0055] Next, in S502, the CPU 261 acquires the document image 113 transmitted from the information processing apparatus 101 to the information processing server 103 in S312 of FIG. A specific example of the document image 113 will now be described with reference to FIG. 6(c). 6(c) is an example of document image 113, which is an invoice having a title 651, an issue date 652, and an order number 653. In the following explanation, it is assumed that document image 650 shown in FIG. 6(c) is acquired as document image 113 in S502 above.

[0056] Next, in S503, the CPU 261 executes the block selection (BS) processing and OCR processing described in S402 of FIG.

[0057] Next, in S504, the CPU 261 acquires, from the storage 265, a command statement template that allows a file name to be suggested for the document image acquired in S502. Here, the instruction template will be explained with reference to FIG. 7(a). FIG. 7(a) shows an example of a directive that uses the sample document image 600 and file name 610 in FIG. 6 to suggest a file name for a document image 650. The directive template refers to the entire text of the directive 700, excluding the file name 610 of the sample document image 600, the OCR result 711 for the sample document image 600, and the OCR result 712 for the document image 650. That is, the directive template for the directive 700 is composed of text 701, text 702, and text 703. When switching between multiple sample document images 600, the directive template may include the file name 610 of the corresponding sample document image 600 and the OCR result 711 for the sample document image 600. This makes it possible to switch the directive template in response to switching between sample document images 600. The CPU 261 may switch the content of the directive 700 depending on the language and destination of the input document image. Furthermore, the instruction template may be one prepared in advance by an engineer or a user, or may be one that has been prepared in advance and edited by the user.

[0058] Next, in S505, the CPU 261 generates an instruction statement by applying the sample document data acquired in S501 and the group of character strings acquired in S503 to the instruction statement template acquired in S504. This process will be described with reference to FIG. 7(a). The CPU 261 generates an instruction statement 910 by applying the file name 610 for the sample document image 600, the OCR result 711, and the OCR result 712 for the document image 650 shown in FIG. 6 to the instruction statement template acquired in S504. At this time, the OCR results 711 and 712 may be replaced with text information necessary and sufficient to form the file name. The necessary and sufficient text information does not refer to all the group of character strings included in the sample document image 600, but rather refers to data including, for example, the elements 601 to 603 that make up the file name 610 and only the surrounding areas.

[0059] Next, in S506, the CPU 261 inputs the instruction statement generated in S505 above to the large-scale language model 116. The CPU 261 inputs, for example, an instruction statement 700 as shown in FIG.

[0060] Next, in S507, the CPU 261 acquires the output result for the instruction statement generated in S505 above from the large-scale language model 116. This output result will be explained using FIG. 7(b) is an example of the output result obtained by inputting the instruction statement 700 to the large-scale language model 116. As a result of inputting the instruction statement 700 to the large-scale language model 116, the CPU 261 obtains an answer result 720 including a file name 721 from the large-scale language model 116.

[0061] Finally, in S508, the CPU 261 extracts the file name proposed by the large-scale language model 116 from the answer result acquired in S507, and transmits the extracted file name to the information processing device 101, where the display unit 158 ​​presents it to the user. Note that the file name extracted from the answer result may be output to the display device 267 and presented to the user. An example of a screen on which the file name is presented to the user will be described with reference to FIG. 8(b).

[0062] FIG. 8(b) is an example of a screen that presents to the user the file name proposed by the large-scale language model 116 for the document image 650 in FIG. 6(b). The CPU 261 extracts a proposed file name 722 from the answer result 720 and controls to display it on the proposal screen 820. At this time, the CPU 261 may indicate which file the proposed file name is for by displaying the document image 650 in Fig. 6 alongside the proposed file name 722. The image displayed alongside may be reduced in size from the original image.

[0063] <User interface> FIG. 8 shows an example of a screen that the user operates when determining the file name of the document image 113. In FIG. FIG. 8A shows an example screen in which a user sets the sample document image 114 as a sample for file naming (S301). A specific example of a case in which a user sets the sample document image 600 shown in FIG. 6A as a sample document image will be described below. When the information processing system 100 detects a user's press of a select button 801, it displays a file dialog box that allows the user to select a file from a list of multiple files. Since the file dialog box is implemented using a known method, its description will be omitted. When the information processing system 100 subsequently detects a user's selection of the sample document image 600, it displays a file name 610 of the sample document image 600 in a field 802. At this time, the information processing system 100 may also display a thumbnail or the like of the sample document image 600 in the field 802. When the information processing system 100 subsequently detects a user's press of an OK button 803, it confirms the sample document image 600 as the sample document image to be used as a sample for file naming. If the user presses the select button 801 again, the information processing system 100 displays the file dialog again. At this time, the file list displayed may be the list displayed when the file dialog was last closed, or the list displayed when the file dialog was first opened.

[0064] FIG. 8(b) is an example of a screen on which the user confirms and corrects the file name (S508) for the document image 113 and confirms the file name. A case in which the user corrects the file name proposed as the file name of the document image 650 in FIG. 6 will be specifically described below. When the information processing system 100 detects that the user has pressed the Edit button 821, it changes the display of the file name 722 to an editable state such as a text box, and displays an input cursor 822. The user manually corrects the file name 722 using the input cursor 822. The user can then confirm the manually corrected file name and confirm the file name for the document image 650 by pressing an OK button 823.

[0065] As described above, according to the first embodiment, by inputting a document image that serves as a sample for file naming, a file name can be presented that is not dependent on whether the file name items match the character strings in the document image. For example, consider a case where a file name is assigned to a document image of a "delivery note" that does not include the character string "document number," assuming that it is desired to standardize file names that include the "document number" for all document images. According to the first embodiment, if the document image of the "delivery note" includes the character string "delivery note order number," the character string "delivery note order number" can be included in the file name instead of the character string "document number." In other words, file names can be assigned flexibly without relying on whether the file name items match the character strings in the document image.

[0066] Second Embodiment In the first embodiment, the file name for the document image 113 proposed by the large-scale language model 116 is presented to the user with reference to the sample document image 114. In contrast to this, in the second embodiment, in addition to the file name for the document image 113, the reason why the large-scale language model 116 proposed the file name is also presented to the user. The second embodiment can be realized by replacing Fig. 5 in the first embodiment with Fig. 9, Fig. 6 with Fig. 10, and Fig. 8(b) with Fig. 11(b). Other than this, the second embodiment is similar to the first embodiment, so a description thereof will be omitted.

[0067] <File name suggestion process> Fig. 9 is a flowchart showing an example of a process for proposing a file name for a document image in the second embodiment. Note that an execution program for each step shown in Fig. 9 is stored in the ROM 262, RAM 264, or storage 265 of the information processing server 103, and is executed by the CPU 261 of the information processing server 103. In Fig. 9, the processes having the same reference numerals as those in the flowchart of Fig. 5 are the same as those in the first embodiment, and therefore description thereof will be omitted.

[0068] The CPU 261 executes the processes of S501 to S503, and then proceeds to S904. In S904, the CPU 261 acquires, from the storage 265, an instruction statement template for suggesting a file name such as that shown in Fig. 10(a) for the document image acquired in S502. In the second embodiment, this instruction statement template is used to control so that the reason for the suggestion is also presented along with the file name suggestion.

[0069] Here, the instruction template will be explained with reference to FIG. 10(a) is an example of a directive for displaying not only the file name of the document image 650 but also the reason why the large-scale language model 116 determined that file name. The directive 1000 in FIG. 10(a) is realized by replacing the text 701 in FIG. 7(a) with text 1001. In addition to the directive of text 701, text 1001 includes the directive "Then, please tell me why you named the file that way."

[0070] Next, CPU 261 executes S505 to S507. At this time, the CPU 261 executes the processing by replacing the instruction statement template acquired in S504 of Fig. 5 with the instruction statement template acquired in S904. The answer result obtained by these processes will be described with reference to Fig. 10(b).

[0071] 10(b) is an example of an output of an answer result obtained by inputting the instruction statement 1000 as shown in FIG. 10(a) into the large-scale language model 116. The CPU 261 acquires an answer result 1020 as a result of inputting the instruction statement 1000 into the large-scale language model 116. Note that the answer result 1020 includes a determination reason 1022 that led to the suggestion of the file name 721 in addition to the answer result 720 in FIG. 7(b).

[0072] Hereafter, we return to the explanation of the flowchart in FIG. When the CPU 261 acquires the answer result from the large-scale language model 116 in S507, the process proceeds to S908. In S908, the CPU 261 extracts the file name proposed by the large-scale language model 116 and the reason for its determination from the answer result acquired in S507 above, and transmits the extracted file name to the information processing device 101, where the display unit 158 ​​presents it to the user. Note that the file name extracted from the answer result may be output to the display device 267 and presented to the user. An example of a screen on which the file name is presented to the user will be described with reference to FIG. 11(b).

[0073] FIG. 11(b) is an example of a screen on which the file name proposed by the large-scale language model 116 for the document image 650 in FIG. 6(b) is presented to the user along with the reason for the proposal. The CPU 261 extracts the proposed file name 722 and the reason for determination 1022 from the answer result 1020 as shown in FIG. 10(b), and displays them on the proposal screen 1120. At this time, the CPU 261 displays the reason for determination 1022 on the proposal screen 1120 in a form that makes it clear that it is the "reason for determination," as in notation 1122, for example. The CPU 261 may always display the notation 1122, or may switch whether or not to display it in response to a user operation such as hovering the mouse over the proposed file name 722 or checking a check box. Switching whether or not to display the reason for determination will be described later with reference to FIG. 11(a).

[0074] <User interface> In the second embodiment, the display unit 158 ​​presents to the user, for example, a setting screen 1100 for reason display options as shown in FIG. 11(a) and a file name suggestion screen 820 as shown in FIG. 11(b), which will be described later.

[0075] 11(a) is an example of a setting screen where the user decides whether or not to display the reason for determining the file name of the document image 113. This setting is optional and can be set by the user in advance. The changes in the proposal screen due to this setting will be specifically described below.

[0076] In the information processing system 100 of the second embodiment, when the radio button 1101 is selected and pressing of the button 1110 by the user is detected, the display unit 158 ​​displays the determination reason 1022 on the proposal screen 820. On the other hand, when the radio button 1102 is selected and pressing of the button 1110 by the user is detected, the display unit 158 ​​does not display the determination reason 1022 on the proposal screen 820. That is, in this case, the display unit 158 ​​displays the proposal screen 820 as shown in FIG. 8(b).

[0077] As described above, according to the second embodiment, it is possible to present the reason for the decision that led to the proposal for a file name proposed by a large-scale language model. This not only achieves the effect of the first embodiment, but also allows the user to easily check, from the presented reason for the decision, whether the file name reflects the regularity of the file name that the user intends.

[0078] Third Embodiment In the first embodiment, the sample document image 114 is used for general purposes, whereas in the third embodiment, an overwrite process is performed only on a specific file with a sample document image (not shown) that is different from the sample document image 114. The third embodiment can be realized by replacing FIG. 3 in the first embodiment with FIGS. 12A and 12B, FIG. 5 with FIG. 13, and FIG. 8(b) with FIG. 14(b), and then adding FIG. 14(c). The third embodiment is otherwise similar to the first embodiment, and therefore description thereof will be omitted.

[0079] <Usage sequence> 12A and 12B are diagrams showing a usage sequence of the information processing system 100 in the third embodiment. Hereinafter, FIGS. 12A and 12B will be collectively referred to as "FIG. 12." Note that in FIG. 12, the processes with the same reference numerals as those in the usage sequence of FIG. 3 are the same as those in the first embodiment, and therefore will not be described again.

[0080] In the third embodiment, after steps S311 to S317 described above, the user can specify and input to the information processing device 101 a sample document image (not shown) that is to be applied exceptionally only to a specific file and that is different from the sample document image 114 to which a file name has been assigned. When the user inputs the specification of this sample document image (re-specification of the sample document image), the information processing device 101 accepts this input (S1218).

[0081] Upon receiving the input of re-specifying the sample document image in S1218 above, the information processing device 101 and the information processing server 103 execute the processes of S302 to S304, replacing the sample document image 114 input in S301 with the sample document image (not shown) input in S1218. Thereafter, the information processing server 103 executes steps S314 to S317, replacing the sample document image 114 input in step S301 with a sample document image (not shown) input in step S1218.

[0082] <File name suggestion process> 13 is a flowchart illustrating a flow in the third embodiment in which the information processing server 103 proposes a file name for a document image 113 received from the information processing device 101 using the large-scale language model 116 and outputs the proposal result. Note that an execution program for each step shown in FIG. 13 is stored in the ROM 262, RAM 264, or storage 265 of the information processing server 103, and is executed by the CPU 261 of the information processing server 103. In addition, in FIG. 13, the processes having the same reference numerals as those in the flowchart of FIG. 5 are the same as those in the first embodiment, and therefore description thereof will be omitted.

[0083] After the file name output process in S508, the CPU 261 advances the process to S1309. In S1309, the CPU 261 determines whether or not the user has instructed to apply an exception to the file name output in S508. The criteria for this determination will be described with reference to FIG.

[0084] Fig. 14(b) is an example of a screen in which an apply exception button 1624 has been added to the proposal screen 820 of Fig. 8(b). User operations on this screen are transmitted to the information processing server 103. When the CPU 261 detects that the user has pressed the apply exception button 1624, it determines that the user has instructed to apply the exception. Furthermore, when the CPU 261 detects that the user has pressed the OK button 823, it determines that the user has not instructed to apply the exception.

[0085] In the above S1309, when it is determined that the user has instructed to apply the exception (in the case of Yes in S1309), the CPU 261 proceeds to the process of S1310. Note that, when S1310 has already been executed at the time of S1309, the CPU 261 ends the file name suggestion process. On the other hand, if the user has not instructed to apply the exception (No in S1309), the CPU 261 ends the file name suggestion process.

[0086] In S1310, the CPU 261 receives from the user a sample document image (not shown) that is different from the sample document image 114 and that is to be applied exceptionally only to a specific file, and acquires sample document data from the received sample document image (not shown). Note that the method for acquiring the sample document data is the same as S401 to S404 in FIG. 4, and therefore will not be described here. Furthermore, if the received sample document image (not shown) is a sample document image that has been set exceptionally in the past, the CPU 261 may acquire sample document data corresponding to the sample document image from the storage 265.

[0087] After S1310, the CPU 261 performs control to execute S502 to S508, replacing the sample document data acquired in S501 with the sample document data acquired in S1310.

[0088] <User interface> In the third embodiment, the display unit 158 ​​presents to the user, for example, a file name suggestion screen 820 as shown in FIG. 14(b) and a sample document setting screen when an exception is applied as shown in FIG. 14(c).

[0089] FIG. 14C is an example of a screen on which the user sets a sample document image (not shown) different from the sample document image 114 as a sample document image to be applied exceptionally only to a specific file. In the information processing system 100 of the third embodiment, when it is detected that the user has pressed the apply exception button 1624 on the proposal screen 820 of Figure 14(b), the display unit 158 ​​displays a sample document setting screen 1600 when the exception is applied, as shown in Figure 14(c).

[0090] As a method for setting a sample document image that is applied exceptionally only to a specific file, for example, there are the following first and second methods. The first method is to select a new sample document image using, for example, a file dialog box. The second method is to select a sample document image to be applied exceptionally from those previously set.

[0091] First, the first method (a method for selecting a new sample document image) will be described. In this method, the user presses select button 1601, selects a file from a file dialog, displays the file name of the sample document image in field 1602, and presses OK button 803. The processing corresponding to these operations is the same as in the other embodiments, except that select button 801 in Fig. 8(a) is replaced with select button 1601 in Fig. 14(b) , field 802 is replaced with field 1602, and OK button 803 is replaced with OK button 804, and so a description thereof will be omitted.

[0092] Next, the second method (selecting from previously set items) will be explained below. In this method, when the information processing system 100 detects that a user has pressed a button 1604, it acquires and displays a list 1605 of sample document images that have been set exceptionally in the past from the storage 265 of the information processing server 103. After that, when the information system 100 detects, for example, the selection of a sample document image with a file name 1606 from the list 1605, it re-proposes a file name that references the sample document image.

[0093] As described above, according to the third embodiment, it is possible to exceptionally present a file name for only a specific file, using as a sample a sample document image (not shown) that is separately input by the user and that is different from the sample document image 114. In other words, it is possible to overwrite the file naming rules only for specific files.

[0094] [Fourth embodiment] In the first embodiment, a file name is proposed using only a single large-scale language model 116 set by an engineer. In contrast, in the fourth embodiment, a plurality of file names are proposed by using a plurality of large-scale language models 116. Furthermore, in the first embodiment, a file name is not re-proposed using the large-scale language model 116, but in the fourth embodiment, a configuration is provided in which a file name other than one that has already been proposed is re-proposed.

[0095] The fourth embodiment can be realized by replacing FIG. 3 in the first embodiment with FIG. 15A and FIG. 15B, FIG. 5 with FIG. 16, and FIG. 8(b) with FIG. 18A(b), and then adding FIG. 17, FIG. 18A(a), and FIG. 18B(c). Other than this, the third embodiment is similar to the first embodiment, and therefore a description thereof will be omitted. Hereinafter, FIGS. 18A and 18B will be collectively referred to as "FIG. 18."

[0096] <Usage sequence> 15A and 15B are diagrams showing a usage sequence of the information processing system 100 in the fourth embodiment. Hereinafter, FIGS. 15A and 15B will be collectively referred to as "FIG. 15." Note that in FIG. 15, the processes with the same reference numerals as those in the usage sequence of FIG. 3 are the same as those in the first embodiment, and therefore will not be described again.

[0097] First, the information processing server 103 acquires a setting for a file name suggestion method from the user (S1711). The setting for a file name suggestion method is to select which large-scale language model to use when proposing a file name using the large-scale language model in S314 to S316. This setting can be designated by the user from the information processing device 101, a web browser on a user terminal (not shown), the input device 266, or the like, and the designated setting is acquired by the information processing server 103. The screen for making this setting will be described with reference to FIG. 18(a).

[0098] 18(a) shows an example of a screen for a user to set a filename suggestion method. For example, when "large-scale language model A" has been set as the filename suggestion method, the user may want to also set "large-scale language model B" as the filename suggestion method. The user can set the method by pressing the list down button 2003, selecting "large scale language model B" 2004 from the options for large scale language models that can be set, and pressing the save settings button 2010. If there are other file name suggestion methods that the user wants to set besides "large scale language model B," the user can add input fields (options) for the third and subsequent file name suggestion methods by pressing the add determination method button 2005. Here, an example is shown in which "large scale language model C" has been added. That is, the user can set multiple large-scale language models to which instructions for suggesting a file name to be assigned to the document image 113 will be sent from the file name suggestion method setting screen 2000 shown in FIG. 18(a).

[0099] In response to the acquisition of the file name suggestion method setting in S1711, the information processing server 103 and the external information processing server 105 execute the processes of S315 to S316 for each file name suggestion method acquired in S1711, and present the file name for each file name suggestion method to the user (S317). As in the first embodiment, the file name suggestion is output on the information processing device 101 or the display device 267, for example, on a file name suggestion screen such as that shown in Fig. 18(b) described later.

[0100] After the file name is presented (S317), the information processing server 103 acquires an instruction from the user to re-propose a file name (S1718). This instruction can be given from a file name proposal screen such as that shown in FIG.

[0101] Next, the information processing server 103 generates a directive to re-propose a file name other than the file name acquired by executing the above-mentioned steps S315 to S317 (S1719). In response to this, the information processing server 103 and the external information processing server 105 execute the processes of S315 to S316, replacing the instruction statement generated in S314 with the instruction statement generated in S1719 for each file name suggestion method obtained in S1711, and present the file names for each file name suggestion method to the user again (S317).

[0102] <File name suggestion process> 16 is a flowchart illustrating a flow in the fourth embodiment in which the information processing server 103 proposes multiple file names for a document image 113 received from the information processing device 101 using the large-scale language model 116 and outputs the proposal results. Note that an execution program for each step shown in FIG. 16 is stored in the ROM 262, RAM 264, or storage 265 of the information processing server 103, and is executed by the CPU 261 of the information processing server 103. Note that the processes in FIG. 16 that are designated by the same reference numerals as those in the flowchart of FIG. 5 are the same as those in the first embodiment, and therefore description thereof will be omitted.

[0103] First, in S1801, the CPU 261 acquires the setting of the file name suggestion method as a setting given by the user. Next, the CPU 261 executes the processes of S501 to S505, and proceeds to S1805.

[0104] In S1805 to S1808, the CPU 261 controls to execute the processes of S506 to S507 for each file name suggestion method acquired in S1801. For example, if the user sets "large-scale language model A" and "large-scale language model B" in S1801, the CPU 261 executes the processes of S506 to S507 for each of "large-scale language model A" and "large-scale language model B." That is, in this case, the CPU 261 executes the processes of S506 to S507 twice.

[0105] Next, in S1807, CPU 261 determines whether or not the processing has been completed for all file name suggestion methods acquired in S1801 above, and repeats the processing of S506 to S507 until it is determined that the processing has been completed for all file name suggestion methods. Then, when the processing of S506 to S507 has been completed for all file name suggestion methods, CPU 261 proceeds to the processing of S1808.

[0106] In S1808, the CPU 261 extracts the file names proposed by the file name proposal methods acquired in S1801 above from the answer results acquired by the processes in S1805 to S1807 above, and transmits the extracted file names to the information processing device 101, and the display unit 158 ​​presents them to the user. Note that the file names extracted from the answer results may be output to the display device 267 and presented to the user. An example of a screen on which file names proposed by a plurality of file name proposal methods are presented to the user will be described with reference to FIG. 18(b).

[0107] Figure 18(b) is an example of a file name suggestion screen by the information processing system 100 using the file name suggestion method set in Figure 18(a). Here, it is assumed that text 1910 is a file name suggested by "large-scale language model A" in response to the instruction statement 700 in Figure 7. Also, it is assumed that text 1911 is a file name suggested by "large-scale language model B" in response to the instruction statement 700.

[0108] The CPU 261 displays text 1910 in field 2025 and text 1911 in field 2026, and generates a screen that presents the user with a file name so that the user can select either text 1910 or text 1911. This may be implemented in a format other than a list box, as long as only one file name can be selected. The CPU 261 may also display information about which large-scale language model proposed the file, for example, by adding the name of the large-scale language model next to each of text 1910 and text 1911, as shown in fields 2025 and 2026. If the same file name is proposed by multiple large-scale language models, the CPU 261 may display the file name and information about which large-scale language model proposed the file in a single field. In this example, each large-scale language model proposes one file name, but multiple file names may also be proposed.

[0109] Next, in S1809, the CPU 261 determines whether or not the user has instructed to re-propose the file name output in S1808. The criteria for this determination will be described with reference to FIG.

[0110] 18(b) is transmitted to the information processing server 103. When the CPU 261 detects that the user has pressed the re-proposal button 2023, it determines that the user has instructed a re-proposal. Furthermore, when the CPU 261 detects that the user has pressed the OK button 823, it determines that the user has not instructed a re-proposal. Furthermore, in this example, the CPU 261 executes the re-proposal process only once, but it may execute the re-proposal process multiple times, up to a number of times specified in advance by an engineer or a user.

[0111] In S1809 above, if it is determined that the user has instructed a re-proposal (Yes in S1809), CPU 261 proceeds to the process of S1810. Note that if S1810 has already been performed at the time of S1809, CPU 261 ends the file name proposal process. On the other hand, if the user has not instructed to re-propose a file (No in S1809), the CPU 261 ends the file name proposal process.

[0112] In S1810, the CPU 261 modifies the instruction generated in S505 to an instruction to suggest "a file name other than the one already proposed" using the file name output in S1808. This instruction will be described with reference to Fig. 7(a) and Fig. 17(a).

[0113] Figure 17(a) is an example of a directive that adds a constraint to the directive 700 in Figure 7(a) to suggest a file name other than the "already proposed file name." Directive 1900 is realized by replacing text 701 in Figure 7 with text 1901. Note that text 1901 is configured to include file names (text 1910, 1911) that have already been suggested by large-scale language model A and large-scale language model B for directive 700. The directive 1900 shown in FIG. 17(a) corresponds to a directive for causing the large-scale language model to output a file name other than the file name already proposed for the document image 113.

[0114] After S1810 above, the CPU 261 performs control to execute S1805 to S1809, replacing the instruction statement generated in S505 with the instruction statement corrected in S1810. Regarding S1808 at this time, a supplementary explanation of the processing will be provided with reference to Fig. 17, Fig. 18(b), and Fig. 18(c).

[0115] FIG. 17(b) is an example of a response to the instruction sentence 1900 by the large-scale language model A. FIG. 17(c) shows an example of a response to the instruction sentence 1900 by the large-scale language model B. FIG. 18(c) shows a file name re-proposing screen by the information processing system 100 using the file name proposing method already set in FIG. 18(a).

[0116] 18(c), the re-proposal screen 2040 is obtained by replacing the file names (text 1910, 1911) proposed in response to the instruction statement 700 in FIG. 7 on the proposal screen 2020 with the file names (text 1921, 1941) proposed in response to the instruction statement 1900. Note that in this example, as described above in S1809, the re-proposal process is limited to one time only, and therefore the re-proposal screen 2040 does not include the re-proposal button 2023. However, if the re-proposal process is configured to be able to be performed multiple times, the re-proposal screen 2040 should include the re-proposal button 2023.

[0117] <User interface> In the fourth embodiment, the display unit 158 ​​presents the user with a file name suggestion screen such as that shown in FIG. 18(b) or FIG. 18(c), for example.

[0118] A method for setting a file name by the user will be described with reference to FIGS. 18(b) and 18(c). FIG. 18(b) is an example of a screen that presents a file name to the user based on the file name proposed (S1808) for the document image 650 in FIG. 6. First, a case where the user confirms the file name to text 1911 on the proposal screen 2020 will be described. The user presses the list down button 2030 and then selects the field 2026 indicating text 1911 from among the proposed file names (text 1910, 1911). Thereafter, the user can confirm the file name to text 1911 by pressing the OK button 823. Note that the user can also confirm the file name by correcting the text 1911. This correction method is similar to the user interface section in the first embodiment, and therefore will not be described here.

[0119] Fig. 18(c) is a file name re-proposal screen that is displayed when the information system 100 detects that the user has pressed the re-proposal button 2023 on the proposal screen 2020 of Fig. 18(b). The method for setting a file name on the re-proposal screen 2040 is the same as the method for setting a file name on the proposal screen 2020, and therefore a description thereof will be omitted.

[0120] As described above, according to the fourth embodiment, in addition to the effects of the first embodiment, it is possible to propose a plurality of file names by a plurality of file name suggestion methods. Furthermore, if all of the proposed file names are different from the file name intended by the user, it is possible to re-propose file names other than the above-mentioned plurality of file names. If the proposed file name is different from the file name intended by the user, the configuration of re-proposing a file name other than that file name may be applied to the first to third embodiments described above.

[0121] The screen displayed by the display unit 158 ​​of the information processing device 101 may be a screen generated by the information processing device 101, or a screen generated by the information processing server 103 and transmitted to the information processing device 101. The display unit 158 ​​of the information processing device 101 may be realized by a web browser or the like. Furthermore, input by the user to the information processing server 103 may be performed from an input device 266, and output from the information processing server 103 may be performed on a display device 267.

[0122] According to each embodiment described above, file names can be assigned regardless of whether the file name items match the character strings in the document image. For example, if the aforementioned item "order number on delivery note" exists in the document image, the "order number on delivery note" can also be included in the file name. In this way, file names can be assigned flexibly regardless of whether the file name items match the character strings in the document image.

[0123] It goes without saying that the configurations and contents of the various data described above are not limited to those described above, and that the data may be configured in various configurations and contents depending on the application and purpose. Although one embodiment has been described above, the present invention can be embodied as, for example, a system, an apparatus, a method, a program, a storage medium, etc. Specifically, the present invention may be applied to a system made up of multiple devices, or may be applied to an apparatus made up of a single device. Furthermore, the present invention also includes any combination of the above embodiments.

[0124] Other Embodiments The present invention can also be realized by supplying a program that realizes one or more functions of the above-described embodiments to a system or device via a network or a storage medium, and having one or more processors in the computer of the system or device read and execute the program.The present invention can also be realized by a circuit (e.g., ASIC) that realizes one or more functions. Furthermore, the present invention may be applied to a system made up of multiple devices, or to an apparatus made up of a single device. The present invention is not limited to the above-described embodiments, and various modifications (including organic combinations of the embodiments) are possible based on the spirit of the present invention, and these modifications are not excluded from the scope of the present invention. In other words, all configurations that combine the above-described embodiments and their modifications are included in the present invention.

[0125] The disclosure of this embodiment includes the following configuration, method, and program. (Configuration 1) a first acquisition means for acquiring a character string constituting a first document image and a file name for the first document image; a second acquisition means for acquiring a character string constituting a second document image; a generating means for generating an instruction sentence for causing a large-scale language model to output a file name for the second document image based on the first document image and a file name for the first document image; a transmitting means for transmitting the generated instruction sentence to the large-scale language model; a receiving means for receiving a response from the large-scale language model in response to the transmitted instruction sentence; a presentation means for presenting the file name obtained from the received response; An information processing system comprising: (Configuration 2) the instruction sentence generated by the generation means is an instruction sentence for causing the large-scale language model to, when providing an answer, also present a reason for arriving at the answer; 2. The information processing system according to configuration 1, wherein the presenting means presents the reason together with the file name. (Configuration 3) a first instruction means for receiving an instruction to exceptionally switch the first document image to a third document image; a first control means for controlling each of the means to perform processing using a third document image in place of the first document image in response to an instruction received by the first instruction means; 3. The information processing system according to configuration 1 or 2, comprising: (Configuration 4) a setting means for setting a plurality of the large-scale language models to which the instruction sentence is to be sent; a second control means, when a plurality of large-scale language models are set by the setting means, for causing the transmitting means and the receiving means to execute processing for each of the specified large-scale language models, and for causing the presenting means to present the file name returned for each of the large-scale language models; 4. The information processing system according to any one of configurations 1 to 3, comprising: (Configuration 5) a second instruction means for receiving an instruction to re-propose the file name after the file name is presented by the presenting means; a third control means for controlling the generation means to generate an instruction statement for causing the large-scale language model to output a file name other than the presented file name for the second document image in response to an instruction received by the second instruction means, the generation means to transmit the generated instruction statement to the transmission means, the reception means to receive a response from the large-scale language model in response to the transmitted instruction statement, and the presentation means to present a file name obtained from the received response; 5. The information processing system according to any one of configurations 1 to 4, comprising: (Configuration 6) the first acquisition means acquires a character string constituting the first document image by performing character recognition on the first document image; 6. The information processing system according to any one of configurations 1 to 5, wherein the second acquisition means acquires character strings that constitute the second document image by performing character recognition on the second document image. (Configuration 7) 7. The information processing system according to any one of configurations 1 to 6, wherein the second document image is read from an original and input. (Method 1) An information processing method in an information processing system, comprising: a first acquisition step of acquiring a character string constituting a first document image and a file name for the first document image; a second acquisition step of acquiring character strings constituting a second document image; a generating step of generating an instruction sentence for causing a large-scale language model to output a file name for the second document image based on the first document image and a file name for the first document image; a transmitting step of transmitting the generated instruction sentence to the large-scale language model; receiving a response from the large-scale language model to the transmitted instruction sentence; a presentation step of presenting the file name obtained from the received response; An information processing method comprising: (Program 1) A program for causing a computer to function as each of the means described in any one of configurations 1 to 7. [Explanation of symbols]

[0126] 100 Information Processing Systems 101 Information processing equipment 103 Information Processing Server 104 Network 105 External information processing server

Claims

1. a first acquisition means for acquiring a character string constituting a first document image and a file name for the first document image; a second acquisition means for acquiring a character string constituting a second document image; a generating means for generating an instruction sentence for causing a large-scale language model to output a file name for the second document image based on the first document image and a file name for the first document image; a transmitting means for transmitting the generated instruction sentence to the large-scale language model; a receiving means for receiving a response from the large-scale language model in response to the transmitted instruction sentence; a presentation means for presenting the file name obtained from the received response; An information processing system comprising:

2. the instruction sentence generated by the generation means is an instruction sentence for causing the large-scale language model to, when providing an answer, also present a reason for arriving at the answer; 2. The information processing system according to claim 1, wherein said presenting means presents the reason together with the file name.

3. a first instruction means for receiving an instruction to exceptionally switch the first document image to a third document image; a first control means for controlling each of the means to perform processing using a third document image in place of the first document image in response to an instruction received by the first instruction means; 2. The information processing system according to claim 1, further comprising:

4. a setting means for setting a plurality of the large-scale language models to which the instruction sentence is to be sent; a second control means, when a plurality of large-scale language models are set by the setting means, for causing the transmitting means and the receiving means to execute processing for each of the specified large-scale language models, and for causing the presenting means to present the file name returned for each of the large-scale language models; 2. The information processing system according to claim 1, further comprising:

5. a second instruction means for receiving an instruction to re-propose the file name after the file name is presented by the presenting means; a third control means for controlling the generation means to generate an instruction statement for causing the large-scale language model to output a file name other than the presented file name for the second document image in response to an instruction received by the second instruction means, the generation means to transmit the generated instruction statement, the reception means to receive a response from the large-scale language model in response to the transmitted instruction statement, and the presentation means to present a file name obtained from the received response; 5. The information processing system according to claim 1, further comprising:

6. the first acquisition means acquires a character string constituting the first document image by performing character recognition on the first document image; 5. The information processing system according to claim 1, wherein the second acquisition means acquires character strings constituting the second document image by performing character recognition on the second document image.

7. 5. The information processing system according to claim 1, wherein the second document image is read from an original and input.

8. An information processing method in an information processing system, comprising: a first acquisition step of acquiring a character string constituting a first document image and a file name for the first document image; a second acquisition step of acquiring character strings constituting a second document image; a generating step of generating an instruction sentence for causing a large-scale language model to output a file name for the second document image based on the first document image and a file name for the first document image; a transmitting step of transmitting the generated instruction sentence to the large-scale language model; receiving a response from the large-scale language model to the transmitted instruction sentence; a presentation step of presenting the file name obtained from the received response; An information processing method comprising:

9. A program for causing a computer to function as each of the means according to any one of claims 1 to 4.

Images