File analysis and storage system

Abstract

To expand types of files that can be sorted and use cases, and improve versatility.SOLUTION: A file analysis and storage system 10 includes: a setting unit 11 configured to set analysis keywords; a file analysis unit 12 configured to generate and transmit a first inquiry sentence related to file analysis to cause a generative AI 30 to analyze files based on the analysis keywords; a folder analysis unit 13 configured to generate and transmit a second inquiry sentence related to folder analysis to cause the generative AI 30 to tag each folder in a folder structure for sorting the analyzed files; and a folder sorting unit 15 configured to generate and transmit a third inquiry sentence related to folder sorting to cause the generative AI 30 to suggest folder paths for storing specified files using file and folder analysis responses from the generative AI 30.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a file analysis and storage system. [Background technology]

[0002] A known technique for generating file names and sorting files into folders in computer systems is to determine file names based on file property information managed by the OS, such as the file creation date and update date, and then sort the files into folders. Another known technique, when the target file is received fax data, is to generate file names based on information held by an external system, such as the fax sender number and an address book, and sort the files into folders by year, month, or customer.

[0003] Patent Document 1 discloses a medical information management device that manages files related to medical care, nursing care, and care for each patient who has received medical care, nursing care, or care, with the aim of reducing the effort required to register such files. This medical information management device uses information contained in the file itself, such as the file name, and a table in which patient information is managed to identify the patient ID of the target file and sort it. Patent Document 1 also discloses a configuration that compares the file with past sorting results and sorts it if it is similar. Summary of the Invention [Problem to be solved by the invention]

[0004] However, conventional file sorting methods are limited in the types of files that can be sorted and in the use cases they can be used for.

[0005] The present invention aims to improve versatility by expanding the types of files that can be sorted and the use cases. [Means for solving the problem]

[0006] In order to solve the above-mentioned problems, a file analysis and storage system according to one aspect of the present invention comprises a setting unit that sets analysis keywords for a generation AI to generate a query statement for analyzing a file and folder structure; a file analysis unit that generates a first query statement related to file analysis based on the analysis keywords to have the generation AI analyze files and sends the query statement to the generation AI; a folder analysis unit that generates a second query statement related to folder analysis to have the generation AI tag each folder in the folder structure to sort the files related to the file analysis and sends the query statement to the generation AI; and a folder sorting unit that uses the file analysis and folder analysis responses from the generation AI to generate a third query statement related to folder sorting to have the generation AI suggest a folder path to save a specified file and sends the query statement to the generation AI. [Effects of the Invention]

[0007] The types of files that can be sorted and use cases can be expanded, improving versatility. [Brief explanation of the drawings]

[0008] [Figure 1] Functional block diagram of a file analysis and storage system according to an embodiment [Figure 2A] 1 is a sequence diagram of file analysis settings and analysis processing according to this embodiment. [Figure 2B] 1 is a sequence diagram of file analysis settings and analysis processing according to this embodiment. [Figure 3] FIG. 10 is a diagram showing an example of a main screen for settings used by a setting unit; [Figure 4] A diagram showing an example of an analysis keyword setting screen [Figure 5] An example of an analysis result screen [Figure 6] A diagram showing an example of the hardware configuration of a file analysis and storage system DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, embodiments will be described with reference to the accompanying drawings. To facilitate understanding of the description, the same components in the drawings will be denoted by the same reference numerals as much as possible, and duplicated descriptions will be omitted.

[0010] FIG. 1 is a functional block diagram of a file analysis and storage system 10 according to an embodiment. The file analysis and storage system 10 shown in FIG. 1 performs various processes, such as file analysis, file name generation, folder analysis, and folder allocation, for various files used in a computer system. In particular, in this embodiment, the file analysis and storage system 10 performs these processes using a generation AI 30, such as ChatGPT. For this reason, the file analysis and storage system 10 is communicatively connected to a device, such as a server, on which the generation AI 30 is implemented via a network, such as the Internet. In the following description, the "file analysis and storage system 10" may also be simply referred to as the "system 10."

[0011] Here, the term "generative AI" used in this embodiment refers to an interactive AI service that utilizes trained artificial intelligence (AI) to understand the meaning and purpose of a question posed by a user in natural language and generate an appropriate answer. In particular, in this embodiment, the file analysis and storage system 10 creates a query to request various processes, such as file analysis, file name generation, folder analysis, and folder allocation, from the generator AI 30, and transmits the query to the generator AI 30. The generator AI 30 then replies to the system 10 with answers regarding the various processes described above in accordance with the request content of the received query. The system 10 then uses the answers received from the generator AI 30 to proceed with various processes.

[0012] The file analysis and storage system 10 is also communicably connected via a network to a file recording device 20 in which files and folders used in the various processes described above are recorded. The file recording device 20 may be configured to be provided inside the file analysis and storage system 10 as a component of the file analysis and storage system 10.

[0013] As shown in FIG. 1, the file analysis and storage system 10 includes a setting unit 11, a file analysis unit 12, a folder analysis unit 13, a file name generation unit 14, a folder sorting unit 15, a result display and correction unit 16, and a storage unit 17.

[0014] The setting unit 11 performs necessary settings for file analysis and sorting. In this embodiment, the setting unit 11 can select a file to be analyzed, set analysis keywords, select a folder to sort into, and set a file name format.

[0015] Here, the "analysis keyword" is information for generating a query statement for analyzing the file / folder structure for the generation AI 30. Examples of the analysis keyword include keywords related to file classification, such as document type, document title, and document number.

[0016] The "sorting folder" refers to a group of folders for sorting and storing files related to file analysis. The folder information set here may refer to the top-level folder in a multi-layer folder structure, or may refer to part or the entire folder structure.

[0017] The "file name format" is a template for automatically generating file names based on the file analysis results and analysis keywords. For example, if the analysis keywords are "document type" and "document title," the file name format can be set to, for example, "<document type>_<document title>.(original file extension)."

[0018] The setting process of various information in the setting unit 11 can utilize, for example, an input operation by the user of the system 10 via a user interface (UI). A specific example of the setting process will be described later with reference to FIGS.

[0019] The file analysis unit 12 requests the generation AI 30 to analyze the input file based on the set information. The file analysis unit 12 generates a "first query statement" related to file analysis to have the generation AI 30 analyze the file based on the analysis keywords set by the setting unit 11, and sends it to the generation AI 30. The file analysis unit 12 then causes the generation AI 30 to perform file analysis processing based on the first query statement, and receives a response from the generation AI 30 indicating the processing results, thereby obtaining the file analysis results. Here, the file analysis results are, for example, a group of classification information selected for each of the above-mentioned analysis keywords for the file to be analyzed.

[0020] The folder analysis unit 13 acquires the folder structure for the selected sorting destination folder up to the end and requests the generation AI 30 to tag each folder in that folder structure. The folder analysis unit 13 generates a "second query statement" related to folder analysis to have the generation AI 30 tag each folder in the folder structure for sorting the files related to the file analysis described above, and sends it to the generation AI 30. The folder analysis unit 13 then has the generation AI 30 perform a tagging process based on the second query statement and receives a response from the generation AI 30 indicating the processing results, thereby acquiring the folder analysis results. Here, the folder analysis results are, for example, the folder structure related to file sorting and a group of information about the tags added to each folder in this folder structure.

[0021] The file name generation unit 14 generates a file name based on the result of analysis by the file analysis unit 12 and the file name format set by the setting unit 11. The file name generation unit 14 can automatically generate a file name using the file name format set by the setting unit 11, based on, for example, the file analysis answer obtained from the generation AI 30 using the first query sentence related to file analysis by the file analysis unit 12 and the analysis keyword set by the setting unit 11.

[0022] The folder sorting unit 15 inputs the information analyzed by the file analysis unit 12 and the folder analysis unit 13 into the generation AI 30, causing it to suggest an appropriate storage destination. Using the file analysis and folder analysis results from the generation AI 30, the folder sorting unit 15 generates a "third query statement" related to folder sorting, which causes the generation AI 30 to suggest a folder path for saving the specified file, and transmits the query statement to the generation AI 30. The folder sorting unit 15 then causes the generation AI 30 to perform a folder sorting process based on the third query statement, and receives a processing result response from the generation AI 30, thereby obtaining the folder sorting results. Here, the folder sorting results are, for example, storage destination information indicating which folder in the folder structure related to file sorting each of the files being analyzed will be saved in.

[0023] The result display and correction unit 16 displays the results of the file name generation unit 14 and the folder sorting unit 15 in a list, allowing the system user to confirm the contents proposed by the generation AI 30. The system user can also partially correct the results. The result display and correction unit 16 receives a folder sorting response from the generation AI 30 and displays the results to the user before reflecting the response in the file. The user can also correct some or all of the displayed results as necessary.

[0024] The storage unit 17 stores files based on the results of the result display / modification unit 16. The storage unit 17 causes the file recording device 20 to record information about the group of files stored in each folder of a predetermined folder structure based on the proposal content of the generation AI 30, for example.

[0025] 2A and 2B are sequence diagrams of the file analysis settings and analysis process in this embodiment.

[0026] First, in steps S1 to S4, the setting unit 11 sets various pieces of information necessary for the processing to be performed this time, such as renaming files and sorting files. This setting processing will be described with reference to FIGS. 3 and 4. FIG. 3 is a diagram showing an example of a main setting screen 100 used by the setting unit 11. The main setting screen 100 is displayed on, for example, a display device of the file analysis and storage system 10, and setting operations can be performed via an input device such as a mouse or keyboard.

[0027] In step S1, the setting unit 11 selects a file to be analyzed from the system UI. Examples of file selection methods include drag-and-drop by the system user or use of a folder browse function. In the example of FIG. 3, the setting main screen 100 includes a file list display unit 101. The file list display unit 101 displays a list of file names, paths (save destinations), thumbnail images of the file contents, and other information for files that have already been selected. Note that, as shown in FIG. 1, if the file recording device 20 is a separate device from the system 10, the path of the file to be analyzed will be a location within the file recording device 20. Alternatively, clicking on a line for each file displayed in the file list display unit 101 may display a preview of the file. The setting main screen 100 in FIG. 3 corresponds to the system UI described above.

[0028] Alternatively, an add button 101A may be displayed in the file list display section 101. In this case, for example, when the add button 101A is clicked, the folder structure within the system is displayed in a separate window as a tree or icons. A system user can use this separate window to open the folder in which the desired file is saved and select the desired file. The file selected in this way is added and displayed in the file list display section 101.

[0029] In step S2, the setting unit 11 inputs keywords to be analyzed (analysis keywords) on the system UI. Multiple analysis keywords can be set. In the example of FIG. 3, the setting main screen 100 has an analysis keyword display unit 103. The analysis keyword display unit 103 displays information on analysis keywords that have already been selected. In the example of FIG. 3, examples of analysis keywords displayed include "customer name," "document number," "document type (invoice, estimate, etc.)," ​​and "document title."

[0030] The analysis keyword can be selected from those predefined in the system as a basic keyword, or any character string can be freely entered as a custom keyword. Fig. 4 is a diagram showing an example of an analysis keyword setting screen 200. For example, when the analysis keyword display section 103 on the main setting screen 100 is clicked, the analysis keyword setting screen 200 in Fig. 4 is displayed. The analysis keyword can be set using the analysis keyword setting screen 200, for example.

[0031] The analysis keyword setting screen 200 has a basic keyword display section 201 and a custom keyword display section 202. The basic keyword display section 201 displays basic keywords predefined by the system. Commonly used basic keywords are registered, and in the example of FIG. 4, "document title," "document type," and "document number" are registered and displayed as basic keywords. The custom keyword display section 202 displays custom keywords that can be added by the user. In the example of FIG. 4, "billing amount," "inspection date," "contract date," and "customer name" are registered and displayed as custom keywords.

[0032] In the basic keyword display section 201 and the custom keyword display section 202, when a keyword is selected by clicking or the like, candidate values ​​for that keyword can be defined. For example, a candidate value display section 203 listing candidate values ​​for the selected keyword is additionally displayed, allowing the user to select desired information from these candidate values. In the example of FIG. 4, "Document Type" is selected in the basic keyword display section 201, thereby displaying the candidate value display section 203 related to document type. In this case, the candidate value display section 203 displays "Quote," "Contract," "Invoice," "Inspection Letter," "Consent Letter," "Proposal," and "Report" as candidate values ​​related to document type. When document type is selected as an analysis keyword, the user clearly determines the type to be classified, allowing specific instructions to be given to the generation AI 30, thereby improving the accuracy of answers.

[0033] Furthermore, on the analysis keyword setting screen 200, for example, as shown in Figure 4, by performing a process to change the background color of the display column for the selected keyword, the user can clearly visually understand the selected keyword, making the setting operation easy and reliable.

[0034] The analysis keyword setting screen 200 further has a set button 204 and a cancel button 205. When the set button 204 is clicked, the keyword group currently selected on the analysis keyword setting screen 200 is set as the analysis keywords, the screen returns to the setting main screen 100 in Figure 3, and the analysis keywords displayed in the analysis keyword display section 103 are updated to the latest information. On the other hand, when the cancel button 205 is clicked, the keyword group currently selected on the analysis keyword setting screen 200 is not reflected and the screen returns to the setting main screen 100.

[0035] In step S3, the file name format is set on the UI by combining the keywords entered in step S1. In the example of FIG. 3, the main setting screen 100 has a file name format display section 104. The file name format display section 104 displays the currently set file name format. In the example of FIG. 3, an example of a file name format is displayed as "<document type>_<customer name>_<document number>.(original file extension)". In this example, the part enclosed in <> contains the name of the basic keyword or custom keyword of the analysis keyword set in step S2. Then, based on the file analysis results of step S7 (described later), classification information related to the keyword entered in the corresponding part (for example, the "candidate value" information described with reference to FIG. 4) is input in step S8 to generate a file name.

[0036] It is preferable that the file name format be automatically created by sequentially arranging all of the basic keywords and custom keywords of the analysis keywords selected in step S2. This allows each keyword to be used to easily generate an appropriate file name based on the response of the generation AI 30. The file name format may be set by a user's input operation, or may be automatically created and modified as needed by a user's input operation. The file name format may also include only some of the basic keywords and custom keywords of the analysis keywords selected in step S2.

[0037] In step S4, a destination folder is selected by folder browsing on the system UI. In the example of FIG. 3, the setting main screen 100 has a destination folder path display section 102. The destination folder path display section 102 displays the name of the currently set destination folder. In the example of FIG. 3, a folder named "Data" in the D drive of the system 10 is selected. Note that the destination folder may also be set in the file recording device 20.

[0038] When the processing of steps S1 to S4 above is completed, all information necessary for subsequent processing is set on the setting main screen 100 shown in Fig. 3. The setting main screen 100 has an analysis button 105. When the analysis button 105 is clicked with information set on each of the display sections 101 to 104, the processing proceeds to step S5 and subsequent steps.

[0039] In step S5, the file analysis unit 12 generates a "first query statement" based on the information set in steps S1 to S3. The file analysis unit 12 generates a "first query statement" related to file analysis, based on the analysis keywords set by the setting unit 11, for causing the generation AI 30 to analyze the file.

[0040] The file analyzer 12 creates the first query using a predetermined template. The template for the first query is, for example, as follows:

[0041] <Example of standard sentence for the first inquiry> "For the following document, please extract <Analysis Keyword 1>, <Analysis Keyword 2>, <Analysis Keyword 3>, ..." (*If there are candidate values ​​for analysis keyword 1) For <Analysis Keyword 1>, please select from <Candidate Value 1>, <Candidate Value 2>, <Candidate Value 3>..."

[0042] In the case of the example of analysis keywords shown in Figures 3 and 4, the first query sentence can be created by filling in the set keywords and candidate values ​​in the <> field of the above template sentence, for example, as shown below.

[0043] <Example of the first query> "Please extract the customer name, document number, document type, and document title for the following documents. Please select the document type from among estimate, contract, invoice, and receipt.

[0044] In step S6, the file analysis unit 12 sends the "first query sentence" generated in step S5 to the generation AI 30 to request file analysis.

[0045] In step S7, the generation AI 30 analyzes the contents of the file based on the specified keywords, and the file analyzer 12 of the system 10 obtains a response from the generation AI 30. For example, the response from the generation AI 30 to the first query statement is as follows:

[0046] <Example of response to the first inquiry> "Customer name: A Co., Ltd. Document number: M000001 Document type: Quotation Document title: Quotation

[0047] The file analysis unit 12 outputs the answer information acquired from the generation AI 30 to the file name generation unit 14 .

[0048] In step S8, the file name generation unit 14 generates a file name in accordance with the specified format based on the result of step S7. For example, if the information in the example response of step S7 is applied to the example of file name format display unit 104 in Figure 3, the file name can be generated as follows. In the example below, the type of the target file is a PDF file.

[0049] <Example file name> "Quotation_Co., Ltd. A_M000001.pdf"

[0050] In step S9, the file name generation unit 14 stores in the system 10 the file analysis result acquired in step S7 and the information on the file name generation result generated in step S8.

[0051] In step S10, the folder analysis unit 13 expands the folders set in step S4 to the end and generates the folder structure and instructions for tagging each folder. That is, the folder analysis unit 13 generates a "second query statement" related to folder analysis to cause the generation AI 30 to tag each folder in the folder structure for sorting the files analyzed in step S7.

[0052] The folder analysis unit 13 creates the second query sentence using a predetermined fixed phrase. The fixed phrase of the second query sentence is, for example, as follows:

[0053] <Example of the second query> "Please tag each folder in the folder structure below <sorting folder>, taking into account the parent folder."

[0054] In the case of the example of the sorting folder shown in FIG. 3, the second query sentence can be created by substituting the set sorting folder in the <> field of the above template sentence, for example, as follows:

[0055] <Example of the second query> "Please tag each folder in the folder structure under "D:￥Data" taking into account the parent folder."

[0056] As in the example above, by including the phrase "consider the parent folder" in the query, even if there are multiple folders with the same folder name that are located under different parent folders, different tags can be added to each folder. This makes it possible to distinguish between multiple folders with the same name in subsequent sorting processes.

[0057] In step S11, the folder analysis unit 13 transmits the "second query statement" generated in step S10 to the generation AI 30 to request a folder structure analysis.

[0058] In step S12, the generation AI 30 understands the folder structure and tags each folder according to the second query statement in step S11. The folder analysis unit 13 of the system 10 obtains the response from the generation AI 30. An example of the response from the generation AI 30 to the second query statement is, for example, a group of tag information associated with each folder in the folder structure being analyzed. One method for creating tags is to assign a symbol to each type of analysis keyword or candidate value, and then use a string of characters that combines the symbol of the candidate value and the symbol of its parent folder as the tag for the child folder. This allows different tags to be added to each folder even if the child folders under different parent folders have the same name, making it possible to distinguish them by tag.

[0059] In step S13, the folder analysis unit 13 stores in the system 10 the folder analysis results acquired in step S12, that is, the information on the tags added to each folder to be analyzed.

[0060] In step S14, the folder sorting unit 15 inputs the file analysis result and file name generation result information saved in step S9 and the folder analysis result information saved in step S13 into the generation AI 30, and generates a query statement asking which folder path is best to save it in. By asking for the "folder path", it is more likely that the response from the generation AI 30 will also be a folder path.

[0061] The folder sorting unit 15 generates a "third query" for folder sorting to have the generation AI 30 suggest a folder path for saving the specified file. The folder sorting unit 15 creates the third query using a predetermined template. The template for the third query is, for example, as follows:

[0062] <Example of the third inquiry sentence> The folder structure and document information are shown below. Based on the folder tag information and document information, which folder path should this document <file name> be saved to? Folder structure <Answer to the second question> Document information <Answer to the first inquiry>

[0063] In the example of steps S6, S8, and S12 above, the second query sentence can be created by filling in the <> fields of the above template sentence with the set file name and information for each response, for example, as shown below.

[0064] <Example of the third inquiry sentence> The folder structure and document information are shown below. Based on the folder tag information and document information, in which folder path should this document "Quotation_Co., Ltd. A_M000001.pdf" be saved? Folder structure ... Document information Customer name: A Co., Ltd. Document number: M000001 Document type: Quotation Document title: Quotation

[0065] In step S15, the folder sorting unit 15 transmits the "third inquiry sentence" generated in step S14 to the generation AI 30 to request folder sorting.

[0066] In step S16, the generation AI 30 responds with the most appropriate folder path based on the keywords extracted from the file to be sorted and the information tagged to the folder in accordance with the third query statement in step S15. The folder sorting unit 15 of the system 10 obtains the response from the generation AI 30. The response from the generation AI 30 to the third query statement is, for example, as follows:

[0067] <Example of response to the third inquiry> "Project A / Customer A / Quotation"

[0068] In step S17, the result display / modification unit 16 displays on the UI the file names saved in step S9 and the folder sorting results of step S17. FIG. 5 is a diagram showing an example of an analysis result screen 300. The result display / modification unit 16 displays, for example, the analysis result screen 300 shown in FIG. 5 on a display device of the file analysis / storage system 10. It can also accept various operations, such as correction instructions, from the user via an input device such as a mouse or keyboard. The analysis result screen 300 in FIG. 5 corresponds to the UI described above.

[0069] As shown in FIG. 5, analysis result screen 300 displays a list of suggested file name changes and save destinations for input files. In the example of FIG. 5, analysis result screen 300 displays multiple pieces of information for each file on one line. According to the multiple pieces of information, analysis result screen 300 displays a suggested file name display field 301, a suggested save destination display field 302, an original file name display field 303, a file name reflection selection check field 304, and a save destination application check field 305. Analysis result screen 300 may also be provided with an all-item check field 306. Analysis result screen 300 also displays an reflect button 307, a cancel button 308, and a retry button 309.

[0070] In step S18, the system user changes the file name for each file based on the results of step S17, and modifies the file sorting results on the UI as necessary. For example, the user can modify the file name or save destination by clicking on the suggested file name display field 301 or the suggested save destination display field 302 in the row of the file they want to modify. When the result display / modification unit 16 accepts these modification operations, it updates the display content of the corresponding field on the analysis result screen 300.

[0071] In step S19, based on the results of step S18, the system user selects whether to change the file name and reflect the file sorting results for each file. For example, the user can apply the contents proposed by the generation AI 30 to the file name and save destination by checking the file name reflection selection check box 304 and the save destination application check box 305 for each file row. On the other hand, the user can also not apply the unchecked file name or save destination by unchecking these boxes 304 and 305.

[0072] Also, by checking or unchecking the all item check column 306, the presence or absence of checks in each row of the file name reflection selection check column 304 or the save destination application check column 305 can be changed all at once.

[0073] In step S20, based on the results of step S19, the system 10 changes the file names and sorts the files. When the Reflect button 307 is clicked, the result display / modification unit 16 checks the changes for the items checked in the file name reflection selection check box 304 and the save destination application check box 305 by the above-mentioned operation, and updates the file name or save destination for each file. On the other hand, no changes are made to the original files for items that were not selected. For example, if you want to reflect only the file name change to the original file, you can check the file name reflection selection check box 304 and uncheck the save destination application check box 305.

[0074] Furthermore, if the file name reflection selection check box 304 and the save destination application check box 305 are partially unchecked and the reflect button 307 is clicked, a separate screen may be displayed for the unchecked areas, allowing the user to input the desired file name and specify the save destination. This can make the correction work more efficient.

[0075] If the stop button 308 is clicked, the result display / modification unit 16 will not update any of the file names or save destinations in this analysis process, and will end the process without making any changes to the original files.

[0076] If the retry button 309 on the analysis result screen 300 is clicked in steps S18 to S20, the file analysis and storage system 10 will re-perform the analysis and classification process using the generation AI 30 from step S5 onwards. This allows the user to be given the option of "retry" to have the generation AI 30 re-perform the process and prompt a new proposal if the proposed result is not what the user desires.

[0077] However, in this embodiment, as described above, the query to the generation AI 30 is created using a predetermined template. The reason for this is to suppress changes in the response of the generation AI 30 due to variations in the nuances of the query content, making the response more stable and improving accuracy. Therefore, even if a retry is made using the same template, there is a high possibility that the same suggestion result will be obtained.

[0078] For this reason, when a configuration that performs retries is applied, it is preferable to prepare in advance a plurality of types of fixed phrases for generating query sentences.

[0079] That is, the file analysis unit 12 has multiple standard phrases for the first query, and if it is necessary to redo the file analysis, it recreates the first query using a standard phrase other than the one already used and performs the file analysis again. The folder analysis unit 13 has multiple standard phrases for the second query, and if it is necessary to redo the folder analysis, it recreates the second query using a standard phrase other than the one already used and performs the folder analysis again. The folder sorting unit 15 has multiple standard phrases for the third query, and if it is necessary to redo the folder sorting, it recreates the third query using a standard phrase other than the one already used and performs the folder sorting again.

[0080] By using multiple standard phrases in this way, it is possible to obtain answers for each process from the generation AI 30 using other query phrases by performing a retry, making it easier to change the content of re-proposed file names and save destinations. This makes it easier for users to obtain the desired proposal results, improving the convenience and versatility of the system.

[0081] It is also preferable that priorities are set individually for multiple template messages, and that they are used according to their priorities. This allows the intentions of the system designer to be reflected in the priority of template messages, and therefore, compared to a configuration in which template messages are selected randomly, it is possible to avoid situations in which the proposed file name or save destination changes drastically even when retries are made.

[0082] The folder analysis unit 13 may also be configured to generate a fourth query for folder structure creation to cause the generation AI 30 to create a folder structure for sorting the files to be analyzed, send the query to the generation AI 30, and have the generation AI 30 tag each folder in the folder structure based on the results of the folder structure creation by the generation AI 30. This eliminates the need for the user to create the folder structure in advance, further reducing the workload.

[0083] The file analysis and storage system 10 of this embodiment includes a setting unit 11 that sets analysis keywords, a file analysis unit 12 that generates and transmits a first query related to file analysis to have the generation AI 30 analyze the files based on the analysis keywords, a folder analysis unit 13 that generates and transmits a second query related to folder analysis to have the generation AI 30 tag each folder in the folder structure for sorting the files related to the file analysis, and a folder sorting unit 15 that generates and transmits a third query related to folder sorting using the file analysis and folder analysis responses from the generation AI 30 to have the generation AI 30 suggest a folder path for saving a specified file.

[0084] This configuration allows the generation AI 30 to be used for various processes, such as file analysis, file name generation, folder analysis, and folder allocation, for various files used in a computer system, thereby improving the accuracy of these processes. Furthermore, by appropriately changing the analysis keywords set by the setting unit 11, the above-mentioned various processes can be performed according to various file types and use cases. Furthermore, since query statements for the generation AI 30 to perform various processes are generated based on standard phrases, variations in the nuances of the queries to the generation AI 30 can be suppressed, and the responses from the generation AI 30 can be more stable. As a result, the file analysis and storage system 10 of this embodiment can expand the types of files and use cases that can be sorted, thereby improving versatility.

[0085] The file analysis and storage system 10 according to this embodiment also includes a result display and correction unit 16 that receives a folder sorting response from the generation AI 30, displays the result to the user before reflecting the response in the file, and allows the user to correct the result as needed. This configuration allows the user to check the result and correct it as needed without directly reflecting the results of file analysis and folder sorting using the generation AI 30, making it easier to guide the user toward the desired result and improving versatility and convenience.

[0086] The file analysis and storage system 10 according to this embodiment also includes a file name generation unit 14 that generates a file name using a file name format based on the file analysis answer obtained from the generation AI 30 by the file analysis unit 12 using the first query sentence related to file analysis and the analysis keywords set by the setting unit 11. The file name format is set by the setting unit 11.

[0087] In order to properly save and reference files, it is necessary for file names to be easy to understand. In this embodiment, by providing the file name generation unit 14, the system 10 can suggest new file names based on the analysis results, thereby significantly reducing the effort required for file management.

[0088] FIG. 6 is a diagram illustrating an example of the hardware configuration of the file analysis and storage system 10. As shown in FIG.

[0089] As shown in Figure 6, the file analysis and storage system 10 is constructed by a computer, and as shown in Figure 6, it includes a CPU 501, a ROM 502, a RAM 503, an HD 504, an HDD (Hard Disk Drive) controller 505, a display 506, an external device connection I / F (Interface) 508, a network I / F 509, a data bus 510, a keyboard 511, a pointing device 512, a DVD-RW (Digital Versatile Disk Rewritable) drive 514, and a media I / F 516.

[0090] Of these, the CPU 501 controls the overall operation of the file analysis and storage system 10. The ROM 502 stores programs, such as an IPL, used to drive the CPU 501. The RAM 503 is used as a work area for the CPU 501. The HD 504 stores various data, such as programs. The HDD controller 505 controls the reading and writing of various data from and to the HD 504 under the control of the CPU 501. The display 506 displays various information, such as a cursor, menus, windows, characters, or images. The external device connection I / F 508 is an interface for connecting various external devices. In this case, external devices include, for example, USB (Universal Serial Bus) memory and printers. The network I / F 509 is an interface for data communication using a communication network. The bus line 510 is an address bus, data bus, or the like for electrically connecting the various components, such as the CPU 501, shown in FIG. 6.

[0091] The keyboard 511 is a type of input means having multiple keys for inputting characters, numbers, various instructions, etc. The pointing device 512 is a type of input means for selecting and executing various instructions, selecting a processing target, moving a cursor, etc. The DVD-RW drive 514 controls reading and writing of various data from a DVD-RW 513, which is an example of a removable recording medium. Note that this is not limited to a DVD-RW, and may be a DVD-R, etc. The media I / F 516 controls reading and writing (storing) of data from a recording medium 515, such as a flash memory.

[0092] The present embodiment has been described above with reference to specific examples. However, the present disclosure is not limited to these specific examples. Design modifications to these specific examples made by a person skilled in the art as appropriate are also included within the scope of the present disclosure as long as they comprise the features of the present disclosure. The elements of the above-described specific examples, as well as their arrangement, conditions, shape, etc., are not limited to those exemplified and can be modified as appropriate. The elements of the above-described specific examples can be combined in various ways as appropriate, as long as no technical contradictions arise.

[0093] For example, aspects of the present invention are as follows. <1> A setting unit that sets analysis keywords for generating a query for analyzing file and folder structures for the generation AI; a file analysis unit that generates a first query statement related to file analysis based on the analysis keyword, for causing the generation AI to analyze the file, and transmits the first query statement to the generation AI; A folder analysis unit that generates a second query statement related to folder analysis for each folder in a folder configuration for sorting files related to the file analysis, and transmits the second query statement to the generation AI to cause the generation AI to tag the folder. A folder sorting unit that uses the file analysis and folder analysis responses from the generation AI to generate a third query statement related to folder sorting for causing the generation AI to propose a folder path for saving a specified file, and transmits the third query statement to the generation AI; A file analysis and storage system. <2> A result display and correction unit is provided which receives a response from the generating AI regarding the folder sorting and displays the result to the user before reflecting the response result in the file, and allows the user to correct the result as necessary. The aforementioned <1> The file analysis and storage system described in <3> a file name generation unit that generates a file name using a file name format based on the answer to the file analysis obtained from the generation AI using the first query sentence related to the file analysis and the analysis keyword; The file name format is set by the setting unit. The aforementioned <1> or <2> The file analysis and storage system described in <4> the file analysis unit has a plurality of fixed phrases for the first query sentence, and when it is necessary to redo the file analysis, recreates the first query sentence using a fixed phrase different from the fixed phrase that has already been used, and performs the file analysis again; the folder analysis unit has a plurality of standard phrases for the second query sentence, and when it is necessary to redo the folder analysis, the folder analysis unit recreates the second query sentence using a standard phrase different from the standard phrase that has already been used, and performs the folder analysis again; the folder sorting unit has a plurality of standard phrases for the third query sentence, and when it is necessary to redo the folder sorting, the folder sorting unit recreates the third query sentence using a standard phrase different from the standard phrase that has already been used, and redoes the folder sorting. The aforementioned <1> ~ <3> 10. The file analysis and storage system according to any one of claims 1 to 9. <5> Priorities are set individually for the plurality of standard phrases, and the standard phrases are used according to the priorities. The aforementioned <4> The file analysis and storage system described in <6> the setting unit sets a folder for which the folder analysis unit performs the folder analysis. The aforementioned <1> ~ <5> 10. The file analysis and storage system according to any one of claims 1 to 9. <7> The folder analysis unit generates a fourth query statement related to folder structure creation to cause the generation AI to create a folder structure for sorting files to be analyzed, and transmits the fourth query statement to the generation AI; causing the generation AI to tag each folder in the folder structure based on the result of the folder structure creation by the generation AI; The aforementioned <1> ~ <5> 10. The file analysis and storage system according to any one of claims 1 to 9. [Explanation of symbols]

[0094] 10 File Analysis and Storage System 11 Setting section 12 File Analysis Unit 13 Folder Analysis Section 14 File name generator 15 Folder sorting section 16 Result display / correction section 30 Generation AI [Prior art documents] [Patent documents]

[0095] [Patent Document 1] Japanese Patent Application Publication No. 2019-219979

Claims

1. a setting unit that sets analysis keywords for generating a query for analyzing file and folder structures for the generation AI; a file analysis unit that generates a first query statement related to file analysis based on the analysis keyword to cause the generation AI to analyze the file, and transmits the first query statement to the generation AI; a folder analysis unit that generates a second query statement related to folder analysis for each folder in a folder configuration for sorting files related to the file analysis, and transmits the second query statement related to folder analysis to the generation AI to cause the generation AI to tag the files; A folder sorting unit that uses the file analysis and folder analysis responses from the generation AI to generate a third query statement related to folder sorting to have the generation AI suggest a folder path for saving a specified file, and sends the third query statement to the generation AI; A file analysis and storage system equipped with:

2. a result display / modification unit that receives a response from the generating AI regarding the folder sorting and displays the result to a user before reflecting the response result in the file, and allows the user to modify the result as necessary; The file analysis and storage system according to claim 1 .

3. a file name generation unit that generates a file name using a file name format based on the answer to the file analysis obtained from the generation AI using the first query sentence related to the file analysis and the analysis keyword; The file name format is set by the setting unit.

3. The file analysis and storage system according to claim 1 or 2.

4. the file analysis unit has a plurality of standard phrases for the first query sentence, and when it is necessary to redo the file analysis, the file analysis unit recreates the first query sentence using a standard phrase different from the standard phrase that has already been used, and performs the file analysis again; the folder analysis unit has a plurality of standard phrases for the second query sentence, and when it is necessary to redo the folder analysis, the folder analysis unit recreates the second query sentence using a standard phrase different from the standard phrase that has already been used, and performs the folder analysis again; the folder sorting unit has a plurality of standard phrases for the third inquiry sentence, and when it is necessary to redo the folder sorting, the folder sorting unit recreates the third inquiry sentence using a standard phrase different from the standard phrase that has already been used, and redoes the folder sorting. The file analysis and storage system according to claim 1 .

5. Priorities are set individually for the plurality of standard phrases, and the standard phrases are used according to the priorities. The file analysis and storage system according to claim 4.

6. the setting unit sets a folder for which the folder analysis unit performs the folder analysis. The file analysis and storage system according to claim 1 .

7. The folder analysis unit generates a fourth query statement related to folder structure creation to cause the generation AI to create a folder structure for sorting files to be analyzed, and transmits the fourth query statement to the generation AI; causing the generation AI to tag each folder in the folder structure based on the result of the folder structure creation by the generation AI; The file analysis and storage system according to claim 1 .

Images