Information processing device, method, and program

Abstract

To provide an information processing device, method, and program capable of supporting the matching of structured data in different formats. [Solution] The information processing device according to the embodiment comprises a structure analysis unit and a matching unit. The structure analysis unit analyzes the structure of first structured data and second structured data structured by one or more items to obtain a first item included in the first structured data and a second item included in the second structured data, and determines the correspondence between the first item and the second item for each item. The matching unit matches the first structured data and the second structured data using the correspondence.

Description

Technical Field

[0004] , ,

[0001] Embodiments of the present invention relate to an information processing apparatus, method, and program.

Background Art

[0002] Regarding structured data (hereinafter referred to as structured data), it may be necessary to perform collation between structured data having different structures. Structured data is, for example, data composed of one or more items. Each item includes, for example, text data of an item name and content. For example, in the manufacturing industry, there is a trouble management system for managing the content of troubles that have occurred. The trouble management system is a system for abstracting the content of a report (hereinafter referred to as a trouble report) created when a trouble occurs and systematically managing the trouble events. A trouble report is generally structured data composed of one or more items such as the situation where a trouble occurred, the location where the trouble occurred, the cause of the trouble, and subsequent countermeasures. A trouble report is created for each trouble case.

[0003] In the trouble management system, the content of the trouble report is systematized for each item such as the type of trouble, the location where the trouble occurred, and the content of the response to the trouble, and an item is created for each trouble content, and the trouble is managed using a trouble list in which the items are listed. In the trouble management system, a list of troubles that have occurred in the past and troubles that are likely to occur in the future is listed and recorded, and when a new trouble occurs, by referring to the trouble list, it can be used for investigating the cause of the trouble, responding to the trouble, and failure mode effect analysis (FMEA).

[0004] In this type of trouble management system, when a trouble report is entered, the contents of the entered trouble report are compared with the contents of the trouble list. If the contents of the trouble report are not included in the trouble list, a new item is created using the contents of the trouble report and added to the trouble list. On the other hand, if the contents of the trouble report are already included in the trouble list, the consistency between the contents of the trouble report and the corresponding existing item is checked, the contents of the trouble report are added to the corresponding existing item, or the corresponding item and the trouble report are linked.

[0005] In this way, when a new trouble report is entered into the trouble management system, the system uses the entered document as the source and the trouble list as the destination to perform a comparison between the trouble report and each item in the trouble list. Based on the comparison results, the system creates new items or adds information to existing items. Each of the above operations may be performed automatically by a computer or manually by an administrator.

[0006] During the matching process, the structured data format may differ between the source and the referenced document. For example, the item names in the source and the referenced document may differ. Specifically, the source document may not have an item called "Type of Trouble," while the referenced document does. Furthermore, even if the content is the same, the degree of abstraction may differ between the source and the referenced document. For example, content described in a single item, "Location of Occurrence," in the source document may be divided into multiple items such as "Building Location," "Device Name," and "Inside the Device" in the referenced document. In addition, the format of expression may differ between the source and the referenced document. For example, content described as "Hair Contamination" in the source document may be abstracted and described as "Foreign Object Contamination" in the referenced document.

[0007] As described above, when the structure and presentation of references differ between the source and the referenced, accurate matching requires the administrator's knowledge and experience. Furthermore, this increases the administrator's workload and the amount of work required for the matching process. [Prior art documents] [Patent Documents]

[0008] [Patent Document 1] Japanese Patent Publication No. 2023-3933 [Overview of the Initiative] [Problems that the invention aims to solve]

[0009] The problem that this invention aims to solve is to provide an information processing device, method, and program capable of supporting the matching of structured data of different formats. [Means for solving the problem]

[0010] The information processing device according to the embodiment comprises a structure analysis unit and a matching unit. The structure analysis unit analyzes the structure of first structured data and second structured data structured by one or more items to obtain a first item included in the first structured data and a second item included in the second structured data, and determines the correspondence between the first item and the second item for each item. The matching unit matches the first structured data and the second structured data using the correspondence. [Brief explanation of the drawing]

[0011] [Figure 1] A diagram showing an example of the configuration of an information processing system. [Figure 2] A diagram showing an example of structured data in document format. [Figure 3] This figure shows an example of structured data in tabular format. [Figure 4] A diagram illustrating another example of structured data in document format. [Figure 5]This figure shows an example of a different type of structured data in tabular format. [Figure 6] A diagram showing an example of the configuration of an information processing device. [Figure 7] A flowchart illustrating the processing procedure for matching by an information processing device. [Figure 8] A diagram showing an example of a trouble report. [Figure 9] A diagram showing an example of items on a troubleshooting list. [Figure 10] A diagram showing an example of the editing screen. [Figure 11] A diagram showing another example of the editing screen. [Modes for carrying out the invention]

[0012] The information processing apparatus, method, and program according to the embodiment will be described below with reference to the drawings.

[0013] (First embodiment) Figure 1 shows an example configuration of an information processing system 100 including an information processing device 10. The information processing system 100 is a system that supports the management of structured data. For example, the information processing system 100 is a trouble management system for managing the details of troubles that occur in the manufacturing industry.

[0014] Structured data is data that contains one or more items and is structured by one or more items. Structured data is typically structured by multiple items. Structured data can be, for example, in document format or tabular format. Figure 2 shows an example of structured data in document format. The document in Figure 2 contains N items. Figure 3 shows an example of structured data in tabular format. In Figure 3, each of the N items contains M items of content. Note that structured data may also be data converted from other formats to document format, tabular format, etc. Other formats of data can include, for example, free-form text, bulleted lists, data formats such as JSON and YAML, data formats written in markup languages ​​such as HTML and XML, electronic document formats such as presentation files and PDFs, and data in various database formats. Furthermore, other formats of data may also be text data digitized by applying OCR or similar methods to non-digitized documents or data.

[0015] Each item in structured data consists of its content and associated information. The content of an item can include, for example, natural language, symbol sequences, functions, macro language, markup language, and media data. Natural language can include words, phrases, or sentences. Symbol sequences can include URLs or IDs. Media data can include image data, video data, or audio data. Associated information for an item can include, for example, the item name (title), update time, edit time, last editor, edit history, size (size of the input field, amount of text, data size), content layout information, and placement (layout) information within the structured data. Associated information may be explicitly displayed when the structured data is viewed, or it may not be explicitly displayed. For example, the name (title) is an example of information that is explicitly displayed when the structured data is viewed. Furthermore, the associated information may also be information about the structured data containing that item. For example, the associated information may include the file name, edit time, last editor, and data size of the structured data containing that item.

[0016] Also, as shown in FIGS. 4 and 5, the content of an item may include other items composed of content and attached information. FIG. 4 is a diagram showing another example of structured data in document format. In FIG. 4, each of the contents of N items includes M items including item names 1 to M and contents 1 to M. FIG. 5 is a diagram showing another example of structured data in table format. In FIG. 5, each of the N items includes M contents, and each content includes L items 1 to L.

[0017] As shown in FIG. 1, the information processing system 100 includes an information processing apparatus 10, an input apparatus 20, a structured data storage apparatus 30, and an editing apparatus 40. The information processing apparatus 10 is connected to each of the input apparatus 20, the structured data storage apparatus 30, and the editing apparatus 40 via a network, and can transmit and receive data to and from each of the input apparatus 20, the structured data storage apparatus 30, and the editing apparatus 40.

[0018] The input apparatus 20 is a computer terminal to which first structured data is input. The first structured data is, for example, a trouble report regarding a newly occurred trouble. The input apparatus 20 includes a display device that displays an input screen for inputting the first structured data, and an input device for inputting the content of the first structured data. As the input device, a keyboard, a mouse, various switches, a touch pad, a touch panel display, or the like can be used. The display device displays various data. As the display device, a CRT (Cathode-Ray Tube) display, a liquid crystal display, an organic EL (Electro Luminescence) display, an LED (Light-Emitting Diode) display, a plasma display, or any other arbitrary display can be appropriately used. The display device may be a projector. The input apparatus 20 is an example of a storage unit that records the first structured data, and may be mounted inside the information processing apparatus 10. The input apparatus 20 creates the first structured data based on the input information, and transmits the created first structured data to the information processing apparatus 10. The input apparatus 20 is an example of an input unit for inputting the first structured data to the information processing apparatus 10.

[0019] The structured data storage device 30 is a storage medium that stores second structured data. The second structured data is, for example, each item included in the trouble list. Hereinafter, an item included in the trouble list is referred to as a trouble item. A trouble item may also be called a past trouble or a past trouble. The structured data storage device 30 may be a storage medium readable by a processor, such as a magnetic storage medium, an electromagnetic storage medium, an optical storage medium, or a semiconductor memory, or may be a driving device that reads and writes data to and from the above storage medium. The structured data storage device 30 transmits the second structured data to the information processing device 10 in response to an instruction from the information processing device 10, and updates the second structured data in response to the information transmitted from the information processing device 10. The structured data storage device 30 is an example of a storage unit that stores the second structured data, and may be mounted inside the information processing device 10.

[0020] The information processing device 10 acquires first structured data from the input device 20 and acquires second structured data from the structured data storage device 30. The information processing device 10 collates the first structured data with each second structured data, and updates the second structured data using the collation result. At this time, the information processing device 10 searches for collation data from the second structured data, and adds the content of the first structured data to the collation data. In the present embodiment, "collate" means comparing the contents between two pieces of data and determining whether the data are similar to each other, or determining the degree of similarity between the data. The collation data is, for example, second structured data similar to the first structured data. If the collation data is not found, the information processing device 10 creates new third structured data by converting the first structured data. The third structured data is, for example, created by converting a trouble report and is a new item (hereinafter referred to as a new item) added to the trouble list. The information processing device 10 transmits the updated collation data and the third structured data to the structured data storage device 30. Further, the information processing device 10 transmits the third structured data to the editing device 40.

[0021] The editing device 40 is a computer terminal for editing third-structured data. The editing device 40 includes a display device for displaying the editing screen for the third-structured data and an input device for inputting the editing content. The input device can include a keyboard, mouse, various switches, touchpad, touch panel display, etc. The display device displays various data. The display device can include a CRT (Cathode-Ray Tube) display, liquid crystal display, organic EL (Electro-Luminescence) display, LED (Light-Emitting Diode) display, plasma display, or any other display as appropriate. The display device may also be a projector.

[0022] The editing device 40 receives the third structured data from the information processing device 10 and displays an editing screen for editing the received third structured data. The user inputs the content to be edited for the third structured data while viewing the editing screen. The editing device 40 transmits the input information to the information processing device 10. The editing device 40 is an example of an input unit for inputting editing information for the second structured data to the information processing device 10.

[0023] The information processing device 10 updates the content of the third structured data based on the editing information acquired from the editing device 40, and transmits the updated third structured data to the structured data storage device 30.

[0024] Next, we will explain the configuration of the information processing device 10 in detail. Figure 6 shows an example of the hardware configuration of the information processing device 10. As shown in Figure 6, the information processing device 10 is a computer having a processor 1, memory 2, and communication device 3. Data and various signals are transmitted and received between the processor 1, memory 2, and communication device 3 via a bus.

[0025] Processor 1 is an integrated circuit that controls the overall operation of the information processing device 10. For example, processor 1 has a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a DSP (Digital Signal Processor), and / or an FPU (Floating-Point Unit). Processor 1 may also have internal memory and I / O interfaces. Processor 1 performs various processes by interpreting and calculating programs pre-stored in memory 2. Processor 1 may be partially or entirely implemented by hardware such as an ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array). Processor 1 may also be called a processing circuit.

[0026] Memory 2 is a storage medium that includes, for example, ROM (Read Only Memory), RAM (Random Access Memory), and auxiliary storage. ROM is a non-volatile memory that stores various types of data. For example, ROM stores data and settings used by processor 1 when executing various processes. ROM may also have a non-temporary computer-readable storage medium that stores programs executed by processor 1. RAM is a volatile memory used for reading and writing data. RAM temporarily stores data used by processor 1 when executing various processes. RAM provides the work area for processor 1. Auxiliary storage is a non-volatile memory that stores various types of data. For example, auxiliary storage stores data and settings used by processor 1 when executing various processes, data generated by various processes on processor 1, etc. Auxiliary storage consists of HDD (Hard Disk Drive), SSD (Solid State Drive), integrated circuit storage, etc. Auxiliary storage may also have a non-temporary computer-readable storage medium that stores programs executed by processor 1.

[0027] The communication device 3 includes a communication interface such as a network interface card (NIC) for data communication with various devices connected to the information processing device 10 via a network. Operation signals may be supplied from a computer connected via the communication device 3 or from an input device on that computer, and various data may be displayed on a display device on a computer connected via the communication device 3.

[0028] The information processing device 10 does not need to have all of the memory 2 and communication equipment 3. Some of the memory 2 and communication equipment 3 may be omitted as needed. The information processing device 10 may be equipped with any additional hardware equipment useful for executing the processing according to this embodiment. The information processing device 10 does not need to consist of a single physical computer; it may consist of a computer system having multiple computers connected via wired or network lines for communication. The assignment of the series of processes according to this embodiment to the multiple processors 1 implemented in each of the multiple computers can be arbitrarily configured. All processors 1 may execute all processes in parallel, or specific processes may be assigned to one or some of the processors 1, and the series of processes according to this embodiment may be executed by the entire computer system.

[0029] As shown in Figure 6, the processor 1 has a functional configuration including a structural analysis unit 11, a matching unit 12, an item creation unit 13, and an editing unit 14.

[0030] The structural analysis unit 11 analyzes the structure of the first structured data and the second structured data. "Analyzing the structure" means identifying one or more items that make up the structured data and determining the correspondence between the items. For example, the structural analysis unit 11 uses information about the items included in the first structured data to identify the item name and content of the first item that makes up the first structured data, and uses information about the items included in the second structured data to identify the item name and content of the second item that makes up the second structured data. Alternatively, the structure of the structured data may be analyzed using known methods to identify the first and second items.

[0031] Furthermore, the structural analysis unit 11 determines the correspondence between the first item and the second item for each item. The correspondence includes, for example, information indicating whether or not there is a correspondence between the first item and the second item, and information indicating the degree of correspondence between the first item and the second item. For example, for each combination of the first item and the second item, it is determined whether there is a "correspondence" or "no correspondence".

[0032] In determining the correspondence, the structural analysis unit 11 analyzes at least one of the following for each item in the first structured data and the second structured data: words, sentences, morphemes, and feature vectors. The degree of agreement of the analysis results is used to determine the correspondence between the first and second items. For example, the structural analysis unit 11 performs natural language analysis on each item in the first and second items, calculates the similarity between words, sentences, morphemes, or feature vectors contained in the first and second items, and determines whether the first and second items correspond by determining whether the calculated similarity is above a threshold.

[0033] Alternatively, the correspondence between the first and second items may be determined based on the similarity between the supplementary information attached to the first item and the supplementary information attached to the second item. For example, the structural analysis unit 11 calculates the similarity between the item name of the first item and the item name of the second item, and determines that the items correspond if the similarity is above a threshold.

[0034] Alternatively, the system may acquire correspondence information that predefines the relationship between the first and second items, and determine the relationship between the first and second items based on this correspondence information. For example, the structural analysis unit 11 acquires correspondence information that predefines the relationship between the first and second items, and uses this correspondence information as the result of determining the relationship.

[0035] Furthermore, an index indicating the degree of correspondence between the first and second items may be calculated.

[0036] The matching unit 12 matches the first structured data and the second structured data using the correspondence between the first item and the second item. For example, the matching unit 12 calculates the similarity between the first item and the second item for each item (hereinafter referred to as inter-item similarity), and uses the calculated inter-item similarity to calculate the similarity between the first structured data and the second structured data (hereinafter referred to as inter-data similarity). Subsequently, the matching unit 12 searches for the combination of the first structured data and the second structured data that has the highest inter-data similarity among the combinations where the inter-data similarity is above a threshold, and determines the found combination as matching data. On the other hand, if no combination with inter-data similarity above a threshold is found, the matching unit 12 determines that there is no matching data. The matching unit 12 generates matching information indicating either a combination of matching data or no combination of matching data, and transmits it to the item creation unit 13. The matching information is information that includes the matching result.

[0037] Item similarity is calculated, for example, using the degree of agreement between content items or the degree of agreement between supplementary information items. For example, the similarity of words and sentences contained in the content and supplementary information may be used, or the similarity of morphemes, words, feature vectors, etc., obtained by text analysis of the words and sentences may be used. Furthermore, item similarity may be calculated using both content and supplementary information.

[0038] Data similarity is calculated, for example, by integrating the item similarity of items included in the structured data. For example, the sum, mean, or median of the item similarity for each item can be used as data similarity. In calculating data similarity, for example, the matching unit 12 calculates data similarity using only items that have a corresponding relationship for both the first structured data and the second structured data. In this case, the matching unit 12 calculates item similarity for combinations of the first and second items that have a corresponding relationship, and uses only the calculated item similarity to calculate data similarity between the first structured data and the second structured data. Therefore, item similarity for combinations that do not have a corresponding relationship is not used in the calculation of data similarity. That is, items that do not have a corresponding relationship with all items in other structured data are not used in matching between structured data. Alternatively, data similarity may be calculated using only the item similarity for combinations that do not have a corresponding relationship.

[0039] In other methods for calculating inter-data similarity, the matching unit 12 may use only items that have a corresponding relationship in one of the first structured data and the second structured data, and use all items in the other to calculate inter-data similarity. For example, for a first item that has a corresponding relationship with a second item, the similarity with all second items is used to calculate inter-data similarity. In this case, only some items in the first item that have a corresponding relationship with a second item are used, but all items in the second item are used. Alternatively, for a second item that has a corresponding relationship with a first item, the similarity with all first items may be used to calculate inter-data similarity. In this case, only some items in the second item that have a corresponding relationship with a first item are used, but all items in the first item are used.

[0040] Furthermore, a method using only items that have a correspondence relationship between the first and second structured data, and a method using only items that have a correspondence relationship in one of the first and second structured data sets, and using all items in the other, may be used in combination. For example, the similarity between items in the first and second structured data sets that have a correspondence relationship, and the similarity between the first item and the entire second structured data set can be calculated, and these can be combined to calculate the similarity between data sets. As the similarity between the first item and the entire second structured data set, for example, numerical distance indicators or linguistic distance indicators can be used. Alternatively, the similarity between items in the first and second structured data sets that have a correspondence relationship, and the similarity between a specific second item and the first item can be calculated, and these can be combined to calculate the similarity between data sets. As a specific second item, for example, a second item for which similarity with the first item can be calculated, or a pre-defined second item can be used.

[0041] Furthermore, when using the degree of correspondence between the first and second items as the relationship, the similarity between corresponding items may be adjusted according to the degree of correspondence.

[0042] Furthermore, in addition to the matching results described above using the correspondence between the first and second items, the matching results of the entire first structured data and the entire second structured data may also be used in combination. For example, the similarity between the first and second items that have a correspondence, and the similarity between the entire first structured data and the entire second structured data can be calculated, and these can be integrated to calculate the similarity between the data.

[0043] Furthermore, matching information may be used as an auxiliary tool. This matching information may include, for example, past matching results or matching results from external matching devices. For instance, if a matching process has been performed in the past using structured data similar to the structured data being matched, those matching results may be used as an auxiliary tool. Additionally, matching results from other matching devices with matching capabilities may be used as an auxiliary tool.

[0044] Furthermore, when comparing the first structured data and the second structured data, the system may obtain other structured data linked to the second structured data (hereinafter referred to as related structured data) and use the related structured data indirectly. Related structured data is other structured data created for comparison. For example, if the first structured data is a trouble report and the second structured data is a trouble item, an FMEA sheet linked to the trouble item can be used as related structured data. An FMEA sheet is structured data that includes the results of an analysis of the impact of the trouble described in the trouble item on quality and manufacturing, and one FMEA sheet is linked to one trouble item. In this case, the comparison unit 12 can compare the related structured data (e.g., an FMEA sheet linked to each trouble item) with the first structured data (e.g., a trouble report) and use the comparison result as the comparison result between the first structured data (e.g., a trouble report) and the second structured data (e.g., a trouble item). Note that multiple related structured data may be linked to one second structured data. In this case, one of several related structured data sets linked to the second structured data can be selected and used.

[0045] The item creation unit 13 uses the matching information to create new items that conform to the format of structured data having different structures. For example, if matching data for the first structured data is found in the second structured data, the item creation unit 13 modifies the second structured data, which was found as matching data for the first structured data, based on the first structured data. On the other hand, if no matching data for the first structured data is found in the second structured data, the item creation unit 13 transforms the first structured data to create one or more third items and creates third structured data composed of the third items. The third item is a new item created by transforming the first structured data and consists of supplementary information and content. The third structured data is structured data composed of one or more third items. The item creation unit 13 transmits the created third items and third structured data to the editing unit 14.

[0046] The third structured data is created, for example, by converting the first structured data in accordance with the format of the existing second structured data. In this case, the item creation unit 13 creates one corresponding third item for each second item included in the second structured data. This creates the same number of third items as there are second items in the second structured data. Each third item corresponds to one second item and is created by converting the first structured data using the corresponding second item. For example, the item name of the corresponding second item can be used as the item name of the third item, and the text obtained by searching for and converting the part related to that item name from the first structured data can be created as the content of the third item.

[0047] Furthermore, when creating the third item, the correspondence between the first and second items may be used. For example, for each third item, the item name and content may be created using only the first item that has a corresponding relationship with the corresponding second item.

[0048] In creating the third structured data, the item creation unit 13 creates new items in accordance with the structure of existing trouble items. In this case, the item creation unit 13 creates new items corresponding to each second item that makes up the trouble item, uses the item name of the corresponding second item as the item name of the new item, and creates the content of the new item by searching for content related to the item name in the trouble report. If no related content is found when searching for content related to the item name of the new item in the trouble report, only the item name of the new item may be created, and the content of the new item may not be created. In this case, only implicit supplementary information may be created.

[0049] Furthermore, the item name for the third item may be the same as the item name for the corresponding second item, or a part of the item name for the corresponding second item may be used. Alternatively, the corresponding second item may be parsed or transformed and used to create the item name and content for the third item.

[0050] Furthermore, the item creation unit 13 may obtain format information regarding the format of the content of the second item from an external source and create the third item using the obtained format information. The format information includes, for example, the data type and format of the content of the second item. The format information is used to match the format of the third item, which is created using the first item, to the format of the second item when the format of the content of the first item and the format of the content of the second item are different. For example, a conversion dictionary that converts "hair contamination" to "foreign object contamination" can be used as format information. In this case, when creating a new trouble item using the content of the trouble report, the expression "hair contamination" used in the trouble report can be converted to the expression "foreign object contamination" used in the trouble list, thereby unifying the inconsistencies in notation between the first structured data and the second structured data. Alternatively, natural language processing may be used to unify the inconsistencies in notation instead of a conversion dictionary. In addition, conversion information that converts the display format of the year, month, and day may be used as format information. In this case, when creating a new trouble item using the contents of the trouble report, the date format can be standardized by converting the "2024 / 01 / 01" format used in the trouble report to the "01 / 01 / 2024" format used in the trouble list. The format information is used as the format for the contents of the third item to be created.

[0051] Furthermore, the item creation unit 13 may create the third structured data in a format different from that of the second structured data by creating the third item using structural information obtained from an external source. Structural information is information about the structure and format of the third structured data to be created. Structural information includes, for example, the number of items in the third item that constitutes the third structured data, the item name of the third item, the data type of the third item, and the display order and layout of the third item. The item creation unit 13 generates the third item and the third structured data using the first item of the first structured data and the structural information.

[0052] Furthermore, the item creation unit 13 may convert the first structured data into third items using statistical information of third items created in the past. For example, frequently used item names from the past can be extracted from the statistical information and used as the item names for the third items. The statistical information may include information of third items created in the past using the information processing device 10, or it may include information of third items created in the past without using the information processing device 10. In addition, the format and conversion method of the third items may be manually set by the user.

[0053] Furthermore, when creating the third item, the item creation unit 13 may use the first item of the first structured data as is. Alternatively, it may use the item name of the first item as the item name of the third item and use a part of the content of the first item as the content of the third item. Alternatively, it may extract only some of the first items from among the multiple first items included in the first structured data and use the extracted first items as the third item.

[0054] The editing unit 14 transmits the third item to the editing device 40 and displays an editing screen for editing the third item on the display unit of the editing device 40. The editing unit 14 also acquires the editing information entered by the user in the editing device 40 and edits the third item based on the editing information. Subsequently, the editing unit 14 creates third structured data composed of the edited third item and transmits the created third structured data to the structured data storage device 30. The editing unit 14 may also be called an update unit that updates the third item and the third structured data based on the editing information.

[0055] Furthermore, the editorial department 14 may display auxiliary information used for editing the third item on the editing screen using the first or second structured data. For example, the editorial department 14 may display the text of the first structured data on the editing screen for the third item. Alternatively, the editorial department 14 may highlight and display portions of the text of the first structured data that are related to the item name of the third item. In addition, suggested words or sentences to be entered as the content of the third item may be displayed on the editing screen to present the editing content to the user.

[0056] Alternatively, the first item of the first structured data may be used as the third item without converting the first structured data. In this case, the editorial department 14 displays the first item as the third item on the editing screen and edits the third item according to the editing content entered by the user.

[0057] Next, an example of the operation of the information processing device 10 will be described.

[0058] Figure 7 shows the processing steps for the matching process. The matching process involves comparing newly created first structured data with existing second structured data, and updating the second structured data or creating third structured data according to the matching results. Note that the processing steps for the automatic generation of the molecular structure of the complex molecule shown in Figure 7 are just one example, and this embodiment is not limited thereto; various elements can be deleted, added, and / or modified.

[0059] The following explanation uses a trouble management system for managing the details of problems that occur in the manufacturing industry as an example. The information processing system 100 is a system that compares newly created trouble reports with each trouble item included in the trouble list, and adds the details of the troubles described in the trouble reports to the trouble list based on the comparison results.

[0060] As shown in Figure 7, in the matching process, the structural analysis unit 11 first acquires the first structured data and the second structured data (step S101). At this time, the structural analysis unit 11 acquires a trouble report from the input device 20 and a trouble list containing multiple trouble items from the structured data storage device 30.

[0061] The input device 20 generates a trouble report regarding the newly occurring problem. The user enters the details of the newly occurring problem while viewing the input screen. The input device 20 creates a new trouble report based on the entered information and sends the created trouble report to the information processing device 10.

[0062] A trouble report is structured data consisting of one or more items, such as the circumstances under which the trouble occurred, the location where the trouble occurred, the cause of the trouble, and the subsequent actions taken. A trouble report is a reference document and is an example of the first type of structured data. The items in a trouble report are also examples of the first type of items. A trouble report is created for each incident of trouble. A trouble report may be a document created at the time the trouble occurred, or it may be a product development document that includes a description of the incident. A product development document is a document created during product design or product development and is structured data consisting of one or more items.

[0063] Figure 8 shows an example of a trouble report in the manufacturing industry. Each item in the trouble report consists of the item's content and its accompanying information. The item's content may include, for example, a free-form description of the trouble and images related to the trouble. Images related to the trouble may include, for example, an impression of an electronic seal or a photograph of the equipment where the trouble occurred. The accompanying information may include, for example, the name of the trouble, the date and time, the location of occurrence, the affected product, the circumstances of the occurrence, the initial response, the cause analysis, and permanent countermeasures. The structure of a trouble report may vary depending on the company and the person who prepared it. For example, the number and types of items included in a trouble report may vary depending on the company and the person who prepared it.

[0064] The structured data storage device 30 stores a trouble list containing multiple trouble items. The trouble list is a dataset that systematizes the content of troubles described in trouble reports created in the past, and lists trouble items created for each type of trouble. Each trouble item included in the trouble list is structured data consisting of one or more items such as the circumstances in which the trouble occurred, the location in which the trouble occurred, the cause of the trouble, and the subsequent actions taken. Each trouble item is document data created in the past for each incident of trouble. Each trouble item is a referenced document and an example of the second structured data, and the items of the trouble item are examples of the second items.

[0065] Figure 9 shows an example of a trouble item in the manufacturing industry. Each item that makes up a trouble item consists of the item's content and its accompanying information. The item's content is, for example, the description of the trouble written in free text. The accompanying information may include, for example, the name (title) of the trouble, the location where it occurred, the affected product, reference materials, a summary of the trouble, the cause of the trouble, the initial response, and permanent countermeasures. The structure of a trouble item may be set in advance when the trouble list is generated, or it may be set by the user when creating a new one.

[0066] Next, the structural analysis unit 11 analyzes the structure of the structured data acquired in step S101 (step S102). In this process, the structural analysis unit 11 identifies the items of the trouble report by analyzing the structure of the trouble report, and identifies the items of the trouble item by analyzing the structure of the trouble item.

[0067] Furthermore, the structural analysis unit 11 determines the correspondence between the items in the trouble report obtained through structural analysis and the items in the trouble item, item by item. For example, the structural analysis unit 11 calculates the similarity between items for each combination of items in the trouble report and items in the trouble item, and determines whether each combination "has a correspondence" or "does not have a correspondence" by determining a threshold for the similarity between items.

[0068] Next, the matching unit 12 matches the first structured data and the second structured data using the correspondence between items (step S103). At this time, the matching unit 12 matches the trouble report with each trouble item included in the trouble list and searches for matching data similar to the trouble report from among the trouble items included in the trouble list. For example, the matching unit 12 calculates the similarity between items in the trouble report and items in the trouble item that have been determined to have a correspondence, and uses the calculated similarity between items to calculate the similarity between data between the trouble report and each trouble item. After that, the matching unit 12 searches for the trouble item with the highest similarity between data, and whose similarity between data is above a threshold, and uses that item as matching data.

[0069] If matching data is found during the processing in step S103 (step S104-Yes), the item creation unit 13 updates the matching data using the first structured data (step S105). At this time, the item creation unit 13 adds the contents of the trouble report to the matching data. After that, the matching process is terminated.

[0070] On the other hand, if no matching data is found during the processing in step S103 (step S104-No), the item creation unit 13 creates a third item using the first structured data (step S106). For example, if there is no matching item in the trouble list, the item creation unit 13 creates a new item by converting the trouble report into the format of a trouble item. In this case, the item creation unit 13 creates the new item using the correspondence between the items in the trouble report and the items in the trouble item. Subsequently, the item creation unit 13 creates a new item using the new item. The new item is an example of a third item, and the new item is an example of third structured data.

[0071] Next, the editorial department 14 sends the newly created item to the editing device 40, which displays the editing screen on its display unit. The editing screen displays the item name and content of each newly created item.

[0072] Figure 10 shows an example of the editing screen 50. As shown in Figure 10, the editing screen 50 has a first display unit 51 and a second display unit 52. The first display unit 51 displays the item name and content of each new item that constitutes a new item. The second display unit 52 displays the item name and content of each item in the trouble report. In Figure 10, M items are displayed in the second display unit 52 and N items are displayed in the first display unit 51. The item names and content of the trouble report items in the second display unit 52 are displayed automatically. The item names of new items in the first display unit 51 are also displayed automatically. The content of new items in the first display unit 51 is entered by the editor. Alternatively, the content of new items may be pre-displayed based on the trouble report and made editable by the editor. Additionally, parts of the trouble report content displayed in the first display unit 51 that are related to the item name of the new item or parts used when the content of the new item was automatically created may be highlighted. Note that the second display unit 52 does not need to be displayed on the editing screen 50.

[0073] Furthermore, as shown in Figure 11, a third display unit 53 that presents candidate words to be entered into the content of a new item may be displayed. For example, as shown in Figure 11, multiple candidate item names may be extracted using the frequency of occurrence of each word in the trouble report, and the degree of relevance between the item name and the trouble report may be displayed for each candidate. Alternatively, the first display unit 51 may highlight and display the parts of the trouble report that are relevant to each candidate.

[0074] The user inputs the content of a new item while viewing the editing screen by operating the editing device 40. The editing unit 14 receives the input content from the editing device 40 and edits the new item according to the content (step S107).

[0075] Next, the item creation unit 13 creates a third structured data composed of the new items (step S108). At this time, the item creation unit 13 creates a new item using the edited new items and updates the trouble list by adding the created new item to the trouble list. After that, the information processing device 10 sends the updated trouble list to the structured data storage device 30 and terminates the matching process.

[0076] As described in the above embodiment, the information processing device 10 includes a structure analysis unit 11, a matching unit 12, an item creation unit 13, and an editing unit 14. The structure analysis unit 11 analyzes the structure of the first structured data and the second structured data, which are structured by one or more items, to obtain the first item included in the first structured data and the second item included in the second structured data, and determines the correspondence between the first item and the second item for each item. The matching unit 12 uses the above correspondence to compare the first structured data and the second structured data.

[0077] According to the above configuration, the structures of the first and second structured data, which have different structures, can be analyzed, the correspondence between items in the first and second structured data can be determined, and the first and second structured data can be compared using these correspondences. In this case, by performing the comparison using the correspondence between items, the accuracy of the comparison can be improved even when comparing documents with different formats. This reduces the burden on the user during the comparison process and reduces the amount of work required.

[0078] When the structure and format of the first and second structured data differ, simply comparing the entirety of the first and second structured data may not result in accurate matching. For example, trouble reports often include dates and specific names. On the other hand, trouble lists are generalized collections of multiple trouble cases with the same content, so trouble items often omit dates and use generalized names. Therefore, even if trouble reports and trouble items are documents for the same trouble, the types of items may differ, or items included in one may not be included in the other. In this way, when items are included only in one document, the similarity may be reduced due to differences in item types and formats, even if the content of the documents is similar. According to this embodiment, for example, by performing matching without using items that do not have a corresponding relationship with other items, it is possible to prevent the similarity from being reduced due to differences in format, even if the content of the documents is similar. Also, if the same word happens to be included in documents with low relationships, the similarity between documents may be increased. According to this embodiment, for example, by performing a comparison using only the similarity between items that have a corresponding relationship, it is possible to prevent the similarity between documents from becoming high even if the content of the documents is unrelated, simply because the same word is included in a section that does not have a corresponding relationship.

[0079] Furthermore, different terminology may be used in trouble reports and trouble items to refer to the same substance. In this way, when the same content is described using different expressions, the similarity between structured data may decrease even if the content is similar. According to this embodiment, for example, by using correspondence information that shows the correspondence between items and pre-setting items that contain similar content, it is possible to prevent the similarity between structured data from decreasing even when the content of the documents is similar but the expressions are different.

[0080] Furthermore, it may be possible to arbitrarily select the items to be used for matching from the first and second items. For example, if the user wants to perform matching focusing on the location where a problem occurred, the first and second items obtained through structural analysis can be presented to the user, and the user can select the items to be used for matching, allowing them to search by focusing on any item. In addition, the importance level when determining correspondence relationships or performing matching may be set for each item. In this case, the user can determine correspondence relationships and search for matching data while considering the importance level of the items. According to the above configuration, it is possible to narrow down the matching to any item within the structured data, or to weight the degree of attention given to each item within the structured data and perform matching.

[0081] Furthermore, the system may present the user with the results of the correspondence determination and the similarity scores between the items used in the determination, and allow the user to modify the correspondence determination results. In addition, during matching, the system may present the user with the search results of the matching data and the similarity scores between the data used, and allow the user to modify the matching results.

[0082] Furthermore, the information processing device 10 further includes an item creation unit 13. When matching data similar to the first structured data is found in the second structured data, the item creation unit 13 appends the contents of the first structured data to the matching data. With the above configuration, for example, by appending the contents of the source trouble report to the target trouble item, it is possible to support the appending work to the trouble list.

[0083] Furthermore, if no matching data is found, the item creation unit 13 creates a third item converted from the first structured data, and a third structured data composed of the third item. With the above configuration, for example, the process of creating new items in the trouble list can be supported by converting the source trouble report into a new item that conforms to the format of the target trouble item, and creating a new item composed of the new item.

[0084] Furthermore, when creating the third item, the correspondence between the first and second items may be used. For example, the item name of the third item can be the same as the item name of the second item, and the supplementary information and content of the third item can be created using the first item that has a corresponding relationship with the second item. In this case, since the third item is created using only the first item related to the item name, the accuracy of the creation of the third item is improved.

[0085] The information processing device 10 also includes an editing unit 14. The editing unit 14 displays an editing screen for editing the third item on the display unit of the editing device 40, and edits the third item based on the editing information for the third item obtained from the editing device 40. With the above configuration, it is possible to support the creation of new structured data by editing the third item based on the editing information. Note that the processing by the editing unit 14 may be omitted. The editing unit 14 may also display auxiliary information used for editing the third item on the editing screen using the first structured data or the second structured data. For example, by displaying the contents of the first structured data or input candidates on the editing screen, it is possible to support the user's editing work.

[0086] (Second embodiment) In the first embodiment, an example was described of a document management system that adds trouble reports to a trouble list, where the trouble reports and each trouble item included in the trouble list are compared. However, the configuration of the information processing device 10 and information processing system 100 of the first embodiment may also be applied to a system that compares infrastructure-related equipment data.

[0087] For example, the correspondence between items in an elevator maintenance work field report and items in a trouble management record ledger can be determined item by item. This correspondence can then be used to compare the field work report with each trouble management record, and existing trouble management records can be updated or new trouble management records can be created based on the comparison results. In this case, the field work report corresponds to the first structured data source, the existing trouble management records in the trouble management record ledger correspond to the second structured data source, and the newly created trouble management records correspond to the third structured data source.

[0088] (Third embodiment) Furthermore, the configuration of the information processing device 10 and information processing system 100 of the first embodiment may be applied to a system for comparing power system data and plant system data.

[0089] For example, the correspondence between items in the plant's maintenance and inspection history and items in the trouble management records can be determined item by item. This correspondence can then be used to compare the maintenance and inspection history with each trouble management item, and existing trouble management items can be updated or new trouble management items can be created based on the comparison results. In this case, the maintenance and inspection history corresponds to the first structured data of the source, existing trouble management items in the trouble management records correspond to the second structured data of the reference destination, and newly created trouble management items correspond to the third structured data.

[0090] Thus, according to this embodiment, it is possible to provide an information processing device, method, and program that can support the matching of structured data of different formats.

[0091] While several embodiments of the present invention have been described, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These novel embodiments can be carried out in a variety of other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims and their equivalents. [Explanation of Symbols]

[0092] 100... Information processing system, 10... Information processing device, 20... Input device, 30... Structured data storage device, 40... Editing device, 50... Editing screen, 1... Processor, 2... Memory, 3... Communication equipment, 11... Structural analysis unit, 12... Verification unit, 13... Item creation unit, 14... Editing unit.

Claims

1. A structural analysis unit analyzes the structure of first structured data and second structured data, which are structured by one or more items, to obtain the first item included in the first structured data and the second item included in the second structured data, and determines the correspondence between the first item and the second item for each item. A matching unit that compares the first structured data and the second structured data using the aforementioned correspondence relationship, An information processing device equipped with the following:

2. The structural analysis unit analyzes at least one of the following for each of the first structured data and the second structured data: a word, a sentence, a morpheme, and a feature vector. Using the analysis results, it determines the correspondence between the first item and the second item. The information processing apparatus according to claim 1.

3. The structural analysis unit determines the correspondence between the first item and the second item based on the degree of agreement between the supplementary information attached to the first item and the supplementary information attached to the second item. The information processing apparatus according to claim 1.

4. The structural analysis unit acquires correspondence information that predefines the correspondence between the first item and the second item, and determines the correspondence between the first item and the second item based on the correspondence information. The information processing apparatus according to claim 1.

5. The aforementioned second structured data is data generated in the past, The structural analysis unit acquires a plurality of the second structured data and determines the correspondence between the first item and the second item for each of the plurality of second structured data. The information processing apparatus according to claim 1.

6. The system further comprises a storage unit for recording the first structured data and the second structured data. The information processing apparatus according to claim 1.

7. The matching unit calculates the similarity between items for each combination of the first and second items that have a corresponding relationship, and uses the similarity between items to compare the first structured data and the second structured data. The information processing apparatus according to claim 1.

8. The matching unit calculates the similarity between items based on the degree of agreement between the analysis results of the first and second items that have a corresponding relationship. The information processing apparatus according to claim 7.

9. The matching unit matches the first structured data with the second structured data using at least one of the similarity between the first item and the second item that has a corresponding relationship with the second item, and the similarity between the second item and the first item that has a corresponding relationship with the first item. The information processing apparatus according to claim 1.

10. The matching unit obtains related structured data linked to the second structured data, and matches the first structured data with the related structured data, thereby matching the first structured data with the second structured data. The information processing apparatus according to claim 1.

11. The system further comprises a third item obtained by converting the first structured data, and an item creation unit that creates a third structured data composed of the third item. The information processing apparatus according to claim 1.

12. The item creation unit creates a third item corresponding to each second item, and creates supplementary information or content using the first item that has a corresponding relationship with the second item corresponding to the third item. The information processing apparatus according to claim 11.

13. The item creation unit acquires format information relating to the format of the content of the second item, and uses the format information to create the third item. The information processing apparatus according to claim 11.

14. The item creation unit acquires structural information relating to the structure of the structured data, and creates the third item from the first structured data based on the structural information. The information processing apparatus according to claim 11.

15. The system further includes an editing unit that displays the third item on a display unit and edits the third item based on editing information for the third item, The editorial department displays auxiliary information for editing on the display unit using the first structured data or the second structured data. The information processing apparatus according to claim 11.

16. If the item creation unit finds a second structured data similar to the first structured data through the matching process, it edits the content of the second structured data based on the content of the first structured data. The information processing apparatus according to claim 11.

17. The structural analysis unit analyzes the structure of the first structured data and the second structured data, which are structured by one or more items, to obtain the first item included in the first structured data and the second item included in the second structured data, and determines the correspondence between the first item and the second item for each item. The matching unit compares the first structured data and the second structured data using the correspondence relationship, A method that includes the following.

18. In the processor, The system includes a function to analyze the structure of a first structured data and a second structured data, which are structured by one or more items, to obtain the first item included in the first structured data and the second item included in the second structured data, and to determine the correspondence between the first item and the second item for each item. The matching unit has the function of matching the first structured data and the second structured data using the correspondence relationship, A program to achieve this.

Images