Information processing apparatus, information processing method, and program

Abstract

To provide an information processing apparatus, an information processing method, and a program capable of properly protecting personal information while smoothly implementing a certification process of long-term care insurance.SOLUTION: An information processing apparatus is configured to: acquire, from an administrative terminal, electronic data of a long-term care certification committee material, which is a review material in a certification process of long-term care insurance; classify unique expressions included in sentences of the acquired long-term care certification committee material; determine whether the classification requires masking or not based on masking setting information stored in advance; mask the unique expressions according to the content set in advance, based on a result of the determination; and provide a long-term care certification committee material obtained after the masking to the administrative terminal.SELECTED DRAWING: Figure 7

Description

[Technical Field]

[0001] The present invention relates to an information processing device, an information processing method, and a program. [Background technology]

[0002] Various systems have been proposed for preventing the leakage of personal information. For example, Patent Document 1 discloses a method for protecting personal information stored in a character input assistance server. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 629853 Summary of the Invention [Problem to be solved by the invention]

[0004] However, the technology described in Patent Document 1 may lack the flexibility to appropriately protect only personal information while leaving information necessary for the work intact. For example, when protecting personal information contained in materials for the nursing care certification examination committee, which are examination documents for the nursing care insurance certification work, it is necessary to mask only the personal information while leaving the information necessary for the nursing care insurance certification work intact. Therefore, there is room for improvement from the perspective of appropriately protecting personal information while realizing the smooth execution of the nursing care insurance certification work.

[0005] The present invention has been made in consideration of the above-mentioned circumstances, and aims to provide an information processing device, information processing method, and program that can effectively protect personal information while enabling the smooth execution of nursing care insurance certification work. [Means for solving the problem]

[0006] In order to achieve the above object, an information processing device according to a first aspect of the present invention comprises: a means for acquiring, from an administrative terminal, electronic data of materials for the nursing care certification examination committee, which are examination documents for the nursing care insurance certification work; a classification determination means for classifying named entities contained in sentences described in the nursing care certification examination committee materials acquired by the nursing care certification examination committee material acquisition means; a masking target determination means for determining whether the classification of the named entity determined by the classification determination means is a classification to be masked based on pre-stored masking setting information; a masking processing means for masking the named entity in accordance with preset content based on the determination result by the masking target determination means; an output means for providing the nursing care certification examination committee materials after the masking processing means has performed the masking to the administrative terminal; Equipped with.

[0007] the masking processing means performs masking by converting the named entity into content specified by the masking setting information; This may be done.

[0008] the masking processing means sets the display mode of the named entity after the masking to a specific mode that is easy for a user to visually recognize. This may be done.

[0009] the classification determination means performs a morphological analysis of the sentences described in the nursing care certification examination committee materials and classifies the named entities using a learning model that has been machine-learned in advance by supervised learning; the masking processing means changes the display mode of the named entity after the masking process depending on the reliability of the classification of the named entity using the learning model in the classification determination means. This may be done.

[0010] When the named entity includes a predetermined specific character string, the masking processing means performs masking by converting the named entity into a special character string different from the content specified by the masking setting information. This may be done.

[0011] In order to achieve the above object, an information processing method according to a second aspect of the present invention comprises: An information processing method by an information processing device, a step of acquiring nursing care certification examination committee materials from an administrative terminal, the electronic data of which is a document for examination in nursing care insurance certification work; a classification determination step of classifying named entities included in the sentences described in the nursing care certification examination committee materials acquired in the nursing care certification examination committee materials acquisition step; a masking target determination step of determining whether the classification of the named entity determined in the classification determination step is a classification to be masked based on pre-stored masking setting information; a masking process step of masking the named entity in accordance with preset content based on the determination result in the masking target determination step; an output step of providing the nursing care certification examination committee materials after the masking process in the masking processing step to the administrative terminal; Equipped with.

[0012] In order to achieve the above object, a program according to a third aspect of the present invention comprises: Computer, A means for acquiring nursing care certification examination committee materials from an administrative terminal, the electronic data of nursing care certification examination committee materials being examination documents in nursing care insurance certification work; a classification determination means for classifying named entities contained in sentences described in the nursing care certification examination committee materials acquired by the nursing care certification examination committee material acquisition means; a masking target determination means for determining whether the classification of the named entity determined by the classification determination means is a classification to be masked based on pre-stored masking setting information; a masking processing means for masking the named entity in accordance with preset content based on the determination result by the masking target determination means; an output means for providing the nursing care certification examination committee materials after the masking process by the masking processing means to the administrative terminal; Function as. [Effects of the Invention]

[0013] According to the present invention, it is possible to realize smooth execution of nursing care insurance certification work while suitably protecting personal information. [Brief explanation of the drawings]

[0014] [Figure 1] FIG. 1 is a block diagram illustrating an example of an information processing system. [Figure 2] FIG. 1 is a block diagram illustrating an example of an information processing device. [Figure 3] FIG. 10 is an explanatory diagram illustrating an example of masking setting information. [Figure 4] FIG. 10 is an explanatory diagram showing an example of nursing care certification review board materials before masking. [Figure 5] 10 is a flowchart illustrating an example of processing by the information processing system. [Figure 6] 10 is a flowchart illustrating an example of a classification determination process. [Figure 7] FIG. 10 is an explanatory diagram showing an example of nursing care certification review board materials after masking. [Figure 8] FIG. 10 is an explanatory diagram showing an example of a confirmation and correction screen. [Figure 9] FIG. 10 is an explanatory diagram showing an example of reliability setting in a modified example. [Figure 10] FIG. 10 is an explanatory diagram showing an example of a nursing care certification review board document after masking in a modified example. [Figure 11] FIG. 10 is an explanatory diagram showing an example of specific character string setting information in a modified example. [Figure 12] FIG. 10 is an explanatory diagram showing an example of a confirmation and correction screen in a modified example. DETAILED DESCRIPTION OF THE INVENTION

[0015] (Embodiment) An information processing device, an information processing method, and a program according to an embodiment of the present invention will be described in detail with reference to the drawings. The same or corresponding parts in the drawings are denoted by the same reference numerals. The following description will be given using an example in which the information processing device of the present invention is applied to an information processing system 1 shown in FIG.

[0016] As shown in Fig. 1, the information processing system 1 is composed of an administrative terminal 200 and an information processing device 100, which are communicatively connected via a network 210. Although the illustrated example shows a single administrative terminal 200, there may be multiple administrative terminals. The network 210 is a dedicated network such as a comprehensive administrative network.

[0017] The administrative terminal 200 is, for example, an information terminal (so-called computer) such as a smartphone, tablet, or PC (Personal Computer) at an administrative agency that performs nursing care certification work in a local government, and is capable of sending and receiving various data with the information processing device 100 via the network 210.

[0018] The information processing device 100 is an information terminal such as a smartphone, tablet, or PC, and is capable of transmitting and receiving various data to and from an administrative terminal 200 via a network 210. The information processing device 100 has a function of masking only personal information from the information contained in the nursing care certification review board materials transmitted from the administrative terminal 200, while leaving the information necessary for nursing care certification intact.

[0019] Next, the configuration of the information processing device 100 will be described with reference to FIG.

[0020] As shown in FIG. 2, the information processing device 100 includes a storage unit 110, a control unit 120, an input / output unit 130, a communication unit 140, and a system bus (not shown) that interconnects these units.

[0021] The storage unit 110 includes a ROM (Read Only Memory), a RAM (Random Access Memory), etc. The ROM stores a program 111 executed by a CPU (Central Processing Unit) of the control unit 120, various data (not shown) required in advance for executing the program 111, masking setting information 112, and a learning model 113.

[0022] The program 111 is a program for executing the processing of the information processing system described later, and is stored in the storage unit 110 in advance.

[0023] The masking setting information 112 is a list of information (see FIG. 3) for determining whether or not to perform masking and for specifying the character string after masking, and is stored in advance in the storage unit 110. The contents of the masking setting information 112 can be changed by the user as appropriate.

[0024] As shown in Fig. 3, the masking setting information 112 includes information such as "classification," "whether or not to mask," and "character string after masking." Specifically, "classification" indicates the classification of words included in the nursing care certification examination committee materials, and the classification is determined by AI (Artificial Intelligence) trained using the learning model 113 using the function of the classification determination unit 122, which will be described later. For example, in the case of "personal name" shown in Fig. 3, if a word included in the nursing care certification examination committee materials is a person's name (e.g., "Yamada Yōsan"), the classification determination unit 122 determines that "Yamada Yōsan" belongs to the "personal name" category.

[0025] "Masking on / off" is information indicating whether or not masking is performed, and is set by the user in advance. In the example shown in FIG. 3, "masking" is set for categories such as "personal name," "age," and "gender," while "no masking" is set for the "disease name" category. "Masked character string" indicates the character string to be displayed after masking. In the example shown in FIG. 3, for example, the word "Yama-Yo-Otoko," which is included in the nursing care certification review board document and is classified as "personal name," is replaced (masked) with a character string such as "personal name." Note that the masked character string can be changed as appropriate. It is not limited to character strings, and may be symbols or numbers. Furthermore, any form, such as a blank space or a black marker, may be used as long as the unmasked word is not visible, and the display form may be set in advance.

[0026] The learning model 113 is a learning model used when the classification determination unit 122 (described later) performs classification. The learning model 113 is an AI model that has learned information necessary for nursing care certification through supervised learning. More specifically, the learning model 113 is an AI model that has learned, through supervised learning, to which nursing care classification a word to be classified belongs, such as "XX Clinic" belonging to the "medical institution name" classification and "Neosty Minamiyamada" belonging to the "facility name" classification. In this embodiment, the learning model 113 is trained in two stages: pre-training using a large amount of text data and post-training specialized for named entity extraction. In the pre-training, Japanese morphemes are quantified (vectorized), and in the post-training, the model is trained to output a determination result as to whether the quantified token vector is a named entity (such as a person's name, company name, or facility name) or something else. The learning model 113 is, for example, a natural language processing model such as BERT (Bidirectional Encoder Representations from Transformers). The learning model 113 is not limited to a model that performs two stages of pre-learning and post-learning, but may be, for example, a model that performs only pre-learning or a model that performs only post-learning.

[0027] The control unit 120 is configured with a CPU, an ASIC (Application Specific Integrated Circuit), etc. The control unit 120 operates in accordance with a program 111 stored in the storage unit 110, and executes processing in accordance with the program 111. The control unit 120 includes, as main functional units provided by the program 111 stored in the storage unit 110, a nursing care certification examination board material acquisition and conversion unit 121, a classification determination unit 122, a masking target determination unit 123, a masking processing unit 124, and a correction confirmation output processing unit 125.

[0028] The long-term care certification review board material acquisition and conversion unit 121 is a functional unit that acquires long-term care certification review board materials from the administrative terminal 200. The long-term care certification review board materials include the initial assessment result, the certification survey form, and the attending physician's opinion. The initial assessment result is the result of an initial assessment based on the certification survey form and the attending physician's opinion, and is the result of the assessment made by a system introduced by the administration. The certification survey form contains the details that the investigator has heard from the long-term care certification recipient, and includes items such as an overview survey and special notes. The attending physician's opinion is written by the attending physician of the long-term care certification recipient. The long-term care certification review board materials are a document that compiles all of these, and the long-term care certification review board material acquisition and conversion unit 121 has the function of receiving the electronic file (PDF file) of the long-term care certification review board materials.

[0029] The long-term care certification review board material acquisition and conversion unit 121 also has a function of converting long-term care certification review board materials acquired from the administrative terminal 200 from image data to character data by inputting the materials into an OCR (Optical Character Reader). The conversion of character data by the long-term care certification review board material acquisition and conversion unit 121 may be performed by an AIOCR. The conversion to character data by the long-term care certification review board material acquisition and conversion unit 121 may be performed by an external functional unit via the network 210, rather than by the long-term care certification review board material acquisition and conversion unit 121. The long-term care certification review board material acquisition and conversion unit 121 may acquire, for example, data converted from image data to character data on the administrative terminal 200 side, i.e., the long-term care certification review board materials converted into character data, from the administrative terminal 200.

[0030] The classification determination unit 122 is a functional unit that classifies terms included in the nursing care certification examination committee materials acquired by the nursing care certification examination committee material acquisition and conversion unit 121 and converted into character data. Specifically, the classification determination unit 122 is a functional unit that classifies named entities included in sentences in the nursing care certification examination committee materials using a named entity extraction function such as BERT. That is, the classification determination unit 122 has a function to extract named entities included in a given sentence (nursing care certification examination committee materials) and classify the named entities into predefined labels. For example, if the nursing care certification examination committee materials contain a statement such as "After being admitted to Neosty Minamiyamada in 1997, he was admitted to the current paid nursing home, Best Life Yamada, in July 2022," the classification determination unit 122 uses the learning model 113 to perform classification using AI and determines that "Neosty Minamiyamada" and "Home Best Life Yamada" belong to the "facility name" category. The classification determination unit 122 may also function as a classification acquisition unit. That is, the classification determination may be performed by an external functional unit via the network 210, rather than by the classification determination unit 122 in the information processing device 100, and the classification acquisition unit may acquire the classification determined by the external functional unit.

[0031] The masking target determination unit 123 is a functional unit that determines whether or not to mask words in the category determined by the category determination unit 122. Specifically, the masking target determination unit 123 determines whether or not to mask words in the category determined by the category determination unit 122, based on the settings of the masking setting information 112. For example, in the case where the settings of the masking setting information 112 are those shown in FIG. 3, as described above, if the category determined by the category determination unit 122 is a category such as "person's name," "age," or "gender," the masking target determination unit 123 determines that words in that category should be masked. On the other hand, if the category determined by the category determination unit 122 is a category of "disease name," the masking target determination unit 123 determines that words in that category should not be masked. For example, if the classification determination unit 122 determines that "Neosty Minamiyamada" and "Home Best Life Yamada" belong to the "facility name" classification, the masking target determination unit 123 determines that "Neosty Minamiyamada" and "Home Best Life Yamada" should be masked based on the settings of the masking setting information 112 shown in Figure 3.

[0032] Returning to FIG. 2, the masking processing unit 124 is a functional unit that masks words determined to be masked based on the determination result by the masking target determination unit 123. Specifically, the masking processing unit 124 performs masking by changing the words determined to be masked by the masking target determination unit 123 to "masked character strings" set in the masking setting information 112. For example, if the category determination unit 122 determines that "Neosty Minamiyamada" and "Home Best Life Yamada" belong to the category of "facility name," the masking target determination unit 123 determines that "Neosty Minamiyamada" and "Home Best Life Yamada" are to be masked. Then, the masking processing unit 124 performs masking by changing "Neosty Minamiyamada" and "Home Best Life Yamada" to "facility name," respectively, based on the settings of the masking setting information 112 shown in FIG. 3. In this embodiment, the masking processing unit 124 displays the masked character string in a conspicuous manner (for example, in a specific manner such as bold, red, or underlined) so that the user can easily see that the masking has been performed. The masking processing unit 124 also has a function of notifying the administrative terminal 200 that the masking of the nursing care certification examination committee materials has been completed.

[0033] The correction confirmation output processing unit 125 is a functional unit that has the functions of providing the administrative terminal 200 with a confirmation screen for the nursing care certification review board materials after masking by the masking processing unit 124, accepting and executing corrections, and outputting the results. Specifically, the correction confirmation output processing unit 125 is a functional unit that displays, in parallel on the administrative terminal 200, the nursing care certification review board materials after masking by the masking processing unit 124 and the acquired nursing care certification review board materials before masking. The correction confirmation output processing unit 125 is also a functional unit that corrects the contents of the masked nursing care certification review board materials based on correction instructions from the administrative terminal 200. Furthermore, the correction confirmation output processing unit 125 is a functional unit that outputs the results by providing the masked nursing care certification review board materials as a PDF file in response to a request from the administrative terminal 200.

[0034] The input / output unit 130 is a device that is configured with a keyboard, a mouse, a camera, a microphone, a liquid crystal display, an organic EL (Electro-Luminescence) display, and the like, and is used to input and output various types of data.

[0035] The communication unit 140 is a device that enables the information processing device 100 to communicate with other information terminals such as the administrative terminal 200 via the network 210. The configuration of the information processing device 100 has been described above.

[0036] Next, the operation of the information processing system 1 will be described with reference to Figs. 4 to 8. In this example, a case where masking is performed on the nursing care certification examination committee document shown in Fig. 4 will be described as an example. Specifically, in this embodiment, as shown in Fig. 4, a non-masking area X, which is an area that is not to be masked, and a masking area Y, which is an area that is to be masked, are set in advance. Note that the non-masking area X and the masking area Y can be arbitrarily changed by the user's settings, and the entire area may be set as the masking area Y.

[0037] 5 is a flowchart showing an example of processing of the information processing system 1. First, processing of the administrative terminal is started by a user's operation on the administrative terminal 200. Specifically, processing of the administrative terminal is started by performing an operation on the administrative terminal 200 that is necessary to execute a masking process on the nursing care certification review board materials.

[0038] When the processing of the administrative terminal is started, the administrative terminal 200 makes a processing procedure request by transmitting request information to the information processing device 100 to request the display of a processing screen required to execute the masking process of the nursing care certification review board materials (step S11). Specifically, in the processing of step S11, the administrative terminal 200 transmits the request information to the information processing device 100, for example, by operating to start an application for uploading the nursing care certification review board materials.

[0039] When the information processing device 100 receives the request information, the information processing device starts processing. When the information processing device starts processing, the control unit 120 of the information processing device 100 transmits a processing screen to provide the administrative terminal 200 with a processing screen for uploading the nursing care certification review board materials (step S12).

[0040] Next, when the administrative terminal 200 receives the processing screen, the processing screen is displayed on the administrative terminal 200, and by uploading the nursing care certification review board materials (step S13), a PDF file of the nursing care certification review board materials is sent to the information processing device 100.

[0041] On the information processing device 100 side, the control unit 120 receives the PDF file (electronic data) of the nursing care certification review board materials using the function of the nursing care certification review board material acquisition and conversion unit 121, and converts the received PDF file of the nursing care certification review board materials from image data to character data by OCR input (step S14). The conversion of character data in the processing of step S14 may be performed by the AIOCR or may be performed by an external functional unit via the network 210. The nursing care certification review board material acquisition and conversion unit 121 that receives the PDF file of the nursing care certification review board materials and this step correspond to a nursing care certification review board material acquisition means and a nursing care certification review board material acquisition step, respectively.

[0042] After executing the process of step S14, the control unit 120 executes a classification determination process (step S15) using the function of the classification determination unit 122 to determine the classification of the content described in the nursing care certification review board document converted into character data in the process of step S14. Fig. 6 is a flowchart showing an example of the classification determination process executed in the process of step S15 in Fig. 5. The classification determination unit 122 that executes the process of step S15 and step S15 correspond to the classification determination means and the classification determination step, respectively.

[0043] 6 starts, the control unit 120 uses the function of the classification determination unit 122 to identify a target area that is an area to be classified, that is, a masking target area Y, in the nursing care certification review board materials converted into character data (step S31). Specifically, in the processing of step S31, the classification determination unit 122 identifies the masking target area Y in the nursing care certification review board materials according to a preset setting.

[0044] After executing the process of step S31, the control unit 120 performs morphological analysis on the content (such as a sentence) written in the masking target area Y using the function of the classification determination unit 122 (step S32). For example, using the nursing care certification review board document shown in FIG. 4 as an example, in the process of step S32, the classification determination unit 122 performs morphological analysis on each of the two sentences included in the masking target area Y of the certification survey form and the two sentences included in the masking target area Y of the attending physician's opinion. For example, the statement "After being admitted to Neosty Minamiyamada in 1997, he was admitted to the current paid nursing home, Best Life Yamada, in July 2022" shown in FIG. 4 is divided into its respective morphemes by executing the process of step S32, such as "After being admitted to Neosty Minamiyamada in 1997, he was admitted to the current paid nursing home, Best Life Yamada, in July 2022."

[0045] 6, the control unit 120 uses the function of the classification determination unit 122 to classify the character string determined to be a "noun" by morphological analysis in the processing of step S32 into named entities based on the learning model 113 (step S33), and terminates the classification determination processing. Specifically, in the processing of step S33, the classification determination unit 122 converts each morpheme divided in the processing of step S32 into a vector and determines the classification for each morpheme. More specifically, it classifies whether the term is related to nursing care certification (a named entity frequently used in nursing care certification), and if it is related to nursing care certification, it determines which category it belongs to based on the learning model 113. For example, taking the certification questionnaire in the nursing care certification examination committee materials shown in Figure 4 as an example, after morphological analysis is performed in the process of step S32, the classification determination unit 122 performs classification using AI with the learning model 113 in the process of step S33, and determines that terms such as "cerebral injury," "subarachnoid hemorrhage," "higher brain dysfunction," "Neosty Minamiyamada," and "Home Best Life Yamada" are named entities commonly used in nursing care certification and belong to the categories of "disease name," "disease name," "disease name," "facility name," and "facility name," respectively. Note that the named entities include personal information such as people's names and facility names, as well as age, gender, place names, and occupation.

[0046] Returning to FIG. 5, after executing the process of step S15, the control unit 120 executes masking using the function of the masking processing unit 124 (step S16). Specifically, in the process of step S16, the control unit 120 first determines, using the function of the masking target determination unit 123, whether each of the categories determined in the process of step S15 is a target for masking, based on the masking setting information 112. If a target for masking is determined, the control unit 120 performs masking by converting the named entity belonging to the category into a masked character string set in the masking setting information 112 using the function of the masking processing unit 124. For example, using the certification survey form in the nursing care certification review board materials shown in FIG. 4 as an example, in the process of step S16, the masking target determination unit 123 determines not to mask "cerebral injury," "subarachnoid hemorrhage," and "higher brain dysfunction" belonging to the "disease name" category (see FIG. 3), and determines to mask "Neosty Minamiyamada" and "Home Best Life Yamada" belonging to the "facility name" category (see FIG. 3). Then, the masking processing unit 124 converts "Neosty Minamiyamada" and "Home Best Life Yamada," which are determined to be masked, into the "facility name" that is the masked character string in the masking setting information 112 shown in FIG. 3, thereby performing masking. By performing the processing of step S16 in this manner, the masking target in the nursing care certification review board document shown in FIG. 4 is masked as shown in FIG. 7. FIG. 7 shows an example of the nursing care certification review board document after masking. As shown in the figure, when the masking processing unit 124 converts the masked character string into the masked character string, the converted character string may be displayed in a conspicuous manner so that it is easily recognized by the user. The masking processing unit 124 and step S16, which perform the processing to determine whether each classification is a masking target in the processing of step S16, correspond to a masking target determination means and a masking target determination step, respectively. The masking processing unit 124 and step S16, which perform the masking in the processing of step S16, correspond to a masking processing means and a masking processing step, respectively.

[0047] After executing the processing of step S16 shown in Figure 5, the control unit 120 notifies the administrative terminal 200 that masking has been completed by sending a completion notification indicating that masking has been completed to the administrative terminal 200 using the function of the masking processing unit 124 (step S17).

[0048] When the administrative terminal 200 receives the completion notification, it sends a confirmation / correction request to the information processing device 100, requesting a confirmation / correction screen, which is a screen for confirming and correcting the nursing care certification review board materials after masking, through input operations by the user (step S18).

[0049] When the information processing device 100 receives the confirmation / edit screen request, the control unit 120 uses the function of the edit confirmation output processing unit 125 to transmit a confirmation / edit screen, which is a screen for confirming and editing the masked nursing care certification review board materials, to the administrative terminal 200, thereby providing the confirmation / edit screen to the administrative terminal 200 (step S19). Specifically, in the processing of step S19, the edit confirmation output processing unit 125 provides the administrative terminal 200 with a confirmation / edit screen that displays the nursing care certification review board materials before and after masking side by side, as shown in FIG. 8. As a result, the nursing care certification review board materials before and after masking are displayed side by side on the administrative terminal 200, as shown in FIG. 8. Note that FIG. 8 is an explanatory diagram showing an example of a confirmation / edit screen, illustrating an example in which the nursing care certification review board materials before masking shown in FIG. 4 and the nursing care certification review board materials after masking shown in FIG. 7 are displayed side by side. In this way, the nursing care certification review board materials before and after masking are displayed side by side, making it easier for the user to confirm the legitimacy of the masking and to make corrections. The correction confirmation output processing unit 125 that executes the processing of step S19 and step S19 correspond to an output means and an output step, respectively. In this embodiment, an example has been shown in which a confirmation and correction screen, which is a screen for confirming and correcting the masked nursing care certification review board materials, is transmitted to the administrative terminal 200 based on the reception of a confirmation and correction request that requests a confirmation and correction screen, but the transmission may be independent of a confirmation and correction request. For example, the confirmation and correction screen may be transmitted to the administrative terminal 200 in the processing of step S17.

[0050] 5 is executed, and the nursing care certification examination committee materials before and after masking are displayed on the administrative terminal 200. After that, if the masked nursing care certification examination committee materials need to be corrected, that is, if the user visually determines that corrections are necessary, the user performs an input operation to instruct the corrections (step S20). Specifically, in the processing of step S20, the user performs an input operation to input corrections to the masked nursing care certification examination committee materials, and the contents are transmitted to the information processing device 100 as correction instructions. Note that if no corrections are required, the processing of step S20 may be skipped.

[0051] When the process of step S20 is executed and the correction instruction is sent, the control unit 120 on the information processing device 100 side corrects the masked nursing care certification review board materials in accordance with the correction content indicated by the correction instruction using the function of the correction confirmation output processing unit 125 (step S21). Then, when the correction is made by the process of step S21, the correction content is reflected, and the correction result is displayed on the administrative terminal 200 side. Note that the correction result is displayed in a state where the correction is reflected in the masked nursing care certification review board materials in FIG. 8. Note that the corrected part may be displayed in a manner that is different from the manner of the character string masked by the process of step S16, and may be displayed only when the correction is made. Note that when the user of the administrative terminal 200 determines that the correction content is acceptable, the "Save Correction" button shown in FIG. 8 is pressed, and the correction is reflected.

[0052] After corrections are made in the process of step S21 shown in Fig. 5, or when the process of step S20 is skipped because there are no parts to be corrected, the user of the administrative terminal 200 instructs output of the results (step S22). Specifically, in the process of step S22, the user performs an input operation to download the masked nursing care certification review board materials, thereby instructing output of the results. For example, when the "download" button shown in Fig. 8 is pressed, output of the results is instructed, and an output request is transmitted from the administrative terminal 200 to the information processing device 100.

[0053] When the information processing device 100 receives the output request, the control unit 120 uses the function of the correction confirmation output processing unit 125 to output a PDF file of the masked nursing care certification review board materials as the processing result (step S23), and terminates the processing of the information processing device. In the processing of step S23, the correction confirmation output processing unit 125 converts the masked nursing care certification review board materials from character data to image data, and provides the PDF file to the administrative terminal 200, thereby transmitting the PDF file as the processing result.

[0054] The administrative terminal 200 receives the masked PDF file of the nursing care certification examination committee materials transmitted from the information processing device 100, thereby acquiring the processing results (step S24), and the processing at the administrative terminal ends.

[0055] As a result, the administrative terminal 200 can acquire masked nursing care certification examination committee materials necessary for nursing care certification. The above is the operation of the information processing system 1. As described above, the information processing device of this embodiment makes it possible to mask only personal information while leaving the information necessary for nursing care insurance certification work intact, thereby enabling the smooth execution of nursing care insurance certification work while appropriately protecting personal information. Therefore, nursing care insurance certification work can be appropriately supported.

[0056] (Variation) It should be noted that the present invention is not limited to the above-described embodiment, and various modifications and applications are possible. For example, the information processing device 100 according to the above-described embodiment does not need to have all of the technical features described above, but may have some of the configurations described in the above-described embodiment so as to solve at least one problem in the prior art. Furthermore, at least a portion of each of the following modifications may be combined.

[0057] In the above embodiment, when masking is performed in the process of step S16 shown in FIG. 5, an example has been shown in which the masked character string is displayed in a conspicuous manner so that the user can easily visually recognize that it has been masked. However, this is merely an example. For example, the display manner of the masked character string may be varied depending on the reliability of the classification by AI, i.e., the reliability of the classification determination in the process of step S15 shown in FIG. 5. Specifically, as shown in FIG. 9, the display manner may be set in advance according to the reliability, such as displaying the masked character string in manner 1 when the reliability of the classification determination is 70% or higher, and displaying the masked character string in manner 2 when the reliability is less than 70%. Note that the set value of the reliability may be arbitrarily changed by the user.

[0058] In this case, for example, when masking the nursing care certification review board materials shown in Figure 4, suppose that in the processing of step S15 of Figure 5, "Neosty Minamiyamada" and "Home Best Life Yamada" are determined to belong to the category of "facility name" with a reliability of 70% or more, but "Minamiyamada Hospital" is determined to belong to the category of "medical institution name" with a reliability of less than 70%. Then, by executing the processing of step S16 of Figure 5 based on the settings of Figure 9, as shown in Figure 10, for the portions corresponding to "Neosty Minamiyamada" and "Home Best Life Yamada", the "facility name" is displayed as the masked character string in the display mode of mode 1, while for the portion corresponding to "Minamiyamada Hospital", the "medical institution name" is displayed as the masked character string in the display mode of mode 2. In this example, although the reliability of "Minamiyamada Hospital" is less than 70%, the masking is ultimately correct. However, if, for example, in the processing of step S15 in FIG. 5, it is determined that the character belongs to the category of "person's name" rather than "medical institution name," the portion corresponding to "Minamiyamada Hospital" will be displayed as a masked character string of "person's name" in the display mode of mode 2. In such a case, an instruction to correct it to "medical institution name" is issued in the processing of step S20 in FIG. 5. Note that such correction history can be used as supervised data for the learning model 113, and re-learning can be performed.

[0059] Furthermore, the display modes according to the reliability of the AI ​​are not limited to two modes, and may be multiple modes, such as three or more modes, including mode 3 where the reliability is less than 50% in addition to the above. In such cases, the masked character strings corresponding to classifications with lower reliability may be displayed in a mode that is easier for the user to see. This makes it easier for the user to recognize errors in classification decisions made by the AI, thereby improving work efficiency. Similarly, when there are two display modes, the display mode with lower reliability may be made more noticeable than the display mode with higher reliability.

[0060] In the above embodiment, a word determined to be masked is masked by changing it to a "masked character string" set in the masking setting information 112. However, this is merely an example. In addition, if a specific character string is included, the word may be changed to a pre-set character string. Specifically, the specific character string setting information shown in FIG. 11 is set, and if a word determined to be masked includes the specific character string shown in the figure, the masked character string may be a special character string. For example, as shown in FIG. 11, if the character string "Neosty Minamiyamada" is included, it would normally be masked to "facility name" based on the setting of the masking setting information 112 shown in FIG. 3. However, by setting the specific character string setting information as shown in FIG. 11, it can be masked to "nearby facility" (see FIG. 12). This allows the character string to be converted to a special character string only if it includes a specific character string, thereby improving convenience.

[0061] In the above embodiment, as shown in FIG. 8, an example was shown in which the nursing care certification review board materials before masking and the nursing care certification review board materials after masking are displayed side by side. However, for example, if there is no masking target, the image data before conversion to text data may be displayed as is as the nursing care certification review board materials after masking. This prevents garbled characters due to conversion to text data. Furthermore, if the nursing care certification review board materials are composed of multiple pages, if there is no masking target on that page, the page transition button that transitions to the content of the next page may be pressed to transition to the next page. This reduces the workload of checking and correcting work.

[0062] In the above embodiment, an example is shown in which the pre-masked nursing care certification examination board materials and the masked nursing care certification examination board materials are displayed and corrected in parallel, but this is just one example. In addition, for example, as shown in FIG. 12, when correcting a masked character string, correction candidates may be displayed by selecting the masked character string to be corrected. The user simply selects the desired correction content from the correction candidates. This makes it possible to easily perform the correction work.

[0063] The information processing device 100 according to the above embodiment can be realized using a normal computer, not a dedicated device. For example, the information processing device 100 that executes the above processes may be configured by installing a program for executing any of the above processes on a computer from a recording medium storing the program. Also, one information processing device 100 may be configured by multiple computers operating in cooperation with each other.

[0064] Furthermore, when the above-mentioned functions are realized by sharing the functions between an OS (Operating System) and an application, or by cooperation between the OS and the application, only the parts other than the OS may be stored on the medium.

[0065] It is also possible to superimpose the program on a carrier wave and distribute it via a communication network. For example, the program may be posted on a bulletin board system (BBS) on the communication network and distributed via the network. These programs may then be started and run under the control of an operating system in the same way as other application programs, thereby enabling the above-mentioned processing to be performed. [Explanation of symbols]

[0066] 1 Information processing system 100 Information processing device, 110 Memory unit, 111 Program, 112 Masking setting information, 113 Learning model, 120 Control unit, 121 Nursing care certification examination committee material acquisition and conversion unit, 122 Classification determination unit, 123 Masking target determination unit, 124 Masking processing unit, 125 Correction confirmation output processing unit, 130 Input / output unit, 140 Communication unit, 210 Network, 200 Administrative terminal

Claims

1. a means for acquiring, from an administrative terminal, electronic data of materials for the nursing care certification examination committee, which are examination documents for the nursing care insurance certification work; a classification determination means for classifying named entities contained in sentences described in the nursing care certification examination committee materials acquired by the nursing care certification examination committee material acquisition means; a masking target determination means for determining whether the classification of the named entity determined by the classification determination means is a classification to be masked based on pre-stored masking setting information; a masking processing means for masking the named entity in accordance with preset content based on the determination result by the masking target determination means; an output means for providing the nursing care certification examination committee materials after the masking processing means has performed the masking to the administrative terminal; An information processing device comprising:

2. the masking processing means performs masking by converting the named entity into content specified by the masking setting information; 2. The information processing apparatus according to claim 1, wherein:

3. the masking processing means sets the display mode of the named entity after the masking to a specific mode that is easy for a user to visually recognize.

3. The information processing apparatus according to claim 1, wherein the information processing apparatus is a computer.

4. the classification determination means performs a morphological analysis of the sentences described in the nursing care certification examination committee materials and classifies the named entities using a learning model that has been machine-learned in advance by supervised learning; the masking processing means changes the display mode of the named entity after the masking process depending on the reliability of the classification of the named entity using the learning model in the classification determination means.

3. The information processing apparatus according to claim 1, wherein the information processing apparatus is a computer.

5. When the named entity includes a predetermined specific character string, the masking processing means performs masking by converting the named entity into a special character string different from the content specified by the masking setting information.

2. The information processing apparatus according to claim 1, wherein:

6. An information processing method by an information processing device, a step of acquiring nursing care certification examination committee materials from an administrative terminal, the electronic data of which is a document for examination in nursing care insurance certification work; a classification determination step of classifying named entities included in the sentences described in the nursing care certification examination committee materials acquired in the nursing care certification examination committee materials acquisition step; a masking target determination step of determining whether the classification of the named entity determined in the classification determination step is a classification to be masked based on pre-stored masking setting information; a masking process step of masking the named entity in accordance with preset content based on the determination result in the masking target determination step; an output step of providing the nursing care certification examination committee materials after the masking process in the masking processing step to the administrative terminal; An information processing method comprising:

7. Computer, A means for acquiring nursing care certification examination committee materials from an administrative terminal, the electronic data of nursing care certification examination committee materials being examination documents in nursing care insurance certification work; a classification determination means for classifying named entities contained in sentences described in the nursing care certification examination committee materials acquired by the nursing care certification examination committee material acquisition means; a masking target determination means for determining whether the classification of the named entity determined by the classification determination means is a classification to be masked based on pre-stored masking setting information; a masking processing means for masking the named entity in accordance with preset content based on the determination result by the masking target determination means; an output means for providing the nursing care certification examination committee materials after the masking process by the masking processing means to the administrative terminal; A program that functions as a

Images