Work action evaluation apparatus, work action evaluation system, work action evaluation method, and program

Abstract

To broadly utilize useful work actions of workers rather than traditional work improvement activities based on expert know-how, thereby raising awareness based on wider range of knowledge.SOLUTION: A work action evaluation apparatus includes: an information acquisition unit which acquires information on at least a worker and a response of action of the worker with respect to a work object; a packaging processing unit which divides the information acquired by the information acquisition unit into work-relevant units using features of the information to package them; an evaluation processing unit which evaluates elements included in the packaged information; a topic creation processing unit which creates topic including the element included in the packaged information and a result of evaluating the element; and a display retrieval unit which indicates contradictory elements in the topic in line with a criterion indicating contradiction defined for a series of element evaluation results in the created topic.SELECTED DRAWING: Figure 9

Description

[Technical Field]

[0001] The present invention relates to a work behavior evaluation device, a work behavior evaluation system, a work behavior evaluation method, and a program. [Background technology]

[0002] Traditionally, attempts have been made to formalize tacit knowledge, which is subjective knowledge that has not been put into words, such as know-how based on personal experience and intuition, or skills acquired through repeated work. Here, knowledge can be read as know-how.

[0003] Patent document 1 discloses a technology that extracts knowledge information of experts from work images and work audio data of multiple workers (at least one of whom is an expert) engaged in various tasks, making it manageable and editable, and then appropriately manages the information. Summary of the Invention [Problem to be solved by the invention]

[0004] Meanwhile, workers (not necessarily skilled workers) engaged in various tasks are exposed to a variety of information obtained in their daily work activities from the reactions of work targets (including people) and conversations between workers, etc. It is believed that a great deal of information useful for work is hidden within this variety of information.

[0005] However, with conventional technology, it has been difficult to obtain information useful for work from the various pieces of information obtained in daily work activities, such as reactions from work objects (including people) and conversations between workers.

[0006] The present invention has been made in light of the above, and aims to provide "awareness" based on a wide range of knowledge by making it possible to widely utilize the useful work behaviors of each worker, rather than relying on work improvement activities based on the know-how of experts up to now. [Means for solving the problem]

[0007] In order to solve the above-mentioned problems and achieve the object, the present invention is characterized by comprising an information acquisition unit that acquires information related to at least the worker and the worker's behavioral response to the work object and the work environment; a packaging processing unit that divides the information acquired by the information acquisition unit into units that are meaningful for business purposes using the characteristics of the information and packages them; an evaluation processing unit that evaluates elements included in the information packaged by the packaging processing unit; a topic creation processing unit that creates topics that include the elements included in the information packaged by the packaging processing unit and the evaluation results of the elements; and a display and search unit that shows contradictory elements in the topics according to criteria that indicate contradictions defined for a series of element evaluation results in the topics created by the topic creation processing unit. [Effects of the Invention]

[0008] According to the present invention, it is possible to make wider use of each worker's useful work behavior, rather than relying on the know-how of experts in the past, thereby achieving the effect of providing "awareness" based on a wide range of knowledge. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a block diagram showing the configuration of a work behavior evaluation system according to the first embodiment. [Figure 2] FIG. 2 is a diagram showing an outline of the work behavior evaluation system. [Figure 3] FIG. 3 is a block diagram showing the hardware configuration of the information processing device and the client terminal. [Figure 4] FIG. 4 is a functional block diagram showing functions provided in the information processing device. [Figure 5] FIG. 5 is a flowchart schematically showing the flow of work behavior evaluation processing in an information processing device. [Figure 6]FIG. 6 is a diagram illustrating information packaging. [Figure 7] FIG. 7 is a diagram showing an example of an information package. [Figure 8] FIG. 8 is a diagram showing an example of a list of topics. [Figure 9] FIG. 9 is a diagram showing an example of a business record. [Figure 10] FIG. 10 is a diagram showing an example of a method for creating an information package. [Figure 11] FIG. 11 is a diagram showing an example of a method for creating topics. [Figure 12] FIG. 12 is a diagram showing an example of a description item extraction method. [Figure 13] FIG. 13 is a diagram showing an example of a method for finishing a business record. [Figure 14] FIG. 14 is a flowchart schematically illustrating an example of the flow of a work behavior evaluation process in an information processing device. [Figure 15] FIG. 15 is a diagram showing an example of a method for providing information that provides awareness. [Figure 16] FIG. 16 is a diagram showing another example of a method for providing information that provides awareness. [Figure 17] FIG. 17 is a diagram showing yet another example of a method for providing information that provides awareness. [Figure 18] FIG. 18 is a diagram showing a first modification of the topic creation method. [Figure 19] FIG. 19 is a diagram showing a second variation of the topic creation method. [Figure 20] FIG. 20 is a diagram showing a third variation of the topic creation method. [Figure 21] FIG. 21 is a flowchart schematically illustrating the flow of a work behavior evaluation process in an information processing device according to the second embodiment. [Figure 22] FIG. 22 is a diagram showing an example of a method for providing information that provides awareness. [Figure 23]FIG. 23 is a diagram showing another example of a method for providing information that provides awareness. [Figure 24] FIG. 24 is a flowchart schematically illustrating the flow of a work behavior evaluation process in an information processing device according to the third embodiment. [Figure 25] FIG. 25 is a diagram showing an example of a method for providing information that provides awareness. [Figure 26] FIG. 26 is a diagram showing another example of a method for providing information that provides awareness. DETAILED DESCRIPTION OF THE INVENTION

[0010] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of a work behavior evaluation device, a work behavior evaluation system, a work behavior evaluation method, and a program will be described in detail with reference to the accompanying drawings.

[0011] (First embodiment) FIG. 1 is a block diagram showing the configuration of a work behavior evaluation system 100 according to a first embodiment. As shown in FIG. 1, the work behavior evaluation system 100 includes an information processing device 50 that functions as a work behavior evaluation device, a client terminal 10 used by a user, and various wearable devices 30 worn by the worker. The client terminal 10 is connected to the information processing device 50 via a communication network N such as the Internet. Note that FIG. 1 shows only one information processing device 50 and one client terminal 10, but this is not limiting, and there may be a plurality of information processing devices 50, client terminals 10, and wearable devices 30.

[0012] The information processing device 50 is a so-called server, etc. The information processing device 50 stores various databases (DBs) that store various data, which will be described later. Note that the information processing device 50 may store the various DBs that store the various data in an external storage server that can communicate with the information processing device 50.

[0013] Work behavior evaluation software 50A, which is a web application, is installed on the information processing device 50. Here, the work behavior evaluation software 50A may take a form other than a web application. The work behavior evaluation software 50A is a program for evaluating and notifying work based on behavior records and past model data. The behavior records are generated by extracting characteristic parts (topics) of work elements from audio data, image data, and management data related to the work object and the worker.

[0014] The information processing device 50 may generate various data to be stored in various DBs through the learning effects of machine learning. Here, machine learning refers to a technology that allows a computer to acquire human-like learning capabilities, in which the computer autonomously generates algorithms necessary for judgments such as data identification from previously acquired learning data and applies these algorithms to new data to make predictions. The learning method for machine learning may be any of supervised learning, unsupervised learning, semi-supervised learning, reinforcement learning, and deep learning, or may be a combination of these learning methods. Any learning method for machine learning is acceptable.

[0015] The client terminal 10 is an information processing terminal such as a PC (personal computer) or a smartphone. The client terminal 10 is a device that transmits and receives various data to and from an information processing device 50. A browser 10A for running work behavior evaluation software 50A, which is a web application managed by the information processing device 50, is pre-installed on the client terminal 10.

[0016] For example, the worker inputs various data (audio data, image data, management data, etc.) to work behavior evaluation software 50A managed by information processing device 50 via various wearable devices 30 (see FIG. 2).

[0017] FIG. 2 is a diagram showing an overview of the work behavior evaluation system 100. FIG. 2 shows an example of caregiving. As shown in FIG. 2, various types of data such as image data and audio data are continuously acquired over time by various wearable devices 30 worn by the caregiver, who is the worker. The various wearable devices 30 transmit the various types of data continuously acquired over time to an information processing device 50.

[0018] The information processing device 50 automatically creates daily nursing care records using machine learning based on AI (Artificial Intelligence). In doing so, records based on the concept of "SOAP" are taken into consideration. "SOAP" stands for "Subject (subjective information)," "Object (objective information)," "Assessment," and "Plan."

[0019] In addition, the information processing device 50 automatically records in detail the daily life at the care site (conversations with the person requiring care, the situation at the care site, conversations between caregivers, etc.), and classifies and stores the "sources of awareness" (more detailed S information, O information) that are the triggers for "awareness" obtained by recognizing contradictions buried in mundane daily work.

[0020] By viewing the "sources of awareness" stored in the information processing device 50 and reflecting on their work, users of the client terminal 10 can achieve higher levels of results than those achieved through work improvement activities based on the know-how of previous experts.

[0021] Next, the hardware configuration of the information processing device 50 and the client terminal 10 will be described.

[0022] 3 is a block diagram showing the hardware configuration of the information processing device 50 and the client terminal 10. Here, the hardware configuration of the information processing device 50 will be described.

[0023] The information processing device 50 is constructed by a computer, and as shown in FIG. 3, includes a CPU 501, a ROM 502, a RAM 503, an HD 504, an HDD (Hard Disk Drive) controller 505, a display 506 which is a display unit, an external device connection I / F (Interface) 508, a network I / F 509, a data bus 510, a keyboard 511, a pointing device 512, a DVD-RW (Digital Versatile Disk Rewritable) drive 514, and a media I / F 516.

[0024] Of these, the CPU 501 controls the overall operation of the information processing device 50. The ROM 502 stores programs used to drive the CPU 501, such as an IPL. The RAM 503 is used as a work area for the CPU 501. The HD 504 stores various data such as programs. The HD 504 of the information processing device 50 stores work behavior assessment software 50A, which is a web application, as a program. The HD 504 of the client terminal 10 stores a browser 10A (program) that executes the work behavior assessment software 50A, which is a web application. The HDD controller 505 controls the reading and writing of various data from and to the HD 504 under the control of the CPU 501. Necessary processing may be performed using a GPU (not shown) used for drawing on the display 506.

[0025] The display 506 displays various types of information such as a cursor, menus, windows, characters, or images. The external device connection I / F 508 is an interface for connecting various types of external devices. In this case, the external devices are, for example, USB (Universal Serial Bus) memories, printers, etc. The network I / F 509 is an interface for data communication using the communication network N. The data bus 510 is an address bus, data bus, etc. for electrically connecting the components such as the CPU 501 shown in FIG. 3.

[0026] The keyboard 511 is a type of input means having multiple keys for inputting characters, numbers, various instructions, etc. The pointing device 512 is a type of input means for selecting and executing various instructions, selecting a processing target, moving a cursor, etc. The DVD-RW drive 514 controls reading and writing of various data from a DVD-RW 513, which is an example of a removable recording medium. Note that this is not limited to a DVD-RW, and may be a DVD-R, etc. The media I / F 516 controls reading and writing (storing) of data from a recording medium 515, such as a flash memory.

[0027] Next, functions realized by the CPU 501 of the information processing device 50 operating based on the work behavior evaluation software 50A stored in the HD 504 of the information processing device 50 will be described.

[0028] Fig. 4 is a functional block diagram showing functions of the information processing device 50. As shown in Fig. 4, the CPU 501 of the information processing device 50 operates based on the work behavior evaluation software 50A, and the information processing device 50 includes an information acquisition unit 51, a first processing unit 52, a second processing unit 53, a model creation unit 54, a display search unit 55, a basic data DB 61, a model DB 62, a learning program DB 63, a learning data DB 64, and a document DB 65.

[0029] The information acquisition unit 51 acquires information such as voice data, image data, and time data relating to at least the worker and the worker's behavioral responses to the work recipient (including a person) and the work environment from various wearable devices 30 worn by the worker. Here, it is assumed that the worker is a caregiver and the work recipient is a person requiring care.

[0030] It is desirable that the information acquired by the information acquisition unit 51 be continuous in time (continuously acquiring data). By not imposing any restrictions on data acquisition in this way, it becomes possible to collect and analyze information without omission, eliminating oversights and enabling the extraction of higher quality "awareness." In addition, the configuration is simple, which increases reliability.

[0031] The first processing unit 52 classifies the daily activity records of each worker collected from the wearable device 30 or the like into topics (characteristic parts of work elements).

[0032] The basic data DB 61 manages the evaluation results of the elements included in the information packaged by the first processing unit 52, as will be described in detail later.

[0033] The second processing unit 53, the details of which will be described later, creates a business record that is a care record created from partial information of the topics and to which a link to the original topic is attached.

[0034] The document DB 65 manages the topics classified by the first processing unit 52 and the business records created by the second processing unit 53.

[0035] The model DB 62 manages a trained model for segmentation 621, a trained model for behavior estimation 622, a trained model for behavior evaluation 623, a trained model for record extraction 624, and a trained model for record document creation 625. The trained model for segmentation 621, the trained model for behavior estimation 622, and the trained model for behavior evaluation 623 are used to create and classify topics in the first processing unit 52. The trained model for record extraction 624 and the trained model for record document creation 625 are used to create business records in the second processing unit 53.

[0036] The learning program DB 63 is used to create each model managed in the model DB 62, and manages a division program 631, a behavior estimation program 632, a behavior evaluation program 633, a record extraction program 634, and a record document creation program 635.

[0037] The learning data DB 64 is used to create each model managed in the model DB 62, and manages data for division 641, data for behavior estimation 642, data for behavior evaluation 643, data for record extraction 644, and data for record document creation 645.

[0038] The model creating unit 54 creates the models 621 to 625 managed in the model DB 62 based on the programs 631 to 635 managed in the learning program DB 63 and the data 641 to 645 managed in the learning data DB 64.

[0039] The display search unit 55 provides a search unit 551 and a display unit 552 to the accessed client terminal 10.

[0040] Next, the first processing unit 52 will be described in detail.

[0041] As shown in Figure 4, the first processing unit 52 includes a temporary information storage unit 521, a first information processing unit 522, an information division processing unit 523, an information packaging processing unit 524, a behavior evaluation processing unit 525 which is an evaluation processing unit, and a topic creation processing unit 526.

[0042] The temporary information storage unit 521 temporarily stores in RAM 503 information acquired by the information acquisition unit 51, such as at least the worker and the worker's behavioral responses to the object being worked on (including a person), as well as voice data, image data, time data, and other information related to the work environment.

[0043] The first information processing unit 522 processes the acquired information temporarily stored by the temporary information storage unit 521 into a form that can be analyzed by the information division processing unit 523. Specifically, the processing by the temporary information storage unit 521 involves, for example, transcribing audio data and converting it into character data. Thereafter, as will be described later, the information division processing unit 523 executes division processing.

[0044] The information division processing unit 523 reads the trained model for division 621 and divides the information processed by the first information processing unit 522 into units that are meaningful in terms of business, using the feature quantities of the information.

[0045] The information packaging processing unit 524 is a packaging processing unit that reads the trained model for behavior estimation 622 and packages the information divided by the information division processing unit 523. Packaging will be described later.

[0046] The behavior evaluation processing unit 525 reads the evaluation results of the elements contained in the packaged information managed by the behavior evaluation trained model 623 and the basic data DB 61, evaluates the elements contained in the information packaged by the information packaging processing unit 524, and updates the basic data DB 61 with the evaluation results of the elements contained in the packaged information.

[0047] The topic creation processing unit 526 creates topics that include elements included in the information packaged by the information packaging processing unit 524 and the evaluation results from the behavior evaluation processing unit 525. The topic creation processing unit 526 also stores the created topics in the document DB 65.

[0048] The search unit 551 of the display search unit 55 searches for contradictory elements within the topics according to criteria indicating contradictions defined for a series of element evaluation results within the topics created by the topic creation processing unit 526 and stored in the document DB 65.

[0049] The display section 552 of the display search section 55 displays the contradictory elements in the topics searched by the search section 551.

[0050] Here, the topic has a criterion for indicating inconsistency defined for a series of element evaluation results within the topic, and the display search unit 55 can display inconsistent elements within the topic according to the criterion for indicating inconsistency.

[0051] A possible method for analyzing contradictory elements within a topic according to the criteria for indicating contradictions within a topic is to evaluate the difference from the standard response of the work recipient and specify the parts with the greatest difference.

[0052] The display search unit 55 also clearly indicates the location of the inconsistency, making it easier to identify the cause of the inconsistency. Furthermore, it is possible to reliably identify inconsistencies from a large amount of information, allowing for thorough review.

[0053] The first processing unit 52 updates the various types of learning data in the learning data DB 64 with the various types of learning data used in the above-mentioned units.

[0054] Next, the second processing unit 53 will be described in detail.

[0055] As shown in FIG. 4, the second processing unit 53 includes a second information processing unit 531, a behavior extraction processing unit for recording 532, and a document creation processing unit 533.

[0056] The second information processing unit 531 reads the topics stored in the document DB 65 .

[0057] The record behavior extraction processing unit 532 reads the record extraction trained model 624 and extracts the record behavior.

[0058] The document creation processing unit 533 reads the trained model for creating record documents 625 and creates business records that are created from partial information on topics and have links to the original topics. In other words, the business records are linked to valid related data outside the business records, making it easier to improve business operations. The document creation processing unit 533 also saves the created business records in the document DB 65.

[0059] The second processing unit 53 updates the various types of learning data in the learning data DB 64 with the various types of learning data used in the above-mentioned units.

[0060] Next, the flow of the work behavior evaluation process in the information processing device 50 will be described in detail.

[0061] 5 is a flowchart outlining the flow of work behavior evaluation processing in the information processing device 50. As shown in FIG. 5, first, the information acquisition unit 51 acquires data relating to at least the worker and the worker's behavioral responses to the work recipient (including a person) and the work environment from various wearable devices 30 worn by the worker (step S1). In this example, the worker is a caregiver and the work recipient is a person requiring care.

[0062] Data acquired from the wearable device 30 includes voice data, image data, time data, temperature data, humidity data, weather data, atmospheric pressure data, illuminance data, wind speed data, vital data, and the like.

[0063] Next, the first processing unit 52 (such as the information packaging processing unit 524) creates an information package (step S2).

[0064] Here, we will explain the definition of an information package. An information package is created from divided information, and as shown in Figure 7, it contains various information related to business operations as well as elements (behavioral information). This will be explained in detail below.

[0065] FIG. 6 is a diagram showing information packaging. As shown in FIG. 6, various wearable devices 30 acquire various types of data (audio data, image data, etc.). The acquired data from the wearable devices 30 is input continuously over time. Information packaging means "extracting meaning from data acquired continuously over time and organizing the data into meaningful units." In other words, the organized data group is an information package. The information written in the information package needs to be interpreted as human behavior information, environmental conditions, etc.

[0066] Here, Fig. 7 shows an example of an information package, where Fig. 7(a) shows a general example of an information package, and Fig. 7(b) shows a care example as a specific example of an information package.

[0067] The information package shown in Figure 7 is a collection of human behavior information that has been saved. A portion of the behavior information (the shaded area in Figure 7(b)) is extracted from the information package and used for business records.

[0068] As shown in FIG. 7(a), a portion of raw data (such as image information) may be left in the information package as auxiliary data, for example, by being linked to a link L1.

[0069] 5, the first processing unit 52 (topic creation processing unit 526, etc.) organizes the information packages into topics (step S3). At this time, the first processing unit 52 (behavior evaluation processing unit 525, etc.) evaluates the elements in the information packages and includes the results in the topics.

[0070] FIG. 8 is a diagram showing an example of a list of topics.

[0071] As shown in Figure 8, the daily behavioral information collected from the wearable device 30 for each worker is classified as topics. That is, a topic is an evaluation of each element (behavioral information) of an information package. The topic is linked to the corresponding part of the business record as a link L3.

[0072] Note that behavioral information is not limited to behaviors performed by humans. For example, behavioral information can also include the door opening, a bird singing, the rain leaking, the bed being at a shallow angle, and the room being noisy. In particular, behavioral information can be referred to as environmental information when describing environmental conditions.

[0073] Then, the first processing unit 52 stores the list of topics in the document DB 65 (step S4).

[0074] Thereafter, in response to an analysis / output request from the client terminal 10, i.e., a request to find and present "sources of awareness" from the topics, the display search unit 55 displays related topics to the accessed client terminal 10 (step S5).

[0075] On the other hand, the second processing unit 53 (second information processing unit 531, recording behavior extraction processing unit 532) extracts recording behavior, which is information used for business recording (step S6).

[0076] Next, the second processing unit 53 (the second information processing unit 531, the document creation processing unit 533) creates a business record (step S7).

[0077] Here, Figure 9 is a diagram showing an example of a business record. Figure 9(a) shows a general example of a business record, and Figure 9(b) shows a nursing care example as a specific example of a business record. As shown in Figure 9, the business record records the actions to be recorded. In addition, as shown in Figure 9, the business record has a link L2 attached to the topic used to create the business record. This makes it possible to jump to the topic used to create the business record when looking back on the business record and finding areas that need improvement or when wanting to think about why something went well.

[0078] Next, the method for creating the information package by the first processing unit 52 in step S2 described above will be described in detail.

[0079] FIG. 10 illustrates an example of a method for creating an information package. The items to be classified are models for estimating tasks at a given time from information (e.g., image data, audio data) acquired continuously over time, such as conversations during various tasks and conversations between various tasks. The first processing unit 52 (the information temporary storage unit 521, the first information processing unit 522, and the information segmentation processing unit 523) collects information such as images and audio (e.g., conversation records) while a certain task is being performed, and correlates and trains the information with task names to create the trained model for segmentation 621. The trained model for segmentation 621 can be created using general text classification and image classification techniques. Using this trained model for segmentation 621, the task at that time can be estimated from newly input information. The trained model for segmentation 621 can classify and package not only general tasks but also non-task information (e.g., conversations between workers).

[0080] Note that the time at which the information acquired continuously over time was acquired can be determined by synchronizing the clocks on the various wearable devices 30. The information acquired continuously over time can include image data, audio data, time data, vital signs data, location information data, temperature / humidity data, air pressure data, illuminance data, wind speed data, and the like.

[0081] For estimation, information acquired continuously over time is divided into appropriate time intervals, and task estimation is performed for each division. For example, as shown in FIG. 10(a), the first processing unit 52 (information temporary storage unit 521, first information processing unit 522, information division processing unit 523) performs task prediction by assigning an appropriate time (e.g., the center time of the division) to the time-divided information, and estimates that the task with the highest probability will be performed. If it is determined that the next task has been performed at the time when the task probability has changed (the switching time is estimated using an appropriate moving average, etc.), the task and its start and end times can be estimated. For task estimation, a wide range of general classification methods can be used, such as ensemble learning, which uses multiple estimation methods to perform task prediction and estimates based on majority vote.

[0082] For example, if a worker is wearing a dedicated wearable device 30, the worker can be identified from the ID of the wearable device 30, and the person they are talking to (the person receiving care in the care example) can be identified by extracting names that come up in conversation using methods such as named entity extraction, or by performing pattern matching of facial images obtained from images, and their location can be identified from the location information (at which time the environment of that place at that time can also be identified from image data, temperature / humidity data, etc.). In this way, information organized and divided by task is created.

[0083] The information divided as shown in FIG. 10(a) is in a state where various data within the time period divided by business unit is bundled together without being processed.

[0084] Next, the first processing unit 52 (information packaging processing unit 524) inputs the information segmented as described above into the trained model for activity estimation 622, and estimates what was happening in the segmented information. Image data and voice data are primarily used for activity estimation. The first processing unit 52 (behavior evaluation processing unit 525) performs speaker separation on the voice data using an appropriate method, transcribes the voice, estimates who said what, and then, adding the image data, estimates the activity using methods such as action recognition, object detection, VQA, and sentence classification. Note that the conversational text used in the analysis is left as is in the information package after the speaker is assigned.

[0085] Next, a method for creating topics by the first processing unit 52 in step S3 described above will be described in detail.

[0086] Figure 11 shows an example of how to create topics. Information packages are divided as necessary and organized, for example, as shown in Figure 11. In the example shown in Figure 11, the following are organized as a list of topics: the person in charge (person in charge classification) indicating the worker who is the main actor, the person and location the person in charge is working with (Classification 1), the type of original record (Classification 2), the task name (including information exchanges between workers in addition to normal tasks / Classification 3), the name of the person involved (Classification 4), time information, topics (behavioral information + evaluation of the behavioral information), and, if present, a link L3 to the task record created using that topic.

[0087] The behavioral information includes a plurality of behaviors. Evaluation of the behavioral information (behavior evaluation) may be performed for each of the plurality of behaviors included in the behavioral information, or a group of several behaviors may be evaluated.

[0088] Behavioral evaluation automatically judges behavior and assigns positive, negative, unexpected, or favorable results. For example, a positive result can be applied when the result itself is positive or when the actor has positive emotions. Positive / negative determinations can be made using common negative / positive determination methods used in natural language processing. Unexpected results are obtained when a series of actions in a task deviates significantly from the normal behavioral routine. For example, unexpected results can be determined by classifying each action using commonly used text classification and comparing the flow of that classification with the normal flow.

[0089] Next, a method for providing information that gives notice by the display search unit 55 in step S5 described above will be described in detail.

[0090] Traditionally, work reports include the name of the work, the results and observations of the work, etc. However, because these records are created based on the focus of the work manager, it is difficult to capture casual "realizations."

[0091] However, the topics shown in Figure 11 contain not only the behavioral information used in the business report, but also a wide range of behavioral information related to the business that was not used in the business report, with evaluation attached. This information is used to extract "awareness." Information related to "awareness" can be obtained by identifying deviations from the norm and inconsistencies in communication based on behavioral evaluation. If necessary, similar situations can be displayed, and by comparing and referencing, "awareness" can be expanded and led to good ideas.

[0092] Next, a detailed description will be given of the method for creating a business record by the second processing unit 53 in the above-mentioned steps S6 and S7. The creation of a business record is performed in the following two steps (a step for extracting items to be described, and a step for finishing the business record).

[0093] First, the description item extraction step will be described. Here, FIG. 12 is a diagram showing an example of a description item extraction method. FIG. 12(a) shows a general example of creating a business record, and FIG. 12(b) shows a nursing care example as a specific example of creating a business record. As shown in FIG. 12, the second processing unit 53 (second information processing unit 531) selects elements (behavioral information) from the topics. Then, the second processing unit 53 (recording behavior extraction processing unit 532) determines items (recording behavior) to be recorded in the business record from the selected elements (behavioral information). This is performed using key phrase estimation from the information contained in the information package, or (if human emotions are important) emotion estimation using image data, audio data (contained in image data), and text information.

[0094] Taking the nursing care record shown in Figure 12(b) as an example, if the record is about lunch assistance, key phrases related to lunch (such as the menu and whether or not the person ate) are extracted. In addition, statements that are considered important while eating (such as "I don't need it" indicating refusal) are also extracted, along with emotion estimation.

[0095] Next, the steps for finishing business records will be explained. Here, Fig. 13 is a diagram showing an example of a method for finishing business records. Fig. 13 shows an example of nursing care as a concrete example of finishing business records. As shown in Fig. 12(b), the information extracted from the behavioral information is not written in a form suitable for a business record.

[0096] Therefore, the second processing unit 53 (document creation processing unit 533) creates more natural sentences using a large-scale language model or the like to create the final business record, as shown in Fig. 13. Also, as shown in Fig. 13, the second processing unit 53 attaches a link L4 to the topic analysis so that the topics that are the source of the business record can be reached, so that subsequent information collection can be carried out quickly.

[0097] When reviewing work, the person in charge of creating business records can click on the link to topic analysis (L4), which will display the results of the topic analysis, providing an opportunity for gaining new insights.

[0098] FIG. 14 is a flowchart showing an outline of an embodiment of the flow of work behavior evaluation processing in the information processing device 50. In FIG.

[0099] As shown in FIG. 14, the basic data DB 61 manages the response status to questions posed by caregivers.

[0100] When compiling the information packages into topics, the first processing unit 52 (topic creation processing unit 526, etc.) refers to the basic data DB 61, and if there is a reaction that is different from the normal reaction (especially one obtained from voice or image data), it is recorded in the topic (step S3). For example, the voice of the reply to the caregiver's question is louder than the normal reaction.

[0101] When displaying related topics, the display search unit 55 compares the reaction to the prompt with normal times to determine the communication level (step S5). The "determination" here refers to finding differences from normal times and discrepancies between positive and negative. Criteria for determining differences from normal reactions include the reaction situation (the ratio of nodding and verbal responses) and the strength of the voice when verbal responses are made (tone of voice).

[0102] When presenting topics, the display search unit 55 displays the communication level and indicates the difference from normal. For example, the display search unit 55 displays examples of normal communication levels in similar situations (persons involved, business classifications, etc.) on the same screen.

[0103] As shown in FIG. 8, the daily behavioral information collected for each worker from the wearable device 30 or the like is classified as topics. In other words, topics are evaluations of each element (behavioral information) of an information package. Topics are linked to the corresponding part of the business record as a link L3. This business record link will take you to the topics in FIG. 15, which will be described later.

[0104] Next, a method for providing information that gives notice by the display search unit 55 in step S5 described above will be described in detail.

[0105] Here, Fig. 15 is a diagram showing an example of a method for providing information that provides awareness. Note that the bold parts shown in Fig. 15 are parts written in the business record. As a premise of the example shown in Fig. 15, it is assumed that in step S4, the first processing unit 52 has performed the following process.

[0106] The first processing unit 52 evaluates the state of Mr. C's response to the question posed by caregiver A in the behavioral information. Whether the response was made by conversation or by movement is determined in advance by behavior estimation.

[0107] More specifically, the basic data DB 61 stores data such as the response rate of Mr. Ctani to questions. The first processing unit 52 compares the data stored in the basic data DB 61 with the average value to determine the communication level. The first processing unit 52 considers a set of behavioral information from a question asked by the caregiver to just before the next question, and evaluates whether or not / how Mr. Ctani responded during that time. For example, here, for "7" questions, the number of verbal responses is "4," the number of physical responses is "2," and the number of no response is "1." According to the basic database information in the basic data DB 61, the normal number of verbal responses is 90%, but this time it was 57% (4 / 7), which means that there were fewer verbal responses than usual.

[0108] In the example shown in Fig. 15, when presenting topics, the display search unit 55 sets the communication level to "unusual" and the behavioral evaluation to "few verbal responses." The standard for inconsistency may be that a certain level of difference from the normal response to the caregiver's question is recognized. The value of the difference may be determined appropriately.

[0109] Then, the display search unit 55 searches for a state where the communication level is good as a normal state in the topics shown in FIG. 15, and displays the screen shown in FIG. 16, which will be described later.

[0110] Here, Figure 16 is a diagram showing another example of a method for providing information that raises awareness. The example shown in Figure 16 is an example of a normal communication level displayed on the same screen. Note that the bold part shown in Figure 16 is the part written in the business record.

[0111] In the example shown in FIG. 16, when presenting topics, the display search unit 55 sets the communication level to "normal (good)" and sets the part corresponding to the behavioral evaluation to "providing good verbal responses."

[0112] In the example shown in Figure 16, it would be good if you could realize that the act of moving forward with the salmon without listening to Mr. C's opinion was the cause of the ambiguity (you should realize this on your own). It would also be good if you could realize that Mr. C was not convinced even though he nodded.

[0113] If you feel that even this information is insufficient (or tends to be insufficient), you can search further and gain some insight into how Mr. C communicates from the interactions between caregivers shown in Figure 17, which will be described later. It is also fine if you gain some insight at the stage of the examples in Figures 15 and 16.

[0114] Here, Fig. 17 is a diagram showing yet another example of a method for providing information that gives awareness. The example shown in Fig. 17 is an example of searching for information about Mr. Ctani in a conversation between caregivers.

[0115] 17, the display search unit 55 presents information about conversations between Ms. C. Tani, for example, by pulling it from conversation topics between workers. More specifically, a search word entry section (not shown) can be provided on the screen, and a normal search method can be used to automatically search for conversation information about residents by entering "conversation," "Ms. C. Tani," "resident information," etc.

[0116] In the example shown in Figure 17, it would be good to notice that "we need to have a good conversation to make Mr. Ctani feel comfortable."

[0117] As described above, according to this embodiment, by setting appropriate evaluation criteria, the inconsistencies in the work behavior information included in the topic group are evaluated, and the inconsistencies hidden in a series of work behaviors are clarified, thereby presenting information that encourages "awareness." This allows workers to reflect on their work based on the "awareness" gained by recognizing the inconsistencies hidden in everyday work, and thereby makes it possible to widely utilize each worker's useful work behavior, rather than relying on work improvement activities based on the know-how of previous experts, thereby achieving a high level of results.

[0118] Because ordinary know-how is extracted based on the work behavior of a small number of experts, it is difficult to reach a work level higher than that of those experts. On the other hand, by evaluating widely collected information without selecting workers and using related evaluated information to improve work, the tacit knowledge of more people can be utilized, and a higher work level can be reached.

[0119] Furthermore, because work behavior information is continuously obtained from all workers, regardless of whether they are experts or novices, it is possible to create a group of topics containing useful information, including know-how based on the actions of experts, without missing any information. In other words, it is possible to utilize useful information that has not yet been refined to the level of know-how.

[0120] [Variation 1] Here, Figure 18 is a diagram showing Variation 1 of the topic creation method. Variation 1 shown in Figure 18 is a method in which, during normal business operations, appropriate behavioral evaluations are performed on the contents of information packages, and the results are combined to create categories that are easy to search, thereby creating classified topics. According to Variation 1 shown in Figure 18, the implementation examples branch off depending on the specific behavioral evaluations and how they are used.

[0121] [Variation 2] Here, Fig. 19 is a diagram showing a second variation of the topic creation method. In the second variation shown in Fig. 19, topics are created in conversations between business personnel in the same way as in normal business, but are not linked to business records.

[0122] [Variation 3] FIG. 20 is a diagram showing a third variation of the topic creation method. In the third variation shown in FIG. 20, topics are created including environmental information. For example, if there is anything distinctive in the video captured by the camera (wearable device 30) worn by the worker, it is recorded as environmental information. This allows the difference from usual to be evaluated and used as a topic as an environmental evaluation (corresponding to a behavioral evaluation). Specific examples of environmental evaluation include "The angle of the backrest of Mr. / Ms. X's bed is shallow" and "The room is clean."

[0123] (Second embodiment) Next, a second embodiment will be described.

[0124] The second embodiment differs from the first embodiment in that it uses a method for analyzing contradictory elements in topics by specifying parts where the evaluation results of two or more types of data (images and audio, etc.) do not match, or by specifying parts where the evaluation result of one type of data cannot be determined. In the following explanation of the second embodiment, explanations of parts that are the same as in the first embodiment will be omitted, and only parts that differ from the second embodiment will be explained.

[0125] FIG. 21 is a flowchart schematically showing the flow of work behavior evaluation processing in the information processing device 50 according to the second embodiment.

[0126] 21, the basic data DB 61 is used for comparison and manages the reaction status to questions posed by caregivers. The basic data DB 61 is educated by accumulating topics.

[0127] Then, when compiling the information packages into topics, the first processing unit 52 (topic creation processing unit 526, etc.) uses machine learning to determine whether each piece of content written in the topic is positive or negative, and includes the results in the topic (step S3).

[0128] When displaying related topics, the display search unit 55 judges parts with ambiguous positive / negative judgments as "sources of awareness." The ambiguous evaluation of positive / negative judgments differs depending on whether one piece of acquired data is used or multiple pieces of acquired data are used. When one piece of acquired data is used, it is judged as positive, negative, or unknown, and unknown pieces are evaluated as having an ambiguous positive / negative judgment. On the other hand, when multiple pieces of acquired data are used, after individually judging each piece as positive, negative, or unknown, pieces where the judgment results of all the data do not match, or pieces where all the judgment criteria are unclear, are evaluated as having an ambiguous positive / negative judgment.

[0129] Then, the display search unit 55 presents the topics as shown below. Highlight ambiguous parts. - Highlight the action immediately before the ambiguity occurs. · Display the parts shown in business records in an easy-to-understand manner. For example, the display search unit 55 displays cases with similar situations (such as the person involved, the business classification, etc.) and where the positive / negative judgment is not ambiguous on the same screen. Then, the link in the business record will take you to the topics in Figure 22, which will be described later.

[0130] Next, a method for providing information that gives notice by the display search unit 55 in step S5 described above will be described in detail.

[0131] Here, Fig. 22 is a diagram showing an example of a method for providing information that gives awareness. Note that the bold parts shown in Fig. 22 are the parts written in the business records.

[0132] In the example shown in Fig. 22, the display search unit 55 highlights the behavior immediately before the response became unclear. The criterion for inconsistency is that evaluations using multiple types of data do not match. When only one type of data is used for judgment, evaluation becomes difficult (in the case of negative / positive, when the probability of a negative result and the probability of a positive result are very close to each other).

[0133] For example, by searching for the work record of whether Mr. C had eaten properly or by searching for keywords such as "Mr. C" and "meals," one arrives at Figure 23, which shows other topics described below. By comparing Figure 22 and Figure 23, one can see that Mr. C is not convinced even though he nodded, which leads to an "insight."

[0134] Here, Figure 23 is a diagram showing another example of a method for providing information that gives awareness. The example shown in Figure 23 is an unambiguous example displayed on the same screen. Note that the bold part shown in Figure 23 is the part written in the business record.

[0135] In the example shown in Figure 23, it would be good if you could realize that the act that caused the ambiguity was recommending salmon to Mr. C without listening to his opinion (I hope you can realize this on your own). Also, it would be good if you could realize that even though Mr. C nodded, there were times when he was not convinced.

[0136] Thus, according to this embodiment, workers can reflect on their work based on the "awareness" they gain from recognizing contradictions buried in their everyday work, and this makes it possible to make wider use of each worker's useful work behavior than with work improvement activities based on the know-how of experts to date, thereby achieving a high level of results.

[0137] (Third embodiment) Next, a third embodiment will be described.

[0138] The third embodiment differs from the first embodiment in that it uses a method for analyzing contradictory elements in topics by specifying the part where a contradiction occurs in the situation of the worker and the work object (including a human). In the following explanation of the third embodiment, the explanation of the same parts as in the first embodiment will be omitted, and only the parts that differ from the first embodiment will be explained.

[0139] FIG. 24 is a flowchart schematically showing the flow of work behavior evaluation processing in the information processing device 50 according to the third embodiment.

[0140] 24, when compiling information packages into topics, the first processing unit 52 (topic creation processing unit 526, etc.) judges each content written in the topic as positive or negative by machine learning, and includes the result in the topic (step S3). The positive / negative judgment is classified into three categories: positive, negative, and neutral.

[0141] When displaying related topics, the display search unit 55 considers a question and a response as one set and checks the degree of discrepancy between positive and negative evaluations within one set. The caregiver asks a question, and the responses up to the next question are considered as one set. If the caregiver's state is neutral, it is determined that there is no problem whether Mr. C's response is negative or positive.

[0142] Then, the display search unit 55 presents the topics as shown below. - Highlight areas where there is a discrepancy between the positive and negative images. · Display the parts shown in business records in an easy-to-understand manner. In response to a request, the display search unit 55 displays cases with similar situations (such as the person involved, the business classification, etc.) and where the positive / negative judgment is not ambiguous on the same screen. This allows related information to be searched for, and hints for solving the problem to be found. Then, links in the business records will take you to the topics in Figure 25, which will be described later.

[0143] Next, a method for providing information that gives notice by the display search unit 55 in step S5 described above will be described in detail.

[0144] Here, Fig. 25 is a diagram showing an example of a method for providing information that gives awareness. Note that the bold parts shown in Fig. 25 are the parts written in the business records.

[0145] In the example shown in Fig. 25, the display search unit 55 highlights cases where there is a mismatch between the negative and positive aspects of a set of actions and responses as a criterion for inconsistency. However, if the response is neutral, it is not considered to be inconsistent, regardless of the response.

[0146] For example, by searching for communication with Mr. C, one arrives at Figure 26, which shows a pattern without emotional mismatch, as described below. Here, one should "realize" that moving forward without asking for Mr. C's reply will cause an emotional mismatch. If one does not understand the reason here either, one should "realize" the know-how necessary for communication with Mr. C by referring to conversations between caregivers, as in the first embodiment.

[0147] Here, Figure 26 is a diagram showing another example of a method for providing information that raises awareness. The example shown in Figure 26 is an unambiguous example displayed on the same screen. Note that the bold part shown in Figure 26 is the part written in the business record.

[0148] In the example shown in Figure 26, we notice that Mr. C had negative feelings when we recommended salmon to him, but it may be difficult for him to realize that the reason for this was that we did not ask for a reply. Therefore, we want him to "become aware" by using a method of providing information that will give him awareness, as shown in Figure 17.

[0149] As described above, according to this embodiment, by setting appropriate evaluation criteria, the inconsistencies in the work behavior information included in the topic group are evaluated, and the inconsistencies hidden in a series of work behaviors are clarified, thereby presenting information that encourages "awareness." As a result, workers can reflect on their work based on the "awareness" gained by recognizing the inconsistencies hidden in everyday work, and achieve higher levels of results than previous work improvement activities based on the know-how of experts.

[0150] Each function of the above-described embodiments can be realized by one or more processing circuits. Here, the term "processing circuit" in this specification includes a processor programmed to perform each function by software, such as a processor implemented by an electronic circuit, as well as devices such as an ASIC (Application Specific Integrated Circuit), a DSP (Digital Signal Processor), an FPGA (Field Programmable Gate Array), and conventional circuit modules designed to perform each of the above-described functions.

[0151] The information processing device 50 is not limited to a personal computer as long as it is a device equipped with a communication function. The information processing device 50 may be, for example, an image forming device, a PJ (Projector), an IWB (Interactive White Board: a white board with an electronic blackboard function capable of mutual communication), an output device such as digital signage, a HUD (Head Up Display) device, industrial machinery, an imaging device, a sound collection device, medical equipment, a network home appliance, an automobile (Connected Car), a notebook PC (Personal Computer), a mobile phone, a smartphone, a tablet terminal, a game console, a PDA (Personal Digital Assistant), a digital camera, a wearable PC, a desktop PC, or the like.

[0152] In addition, in Figure 1, data such as voice is acquired using wearable device 30, but this is not limited to this, and measurements can be shared by wearable devices such as glasses or a wristwatch, or measuring equipment such as a surveillance camera installed in a separate location can be used in combination.

[0153] For example, aspects of the present invention are as follows. <1> an information acquisition unit that acquires information relating to at least the worker, the worker's behavioral response to the work object, and the work environment; a packaging processing unit that divides the information acquired by the information acquisition unit into units that are meaningful in business by using features of the information, and packages the divided units; an evaluation processing unit that evaluates elements included in the information packaged by the packaging processing unit; a topic creation processing unit that creates a topic including elements included in the information packaged by the packaging processing unit and evaluation results of the elements; a display search unit that displays contradictory elements in the topic according to a criterion for indicating contradictions defined for a series of element evaluation results in the topic created by the topic creation processing unit; A work behavior evaluation device comprising: <2> The display search unit evaluates the difference from the standard response of the work recipient and specifies the part with the large difference as a method of analyzing the contradictory elements in the topic according to the criteria indicating the contradiction in the topic. Characterized by <1> The work behavior evaluation device according to claim 1. <3> The display search unit specifies a portion where the evaluation results of two or more types of data do not match as a method of analyzing contradictory elements in the topic according to a criterion indicating contradictions in the topic. Characterized by <1> The work behavior evaluation device according to claim 1. <4> The display search unit specifies a part where an evaluation result of one type of data cannot be determined as a method of analyzing contradictory elements in the topic according to a criterion indicating contradictions in the topic. Characterized by <1> The work behavior evaluation device according to claim 1. <5> the display search unit specifies a part where a contradiction occurs between the status of the worker and the workpiece as a method of analyzing contradictory elements in the topic according to a criterion indicating contradictions in the topic; Characterized by <1> The work behavior evaluation device according to claim 1. <6> the display and search unit indicates the location where the contradiction occurs. Characterized by <1> Or <5> 10. The work behavior evaluation device according to claim 9, wherein <7> the information acquired by the information acquisition unit is continuous in time; Characterized by <1> Or <6> 10. The work behavior evaluation device according to claim 9, wherein <8> The system further includes a document creation processing unit that creates a business record that is created from partial information of the topic created by the topic creation processing unit and that has a link to the original topic attached thereto. Characterized by <2> Or <7> 10. The work behavior evaluation device according to claim 9, wherein <9> <1> Or <8> a work behavior evaluation device according to any one of the preceding items; an information processing terminal that displays contradictory elements in the topics indicated by the work behavior evaluation device; A work behavior evaluation system comprising: <10> an information acquisition step of acquiring information relating to at least the worker, the worker's behavioral response to the work object, and the work environment; a packaging processing step of dividing the information acquired by the information acquisition step into units meaningful for business purposes using features of the information and packaging the divided information; an evaluation processing step of evaluating elements included in the information packaged by the packaging processing step; a topic creation process step of creating a topic including elements included in the information packaged by the packaging process step and evaluation results of the elements; a display and search step of displaying contradictory elements in the topic according to a criterion for indicating contradiction defined for a series of element evaluation results in the topic created by the topic creation processing step; A work behavior evaluation method comprising: <11> Computer, an information acquisition unit that acquires information relating to at least the worker, the worker's behavioral response to the work object, and the work environment; a packaging processing unit that divides the information acquired by the information acquisition unit into units that are meaningful in business by using features of the information, and packages the divided units; an evaluation processing unit that evaluates elements included in the information packaged by the packaging processing unit; a topic creation processing unit that creates a topic including elements included in the information packaged by the packaging processing unit and evaluation results of the elements; a display search unit that displays contradictory elements in the topic according to a criterion for indicating contradictions defined for a series of element evaluation results in the topic created by the topic creation processing unit; A program to function as a [Explanation of symbols]

[0154] 10 Information processing terminal 50 Work behavior evaluation device 51 Information Acquisition Department 55 Display search section 100 Work Behavior Evaluation System 524 Packaging Processing Unit 525 Evaluation Processing Unit 526 Topics Creation Processing Unit 533 Document Creation Processing Unit [Prior art documents] [Patent documents]

[0155] [Patent Document 1] Patent No. 7052062

Claims

1. an information acquisition unit that acquires information relating to at least the worker, the worker's behavioral response to the work object, and the work environment; a packaging processing unit that divides the information acquired by the information acquisition unit into units that are meaningful in business by using features of the information, and packages the divided units; an evaluation processing unit that evaluates elements included in the information packaged by the packaging processing unit; a topic creation processing unit that creates a topic including elements included in the information packaged by the packaging processing unit and evaluation results of the elements; a display search unit that displays contradictory elements in the topic according to a criterion for indicating contradictions defined for a series of element evaluation results in the topic created by the topic creation processing unit; A work behavior evaluation device comprising:

2. The display search unit evaluates the difference from the standard response of the work recipient and specifies the part with the large difference as a method of analyzing the contradictory elements in the topic according to the criteria indicating the contradiction in the topic.

2. The work behavior evaluation device according to claim 1, wherein:

3. The display search unit specifies a portion where the evaluation results of two or more types of data do not match as a method of analyzing contradictory elements in the topic according to a criterion indicating contradictions in the topic.

2. The work behavior evaluation device according to claim 1, wherein:

4. The display search unit specifies a part where an evaluation result of one type of data cannot be determined as a method of analyzing contradictory elements in the topic according to a criterion indicating contradictions in the topic.

2. The work behavior evaluation device according to claim 1, wherein:

5. the display search unit specifies a part where a contradiction occurs between the status of the worker and the workpiece as a method of analyzing contradictory elements in the topic according to a criterion indicating contradictions in the topic; 2. The work behavior evaluation device according to claim 1, wherein:

6. the display and search unit indicates the location where the contradiction occurs.

2. The work behavior evaluation device according to claim 1, wherein:

7. the information acquired by the information acquisition unit is continuous in time; 2. The work behavior evaluation device according to claim 1, wherein:

8. The system further includes a document creation processing unit that creates a business record that is created from partial information of the topic created by the topic creation processing unit and that has a link to the original topic attached thereto.

2. The work behavior evaluation device according to claim 1, wherein:

9. The work behavior evaluation device according to any one of claims 1 to 8, an information processing terminal that displays contradictory elements in the topics indicated by the work behavior evaluation device; A work behavior evaluation system comprising:

10. an information acquisition step of acquiring information relating to at least the worker, the worker's behavioral response to the work object, and the work environment; a packaging processing step of dividing the information acquired by the information acquisition step into units meaningful for business purposes using features of the information and packaging the divided information; an evaluation processing step of evaluating elements included in the information packaged by the packaging processing step; a topic creation process step of creating a topic including elements included in the information packaged by the packaging process step and evaluation results of the elements; a display and search step of displaying contradictory elements in the topic according to a criterion for indicating contradiction defined for a series of element evaluation results in the topic created by the topic creation processing step; A work behavior evaluation method comprising:

11. Computer, an information acquisition unit that acquires information relating to at least the worker, the worker's behavioral response to the work object, and the work environment; a packaging processing unit that divides the information acquired by the information acquisition unit into units that are meaningful in business by using features of the information, and packages the divided units; an evaluation processing unit that evaluates elements included in the information packaged by the packaging processing unit; a topic creation processing unit that creates a topic including elements included in the information packaged by the packaging processing unit and evaluation results of the elements; a display search unit that displays contradictory elements in the topic according to a criterion for indicating contradictions defined for a series of element evaluation results in the topic created by the topic creation processing unit; A program to function as a

Images