JP2021018751A - Program, information processing method and information processing device - Google Patents

Abstract

To provide a program, etc. capable of optimally supporting danger prediction activities.SOLUTION: A program causes a computer to execute processing of: displaying an interactive screen continuously disposing an input statement and an output statement, on a display section; receiving, in the interactive screen, an input of the input statement representing underlying danger items in a work site, from a worker; and, when receiving the input of the input statement representing the danger items, displaying, on the interactive screen, the output statement representing countermeasures for the danger items.SELECTED DRAWING: Figure 6

Description

The present invention relates to programs, information processing methods and information processing devices.

At work sites such as construction and manufacturing, in order to prevent accidents, disasters, etc., an initiative called danger prediction training or danger prediction activity (TBM; Tool Box Meeting) is being carried out. Various systems have been proposed to support this effort.

For example, in Patent Document 1, it is a group confirmation activity support system that supports group confirmation activities including danger prediction activities, and accepts designation of work contents on a mobile terminal possessed by a prospective worker and corresponds to the designated work contents. A group confirmation activity support system that displays notes in sheet format is disclosed.

Japanese Patent No. 6510701

However, the invention according to Patent Document 1 merely digitizes the danger prediction activity that has been conventionally performed in writing, and improves the convenience of the worker who performs the danger prediction activity, or the worker engages in the danger prediction activity. It has not yet provided effective motivation.

One aspect is to provide a program or the like that can suitably support risk prediction activities.

The program according to one aspect displays a dialogue screen on which input sentences and output sentences are continuously arranged on the display unit, and on the dialogue screen, an operator inputs the input sentence indicating a potential danger at the work site. When the input of the input sentence representing the dangerous matter is received, the computer is made to execute the process of displaying the output sentence representing the countermeasure for the dangerous matter on the dialogue screen.

On one side, risk prediction activities can be suitably supported.

It is a schematic diagram which shows the configuration example of the danger prediction support system. It is a block diagram which shows the configuration example of a server. It is a block diagram which shows the configuration example of a worker terminal. It is explanatory drawing which shows an example of the record layout of a worker DB, a dialogue table, and a dialogue history DB. It is explanatory drawing which shows an example of the dialogue screen. It is explanatory drawing which shows an example of the dialogue screen. It is explanatory drawing which shows an example of the management screen. It is explanatory drawing which shows an example of the management screen. It is explanatory drawing about the guide processing during work. It is a flowchart which shows an example of the processing procedure of an interactive process. It is a flowchart which shows an example of the processing procedure executed by an administrator terminal. It is a flowchart which shows an example of the processing procedure of a guide process. It is a block diagram which shows the configuration example of the server which concerns on Embodiment 2. FIG. It is explanatory drawing which shows an example of the record layout of a summary table. It is explanatory drawing which shows an example of the evaluation report about risk prediction. It is a flowchart which shows an example of the processing procedure of the report generation processing. It is explanatory drawing about a dialogue model. It is explanatory drawing which shows an example of the dialogue screen which concerns on Embodiment 3. It is a flowchart which shows the procedure of the generation process of an interaction model. It is a flowchart which shows the procedure of the dialogue processing which concerns on Embodiment 3.

Hereinafter, the present invention will be described in detail with reference to the drawings showing the embodiments thereof.
(Embodiment 1)
FIG. 1 is a schematic diagram showing a configuration example of a danger prediction support system. In the present embodiment, a danger prediction support system that supports danger prediction activities performed by a worker performing a predetermined work at a work site will be described. The danger prediction support system includes an information processing device 1, worker terminals 2, 2, 2 ..., And an administrator terminal 3. Each device is communicated and connected via a network N such as the Internet.

The information processing device 1 is an information processing device capable of transmitting and receiving various types of information processing and information, and is, for example, a server device, a personal computer, or the like. In the present embodiment, the information processing device 1 is assumed to be a server device, and in the following, it will be read as server 1 for the sake of brevity. The server 1 is a process that presents measures to prevent accidents, disasters, etc. caused by the dangerous matters that the worker has listed as potential dangerous matters at the work site in order to support the danger prediction activity. I do. Specifically, as will be described later, the server 1 provides a chatbot system that interactively inputs and outputs texts, and presents countermeasures to dangerous matters to workers through dialogue with the chatbot. In the present embodiment, as an example of the work, the description will be given on the assumption that a single worker who patrolls the work site independently performs work such as installation, inspection, and maintenance of electrical equipment (for example, an electric energy meter).

In the following, the work performed by the worker will be described as the work related to the installation, inspection, maintenance, etc. of electrical equipment, but the work performed by the worker is not limited to this, for example, construction, civil engineering, manufacturing. This system may be applied to a work site such as a business. In addition, although the present embodiment targets single workers, it is also possible to support risk prediction activities for group workers.

The worker terminal 2 is a terminal device possessed by each worker, and is, for example, a mobile terminal such as a smartphone or a tablet terminal. In the present embodiment, the worker terminal 2 will be described as a smartphone. The worker terminal 2 communicates with the server 1, displays a dialogue screen (see FIG. 5 and the like) described later, and inputs / outputs texts, images, and the like related to the danger prediction activity on the dialogue screen.

The administrator terminal 3 is a terminal device operated by an administrator who manages a plurality of workers, and is, for example, an information processing terminal such as a personal computer, a tablet terminal, or a smartphone. The server 1 inputs the setting input of the dialogue scenario (combination of the input sentence input from the worker and the output sentence output from the chatbot side with respect to the input sentence) by the chatbot via the administrator terminal 3. At the same time as accepting, the chat status on each worker terminal 2 and the statistical data of the danger prediction activity by each worker are output to the administrator terminal 3 and presented to the administrator.

The goggles 4 are so-called VR (Virtual Reality) goggles, which are wearable devices to which the worker terminal 2 can be attached. For example, during the work, the worker visually recognizes the work site through the worker terminal 2 with the goggles 4, and performs the work while receiving the attention points of the work object and the presentation (guide) of other alerts.

FIG. 2 is a block diagram showing a configuration example of the server 1. The server 1 has a control unit 11, a main storage unit 12, a communication unit 13, and an auxiliary storage unit 14.
The control unit 11 has one or more CPUs (Central Processing Units), MPUs (Micro-Processing Units), GPUs (Graphics Processing Units) and other arithmetic processing units, and stores the program P1 stored in the auxiliary storage unit 14. By reading and executing, various information processing, control processing, etc. are performed. The main storage unit 12 is a temporary storage area for SRAM (Static Random Access Memory), DRAM (Dynamic Random Access Memory), flash memory, etc., and temporarily stores data necessary for the control unit 11 to execute arithmetic processing. Remember. The communication unit 13 is a communication module for performing processing related to communication, and transmits / receives information to / from the outside.

The auxiliary storage unit 14 is a non-volatile storage area such as a hard disk or a large-capacity memory, and stores a program P1 and other data necessary for the control unit 11 to execute processing. In addition, the auxiliary storage unit 14 stores the worker DB 141, the dialogue table 142, and the dialogue history DB 143. The worker DB 141 is a database that stores information on each worker. The dialogue table 142 is a table (dialogue model) that defines a combination of dialogue sentences (input sentences and output sentences). The dialogue history DB 143 is a database that stores the history of dialogues (input / output history of input sentences and output sentences) performed in the worker terminal 2 of each worker.

The auxiliary storage unit 14 may be an external storage device connected to the server 1. Further, the server 1 may be a multi-computer composed of a plurality of computers, or may be a virtual machine virtually constructed by software.

Further, in the present embodiment, the server 1 is not limited to the above configuration, and may include, for example, an input unit that accepts operation input, a display unit that displays an image, and the like. Further, the server 1 is provided with a reading unit that reads a portable storage medium 1a such as a CD (Compact Disk) -ROM, a DVD (Digital Versatile Disc) -ROM, and reads and executes the program P1 from the portable storage medium 1a. You can do it. Alternatively, the server 1 may read the program P1 from the semiconductor memory 1b.

FIG. 3 is a block diagram showing a configuration example of the worker terminal 2. It includes a control unit 21, a main storage unit 22, a communication unit 23, a display unit 24, an input unit 25, an imaging unit 26, a position acquisition unit 27, and an auxiliary storage unit 28.
The control unit 21 has one or more CPUs, MPUs, and other arithmetic processing devices, and by reading and executing the program P2 stored in the auxiliary storage unit 28, various information processing related to the worker terminal 2 can be performed. Perform control processing, etc. The main storage unit 22 is a temporary storage area such as a RAM, and temporarily stores data necessary for the control unit 21 to execute arithmetic processing. The communication unit 23 includes an antenna for communication, a processing circuit, and the like, and transmits / receives information to / from the outside. The display unit 24 is a display device such as a liquid crystal display or an organic EL (Electro Luminescence) display, and displays an image given by the control unit 21. The input unit 25 is an operation interface such as a touch panel, and inputs the operation content to the control unit 21. The image pickup unit 26 is a camera provided with an image pickup element such as a CMOS (Complementary MOS) sensor, and captures an image. The position acquisition unit 27 is, for example, a communication module that receives a GPS (Global Positioning System) signal, and acquires the position information of the worker terminal 2 (work site). The auxiliary storage unit 28 is a non-volatile storage area such as a ROM (Read Only Memory), and stores the program P2 and other data necessary for the control unit 21 to execute processing.

The worker terminal 2 may include a reading unit that reads the portable storage medium 2a, and may read the program P2 from the portable storage medium 2a and execute the program P2. Alternatively, the worker terminal 2 may read the program P2 from the semiconductor memory 2b.

FIG. 4 is an explanatory diagram showing an example of the record layout of the worker DB 141, the dialogue table 142, and the dialogue history DB 143.
The worker DB 141 includes a worker ID column and a name column. The worker ID column stores a worker ID for identifying each worker. The name column stores the worker's name in association with the worker ID.

The dialogue table 142 includes a message ID column, a scenario phase column, an output type column, an output message column, a post-output operation column, and a transition destination column. The message ID string is a response output from the chatbot side, and stores a message ID for identifying each response prepared in advance according to the scenario. The scenario phase column, output type column, output message column, post-output operation column, and transition destination column are associated with the message ID, respectively, and the phase and output type (text) on the scenario that outputs the response indicated by the message ID. (Or another image), the content of the output message (text or image), the type of chatbot operation after output, and the information of the next output (transition) message are stored.

The dialogue history DB 143 includes a session ID column, a worker ID column, a date and time column, an input type column, an input column, an output column, a selection ID column, and an evaluation column. The session ID column stores a session ID for identifying a dialogue session (chat room) between the worker and the chatbot. The worker ID column, date and time column, input type column, input column, output column, selection ID column, and evaluation column are associated with the session ID, respectively, and the worker ID of the worker who interacted with the chatbot in the dialogue session, Date and time when input and output statements were input and output, input type (whether it is a choice format, etc.), text input by the worker, message output from the chatbot (text or image), output message of the chatbot The message ID of the text selected by the worker next and the evaluation entered by the worker are stored.

5 and 6 are explanatory views showing an example of the dialogue screen. 5A to 6B are screens for the worker to interact with the chatbot, respectively, and are an input sentence input by the worker and an output sentence output from the chatbot side (output as a response to the input sentence). In addition to the text, the dialogue screen is shown in which the text actively output from the chatbot side is continuously arranged and displayed. The interactive processing with the chatbot on the worker terminal 2 will be described with reference to FIGS. 5A to 6B.

The dialogue screen related to this system is provided as an API (Application Programmable Interface) using a Web browser as a medium. For example, a dedicated application program may be installed on the worker terminal 2 and the dedicated browser may be launched by executing the application program. The worker terminal 2 communicates with the server 1 based on the operation input from the worker and displays the dialogue screen.

The worker terminal 2 first displays the dialogue screen of FIG. 5A. Specifically, the worker terminal 2 displays a predetermined icon representing a chatbot and displays a predetermined output sentence in the form of a balloon. For example, in addition to the greeting text, the worker terminal 2 displays a question text prompting the selection of the confirmation item and a selection of the confirmation items by default.

The options include, for example, "danger prediction" for conducting dialogues related to danger prediction, "work procedure" for confirming work procedures, and "work points (caution points)" for confirming work procedures. About the points of work ". Note that these items to be confirmed are examples, and other items may be prepared as default menus. The worker terminal 2 accepts a selection input for selecting one of the options.

In the example of FIG. 5A, the case where "danger prediction" is selected is illustrated. In this case, the worker terminal 2 displays the text of "danger prediction", that is, the input sentence input (selected) by the worker in the form of a balloon. When "danger prediction" is selected, the worker terminal 2 displays an output sentence (response sentence) from the chatbot on the dialogue screen in order to proceed with the dialogue related to the danger prediction.

When "About work procedure" or "About work point" is selected, the worker terminal 2 is a message (text, image, etc.) from the chatbot indicating the work procedure, point, etc. that the worker is going to perform. Are displayed in sequence. In the present embodiment, only the dialogue process related to danger prediction will be described, and detailed illustration and description of the dialogue process related to work procedures and points will be omitted.

The combination of the above input and output statements, that is, the scenario of the dialogue, is defined in the dialogue table 142. When the worker terminal 2 receives the input of the input sentence from the worker, the worker terminal 2 sequentially outputs the input sentence to the server 1 and requests the output sentence from the chatbot. The server 1 reads the output sentence corresponding to the input sentence acquired from the worker terminal 2 from the dialogue table 142, outputs it to the worker terminal 2, and sequentially displays it on the dialogue screen. The output message from the chatbot may include not only text (output text) but also images and the like.

Further, in the present embodiment, a case where the input sentence option is presented from the chatbot and the input sentence is accepted by accepting the selection input from the worker will be described. However, the worker terminal 2 has an input field at the bottom of the dialogue screen. It is also possible to accept input of input sentences (text) in free description format. When the input of the input sentence is accepted in the free description format, the server 1 analyzes the input sentence, determines the output sentence to be output from the chatbot side, and outputs it to the worker terminal 2. The method of analyzing the input sentence is not particularly limited. For example, the server 1 performs parsing, determines an input sentence similar to the text input by the worker from the dialogue table 142, and outputs an output sentence corresponding to the determined input sentence. Output to the worker terminal 2.

The description will be continued based on FIG. 5A. When "danger prediction" is selected, the worker terminal 2 further includes a question sentence asking the type of work to be performed by the worker, and a plurality of texts (input sentences) indicating the work type as options for answering the question sentence. Is displayed.

The worker terminal 2 accepts a selection input for selecting one of a plurality of texts indicating the work type. When any text (work type) is selected, the worker terminal 2 displays information on past work sites (cases) for the selected type of work. For example, the worker terminal 2 displays texts and images related to a work site where an accident, disaster, or the like has occurred, as illustrated in FIG. 5B. In this way, the worker terminal 2 displays information on a work site of the same type as the work to be performed by the worker, that is, a similar site similar to the work site on which the worker plans to work, and calls the worker's attention.

In the above, it is determined whether or not the work itself is similar to the past case according to the type of the work itself, but the present embodiment is not limited to this. For example, even if the worker inputs the site name of the work site to be worked on, the model of the work object (electric energy meter), etc. as an input sentence, and judges whether or not it is similar to the past case from the input sentence. Good. As described above, it is sufficient that the information on the similar site can be determined based on the input sentence regarding the work site where the work is scheduled, and the input sentence is not limited to the work type.

In the dialogue table 142, information on similar sites is stored in association with the text (input sentence options) representing the work type illustrated in FIG. 5A. When the work type is selected, the server 1 randomly selects one of the similar sites among the plurality of similar sites associated with the selected work type and outputs the output to the worker terminal 2. In the present specification, "random" means that the data to be output and the order thereof are different each time it is output (displayed), and it does not have to be random in a strict sense. Even when the work of the same work type is performed, by randomly outputting information on different similar sites, the same case is presented each time, and it is possible to prevent the confirmation work of the worker from becoming slow.

The number of similar sites output to the worker terminal 2 is not limited to one, and may be multiple.

After displaying the information on the similar site, the worker terminal 2 displays a question text asking about potential dangers in the work site where the worker is going to work, as illustrated in FIG. 6A. Then, the worker terminal 2 accepts the input of the input sentence indicating the dangerous matter at the work site. For example, the worker terminal 2 displays a plurality of input sentence candidates 61, 61, 61 ... Representing candidates for dangerous matters, and accepts a selection input for selecting one of the displayed input sentence candidates 61.

In the example of FIG. 6A, first, the worker terminal 2 displays a question sentence asking the dangerous matter itself, and accepts the selection input of the input sentence candidate 61. Subsequently, the worker terminal 2 displays a question sentence asking the cause of the selected dangerous matter, and accepts the selection input of the input sentence candidate 61. In this way, the worker terminal 2 displays one or more question sentences, and accepts the selection input of the input sentence candidate 61 for each question sentence. In this way, the worker can naturally confirm the dangerous matters that can be considered at the work site where he / she works, that is, perform the danger prediction activity by sequentially inputting according to the dialogue scenario.

Here, even when a plurality of input sentence candidates 61, 61, 61 ... Are displayed for the same question sentence, the worker terminal 2 randomly changes the display order of the input sentence candidates 61 each time it is displayed. To do. If the input sentence candidates 61 are displayed in the same order each time, the operator may slow down the input work and select a dangerous matter by judging only by the order. Therefore, by randomly changing the display order of the input sentence candidate 61, it is possible to prompt the worker to confirm the input sentence candidate 61 every time, and it is possible to correctly perform the danger prediction activity.

When the selected input of the input sentence candidate 61 is accepted, the worker terminal 2 takes measures to prevent accidents, disasters, etc. due to the dangerous matters represented by the selected input sentence candidate 61, as illustrated in FIG. 6B. Display the output statement that represents. The output sentence is stored in the dialogue table 142 in association with the input sentence candidate 61. The worker terminal 2 outputs the input sentence candidate 61 selected by the worker to the server 1, acquires the output sentence for presenting the countermeasure corresponding to the input sentence candidate 61 to the worker, and displays it on the dialogue screen. .. As a result, the worker can perform danger prediction activities and receive instruction on countermeasures for dangerous matters.

The server 1 may output not only a text (output sentence) of countermeasures against dangerous matters but also a message including an image.

After displaying the output text indicating the countermeasures against the dangerous matters, the worker terminal 2 displays a question text asking the worker whether the displayed output text was useful, and also displays an evaluation button for accepting the evaluation input. 62, 62 is displayed. The worker terminal 2 accepts the evaluation input of the countermeasure (output sentence) presented from the chatbot side by accepting the selection input of the evaluation button 62. The input / output history of input and output statements input / output to the dialogue screen, including the evaluation input by the worker, that is, the dialogue history between the worker and the chatbot is sequentially transmitted to the server 1 and sent to the dialogue history DB 143. Be remembered.

7 and 8 are explanatory views showing an example of the management screen. 7 and 8 show an example of a screen for an administrator who manages each worker. The processing contents in the administrator terminal 3 will be described with reference to FIGS. 7 and 8.

On the management screen, it is possible to set and display various information related to the chatbot, such as setting a dialogue scenario and checking the dialogue status of each worker. In the present embodiment, the server 1 receives the setting input of the dialogue scenario (combination of the input statement and the output statement) from the administrator terminal 3 via the management screen, and stores the setting contents in the dialogue table 142. Detailed illustration and explanation of the setting of the dialogue scenario will be omitted.

The administrator terminal 3 displays the management screen illustrated in FIG. 7 based on the operation input from the administrator. The screen of FIG. 7 is a dashboard that displays the dialogue status of each worker on the worker terminals 2, 2, 2, ... While sequentially updating in real time. When displaying the screen of FIG. 7, the administrator terminal 3 acquires the dialogue sentences (input sentences or output sentences) input / output from each worker terminal 2 and displays them in order from the latest time point. For example, in the dashboard, as illustrated in FIG. 7, the session ID related to each dialogue sentence, the date and time of input / output, the input sentence input by the worker, the output sentence from the chatbot side for the input sentence, etc. are tabulated. Listed in.

When a new dialogue sentence is input / output on any of the worker terminals 2, 2, 2, ..., The dashboard is sequentially updated and the latest dialogue sentence is added to the top row. In this way, the administrator terminal 3 displays the dialogue that each worker is conducting with the chatbot in real time, and enables the administrator to confirm the dialogue status. As a result, the manager can easily confirm whether each worker is engaged in danger prediction activities and what kind of danger matters are raised.

In addition, for example, the administrator terminal 3 may further enable the worker terminal 2 of each worker to be notified of a message via the management screen. The message notification method is not particularly limited, but is performed by, for example, sending a short mail message or posting a comment on the dialogue screen. As a result, the administrator can send instructions to the operator, and necessary logistical support such as thorough risk prediction activities and work advice can be easily provided.

Further, the administrator terminal 3 displays the management screen illustrated in FIG. 8 based on the operation input from the administrator. The screen of FIG. 8 is a display screen for displaying statistical data that aggregates the dialogue history of all workers. When the screen of FIG. 8 is displayed on the administrator terminal 3, the server 1 aggregates the dialogue history (input / output history) of each worker stored in the dialogue history DB 143 based on a predetermined standard, and generates statistical data. Is displayed on the administrator terminal 3.

The above aggregation criteria are not particularly limited, but for example, in addition to the number of accesses from each worker, the number of dialogues (number of inputs and outputs), the number of selections of input sentence candidates presented as candidates for dangerous matters, and the number of free description formats. Includes the number of input sentences and the number of evaluations for the output sentence presented as a countermeasure against dangerous matters. It should be noted that all of the above criteria are examples, and the statistical data may be any aggregated dialogue history of a plurality of workers. As illustrated in FIG. 8, the administrator terminal 3 displays various aggregation results in a graph or tabular format, and presents the administrator with the dialogue status of all the workers. As a result, the manager can easily grasp the frequency of dialogues and the like, and can get a bird's-eye view of the danger prediction activities of all the workers.

FIG. 9 is an explanatory diagram regarding the guide process during the work. In the present embodiment, not only the danger prediction activity before the work is supported, but also the guide process for guiding the work is performed by issuing a warning or the like to the worker during the work.

For example, during work, the worker wears the worker terminal 2 on the head with the goggles 4, and the worker terminal 2 captures and displays the image in real time to the work site or the work object (the present embodiment). Then, visually check the electricity meter). That is, the worker terminal 2 (smartphone) is made to function as a display of so-called smart goggles.

In the present embodiment, the worker terminal 2 is attached, but dedicated glasses (so-called smart glasses) provided with a transmissive display may be used. Further, the configuration in which the wearable device is used to guide the worker is not essential, and the warning information described later may be simply output to the worker terminal 2.

The worker terminal 2 mounted on the worker's head captures the field of view of the worker with a camera (imaging unit 26) facing the worker's line of sight, and displays the captured image on the display unit 24. To do. The worker performs the work while visually recognizing the captured image displayed on the worker terminal 2. During the work, the worker terminal 2 outputs predetermined information in order to guide the worker.

For example, the worker terminal 2 acquires work environment information regarding the work environment in which the worker is working from the outside, and outputs warning information to the worker based on the acquired work environment information. The work environment information is, for example, the current time or the temperature at the work site (current position). The worker terminal 2 determines whether or not the current time has reached a predetermined break time, and if it determines that the break time has come, outputs warning information prompting the break. Further, the worker terminal 2 acquires the position information of the current position by the position acquisition unit 27, and acquires the temperature of the current position from the external system based on the position information. For example, the worker terminal 2 outputs warning information to the effect that heat stroke should be noted when the acquired air temperature is equal to or higher than a predetermined value. The worker terminal 2 may display the warning information in the form of text or the like by superimposing it on the captured image (image of the work site) visually recognized by the worker, or may output it by voice.

In the above, temperature and the like are given as an example of work environment information, but humidity and weather may be used, for example. In addition, the content of the warning is not limited to breaks and heat stroke prevention, and may be any warning related to the work environment.

Further, for example, the worker terminal 2 performs image recognition on the captured image and recognizes a work object to be worked on by the worker. Then, the worker terminal 2 displays the caution points that the worker should pay attention to in the work object in a mode different from other points by color coding or the like. For example, when the work object is a watt hour meter, the worker terminal 2 recognizes a part (for example, a screw) for fixing the watt hour meter as a caution point and displays the part in a color-coded manner. It should be noted that the reason why the screws are color-coded as caution points is that it causes a poor construction of the work related to the watt hour meter. In FIG. 9, for convenience of illustration, the color-coded display is shown by a rectangular frame.

In the above, fixed parts (screws) are mentioned as points to be noted, but other parts such as screws at the cable connection portion, which are likely to cause a short circuit accident, may be used. Further, for example, the worker terminal 2 recognizes a plurality of work points (parts) to be performed by the worker as caution points, and sequentially changes the color-coded display points according to a predetermined work procedure, thereby performing the work procedure. You may try to guide. As described above, the worker terminal 2 may recognize the cautioned portion and display it in a mode different from that of other portions, and the cautioned portion and the display method are not particularly limited.

Further, as shown in FIG. 9, not only the screen display may be used to call attention, but also voice output may be used for guidance.

The worker terminal 2 further recognizes the state of the above-mentioned caution points, and changes the display color (display mode) according to the recognition result. Explaining according to the above example, the worker terminal 2 recognizes the state of the screw, determines whether or not the screw is tightened, and changes the display color if the screw is tightened. In FIG. 9, for convenience of illustration, the change in display color is indicated by the thickness of the rectangular frame line. In this way, the worker terminal 2 determines whether or not the work is properly completed based on the state of the cautioned points, and when the work is completed, changes the display color and presents it to the worker. As a result, the worker can intuitively understand whether or not the work has been completed properly.

From the above, according to the present embodiment, it is possible to perform the danger prediction activity without making the worker bored by inputting the danger matter in the chat format. In addition, the chatbot side can present countermeasures for dangerous matters to improve the convenience of risk prediction activities. Moreover, not only the dialogue between the worker and the chatbot, but also the support to the manager and the support during the work can be provided, and various support including the danger prediction activity can be provided.

FIG. 10 is a flowchart showing an example of the processing procedure of the interactive processing. Based on FIG. 10, a dialogue process between the worker and the chatbot via the dialogue screen will be described.
The worker terminal 2 displays a dialogue screen in which an input sentence input by the worker and an output sentence output to the input sentence are continuously arranged (step S11). For example, in step S11, the worker terminal 2 first displays each confirmation item such as a work procedure and a point as an option in addition to the danger prediction, and prompts the operator to select the confirmation item.

The worker terminal 2 accepts a selection input for selecting one from a plurality of confirmation items (step S12). The worker terminal 2 determines whether or not to perform danger prediction according to the selection result (step S13). When it is determined that the danger prediction is not performed (S13: NO), the worker terminal 2 sequentially displays texts representing the work procedure, work points, etc. according to the operation input from the worker (step S14), and steps Move to S21.

When it is determined that the danger prediction is to be performed (S13: YES), the worker terminal 2 displays a question sentence asking the work type and accepts a selection input for selecting an input sentence indicating the work type (step S15). The worker terminal 2 displays information on a similar site randomly selected based on the selected input sentence (work type) on the dialogue screen (step S16).

The worker terminal 2 displays a plurality of input sentence candidates representing candidates for potential hazards at the work site regarding the work site where the worker is scheduled to work (step S17). The worker terminal 2 receives a selection input for selecting one of the displayed plurality of input sentence candidates from the worker (step S18).

The worker terminal 2 displays an output sentence indicating a countermeasure against a dangerous matter represented by the selected input sentence candidate (step S19). The worker terminal 2 receives an evaluation input from the worker to see if the displayed countermeasure was useful (step S20).

After executing the process of step S14 or S20, the worker terminal 2 determines whether or not to end the series of processes depending on, for example, whether or not the operation of closing the dialogue screen is accepted (step S21). When it is determined that the process is not completed (S21: NO), the worker terminal 2 returns the process to step S12. When it is determined that the process is completed (S21: YES), the worker terminal 2 ends the series of processes.

FIG. 11 is a flowchart showing an example of a processing procedure executed by the administrator terminal 3. The processing contents executed by the administrator terminal 3 will be described with reference to FIG.
The administrator terminal 3 displays a management screen for setting a dialogue scenario between the worker and the chatbot, confirming the dialogue status, and the like (step S41). The administrator terminal 3 determines whether or not to set the dialogue scenario based on the operation input from the administrator (step S42). When it is determined that the dialogue scenario is set (S42: YES), the administrator terminal 3 accepts the dialogue scenario setting input that defines the combination of the input statement and the output statement, and stores the setting contents in the dialogue table 142. (Step S43).

When it is determined that the dialogue scenario is not set (S42: NO), the administrator terminal 3 determines whether or not to confirm the dialogue status between the worker and the chatbot based on the operation input from the administrator. Determine (step S44). When it is determined to confirm the dialogue status (S44: YES), the administrator terminal 3 lists the dialogue sentences (input sentences and output sentences) input / output from the worker terminals 2, 2, 2 ... Of each worker. The dashboard to be shown is displayed (step S45). In this case, the administrator terminal 3 displays the dialogue sentence shown on the dashboard while updating it in real time each time the input sentence and the output sentence are input / output to each worker terminal 2.

When it is determined that the dialogue status is not confirmed (S44: NO), whether or not the administrator terminal 3 confirms the statistical data that aggregates the dialogue history (input / output history of the input sentence and the output sentence) of each of the plurality of workers. (Step S46). When it is determined to confirm the statistical data (S46: YES), the administrator terminal 3 displays the statistical data obtained by totaling the dialogue history of the plurality of workers based on a predetermined standard (step S47). In addition to the number of accesses, the number of dialogues, and the like, the statistical data aggregation criteria include, for example, the number of times input sentence candidates are selected, the number of times input sentences are input in a free description format, and the number of evaluations for output sentences.

If NO in step S46, or after executing the processes of steps S43, S45, and S47, whether or not the administrator terminal 3 ends a series of processes depending on whether or not the operation of closing the management screen is accepted. Is determined (step S48). When it is determined that the process is not completed (S48: NO), the administrator terminal 3 returns the process to step S42. When it is determined that the processing is completed (S48: YES), the administrator terminal 3 ends a series of processing.

FIG. 12 is a flowchart showing an example of the processing procedure of the guide processing. Based on FIG. 12, the contents of the process in which the worker terminal 2 outputs the work guide while the worker is working will be described.
The worker terminal 2 attached to the worker's head by the goggles 4 captures the visual field range of the worker by the image pickup unit 26, and displays the captured image on the display unit 24 (step S61). The worker terminal 2 continues imaging while executing the process of the flow of FIG. 12, and displays the captured image in real time.

The worker terminal 2 acquires work environment information regarding the work environment in which the worker is working (step S62). The work environment information is, for example, the current time or the temperature at the work site. The worker terminal 2 determines whether or not to output the warning information based on the acquired work environment information (step S63). For example, the worker terminal 2 determines whether or not the worker should take a break based on the current time. Further, for example, the worker terminal 2 acquires the air temperature of the work site based on the position information of the worker, and determines whether or not the air temperature is equal to or higher than a predetermined value. When it is determined that the warning information should not be output (S63: NO), the worker terminal 2 shifts the process to step S65. When it is determined that the warning information should be output (S63: YES), the worker terminal 2 outputs the warning information according to the determination result (step S64).

The worker terminal 2 performs image recognition for recognizing a work object, and displays a cautioned portion of the object in a mode different from other locations by color coding or the like (step S65). For example, when the object is a watt-hour meter, the parts (screws) that fix the watt-hour meter are recognized and color-coded.

The worker terminal 2 recognizes the state of the cautioned portion displayed in different colors, and determines whether or not the work is completed according to the recognition result (step S66). If it is determined that the process has not been completed (S66: NO), the worker terminal 2 returns the process to step S61. When it is determined that the completion is completed (S66: YES), the worker terminal 2 changes the display mode by changing the display color of the cautioned portion (step S67), and ends the series of processes.

From the above, according to the first embodiment, it is possible to suitably support the danger prediction activity by inputting the dangerous matter and outputting (displaying) the countermeasure for the dangerous matter in an interactive manner.

Further, according to the first embodiment, it is possible to more preferably support the danger prediction activity by selecting from a plurality of candidates for danger matters.

Further, according to the first embodiment, the danger prediction activity can be more preferably carried out by randomly changing the display order of the candidates for the danger matter.

Further, according to the first embodiment, it is possible to promote the prevention of the occurrence of disasters, accidents, etc. by presenting information on similar sites.

In addition, according to the first embodiment, support during work can also be provided.

In addition, according to the first embodiment, support for the administrator can also be provided.

(Embodiment 2)
In this embodiment, a mode in which a worker is involved in a danger prediction activity based on a dialogue history (input / output history) between a worker and a chatbot will be described. The contents overlapping with the first embodiment are designated by the same reference numerals and the description thereof will be omitted.
FIG. 13 is a block diagram showing a configuration example of the server 1 according to the second embodiment. The auxiliary storage unit 14 of the server 1 according to the present embodiment stores the aggregation table 144. The aggregation table 144 is a table for aggregating the answer results that each worker answers (inputs) about the dangerous matter.

FIG. 14 is an explanatory diagram showing an example of the record layout of the aggregation table 144. The summary table 144 includes a question sequence, a selection column, a selection sequence, and a total answer sequence. The question string is a question sentence output from the chatbot side in order to prompt the input of an input sentence regarding a dangerous matter, and stores an output sentence indicating a work type. The selection column stores input sentence candidates that can be selected by the operator for the question sentence in association with the question sentence. The selection sequence stores the total result of the selection number in which each worker selects each input sentence candidate in association with the question sentence and the input sentence candidate. The total number of answers stores the total number of selections of each input sentence candidate, that is, the total number of answers, in association with the question sentence.

FIG. 15 is an explanatory diagram showing an example of an evaluation report on risk prediction. An outline of the present embodiment will be described with reference to FIG.
For example, the server 1 receives an output request from the administrator terminal 3, generates the report shown in FIG. 15, and outputs the report to the administrator terminal 3. This report is a personal report that describes the worker's evaluation of risk prediction activities.

The server 1 evaluates each worker for danger prediction based on the dialogue history (input / output history of input sentences and output sentences) of each worker stored in the dialogue history DB 143. In the present specification, "evaluating a worker for danger prediction" means to quantitatively or qualitatively evaluate the ability of a worker to detect a dangerous matter or deal with a dangerous matter. The server 1 evaluates the worker based on the history of the input sentence input by the worker as a dangerous matter.

The algorithm for evaluating the worker is not particularly limited, but in the present embodiment, the server 1 aggregates the input sentence input by each worker as a dangerous matter, that is, the answer result of the dangerous matter, and according to the aggregated result of the answer. Evaluate the worker.

For example, the server 1 totals the number of selections selected by each worker for each input sentence candidate for each input sentence candidate output (displayed) as a candidate for a dangerous matter. In addition, the server 1 totals the total number of selections of all input sentence candidates, that is, the total number of responses. Then, the server 1 divides the selection number of each input sentence candidate by the total number of responses, and determines (calculates) the points to be given to the worker who selected each input sentence candidate. The points are points assigned to evaluate whether or not each input sentence candidate should be listed as a dangerous matter and the validity of the judgment. For example, when there are three input sentence candidates and the number of selections of each input sentence candidate is 50, 30, and 20 times, the points of each input sentence candidate are 0.5 points, 0.3 points, and 0 points. . 2 points. The server 1 determines the points to be given for each input sentence candidate, which is an answer option, for each work type.

The server 1 gives points to each question sentence for which the worker answers (inputs) a dangerous matter according to the input sentence candidate (dangerous matter) actually selected by the worker, and the worker accumulates the given points. Calculate the score for each. The score is a score indicating the level of the worker's ability regarding risk prediction. For example, the worker answered the above-mentioned question sentence related to danger prediction three times, and selected 0.5 point, 0.4 point, and 0.3 point input sentence candidates for each question sentence, respectively. In that case, the score of the worker is 1.2 points. For example, the server 1 calculates (cumulatively) the worker's score for each work type.

The server 1 calculates the average value of the scores of all the workers. Then, the server 1 calculates the evaluation value of the worker regarding the danger prediction by dividing the score of each worker by the average value. For example, if the worker's score is 1.2 points and the average value is 1.0, the evaluation value is 1.2. The server 1 calculates an average value for each work type and evaluates the worker.

The above evaluation (calculation) method is an example, and the present embodiment is not limited to this. For example, the server 1 may set predetermined points for each input sentence candidate in advance and calculate the score according to the defined points. Further, for example, the server 1 may total the number of evaluations input by each worker by selecting the above-mentioned evaluation button 62, and weight the points of each input sentence candidate according to the totaling result of the number of evaluations. .. Further, for example, the server 1 may increase or decrease the score of the worker according to the result of the work actually performed by the worker (for example, whether or not an accident has occurred). As described above, the server 1 may be appropriately evaluated based on the history of the input sentence input by the worker as a dangerous matter, and the specific evaluation method thereof is not particularly limited.

The server 1 generates the report of FIG. 15 based on the above processing result. For example, the server 1 generates charts 151 and 152 showing the evaluation values of the workers for each work type. The charts 151 and 152 are evaluation results based on the scores for the past month and the past week, respectively, and the numerical axes of the charts 151 and 152 represent the work type. In addition, the "achieved level" represents the evaluation value of the workers, the "initial value" represents the evaluation value of the previous month or last week, and the "current value average" represents the average of all workers.

In addition, the server 1 generates a table 153 showing the aggregation result of the number of selections of each input sentence candidate by the target worker, and also generates a comment 154 selected by the template according to the evaluation result. The server 1 outputs a report summarizing each of the above information to the administrator terminal 3. As a result, the analysis result of analyzing the tendency of individual risk prediction is presented to the manager, and the risk prediction activity can be more preferably supported.

FIG. 16 is a flowchart showing an example of a processing procedure of the report generation processing. Based on FIG. 16, the contents of the process of evaluating the worker for risk prediction and generating a report will be described.
The server 1 refers to the dialogue history DB 143 and aggregates the input sentences input by each worker as dangerous matters (step S201). For example, the server 1 totals the number of selections selected by each worker for each input sentence candidate which is a candidate for a dangerous matter. Based on the aggregation result, the server 1 determines the points to be given to the worker when each input sentence candidate is selected (step S202). For example, the server 1 determines the points to be given when each input sentence candidate is selected based on the number of selections of each input sentence candidate and the total number of selections of all the input sentence candidates.

The server 1 assigns points according to the input sentence candidates selected by each worker according to the points determined above, and calculates the score, which is the cumulative value of the given points, for each worker (step S203).

The server 1 calculates the average value of the scores of all the workers (step S204). Then, the server 1 evaluates each worker for danger prediction based on the calculated average value and the score of each worker (step S205). For example, the server 1 calculates a value obtained by dividing the score of each worker by an average value as an evaluation value. The server 1 generates a report showing the evaluation result and outputs it to the administrator terminal 3 (step S206). Server 1 ends a series of processes.

From the above, according to the second embodiment, it is possible to present the analysis result of analyzing the tendency of individual risk prediction.

(Embodiment 3)
In the first embodiment, a mode for outputting an output sentence from the chatbot based on the rule-based dialogue table 142 (dialogue model) has been described. In this embodiment, a mode in which machine learning for learning dialogue sentences (input sentences and output sentences) for teachers is performed and an output sentence is output using the dialogue model 142A generated by the machine learning will be described.

FIG. 17 is an explanatory diagram of the dialogue model 142A. In the present embodiment, the server 1 performs machine learning to learn the input sentence input from the worker and the output sentence to be output as a response to the input sentence, and generates the dialogue model 142A.

For example, the dialogue model 142A is a neural network generated by deep learning, and is a Seq2Seq (Sequence to Sequence) which is a kind of RNN (Recurrent Neural Network). The dialogue model 142A has an input layer that accepts input of an input sentence, an intermediate layer that performs an operation based on the input sentence, and an output layer that outputs an output sentence based on the calculation result in the intermediate layer.

The input layer has a plurality of neurons that receive input of each character string obtained by dividing an input sentence into predetermined units (for example, word units, phrase units, etc.). Each neuron accepts the input of each character string according to the order of appearance in the text (input sentence).

The middle layer is composed of multiple neurons called LSTM blocks. Each neuron in the middle layer performs an operation based on the character string input to the corresponding neuron in the input layer. In the LSTM block, the calculation is performed with reference to the calculation result in the previous LSTM block (neuron), that is, the calculation result based on the character string appearing immediately before. As a result, the dialogue model 142A predicts the output sentence based on the sequence (appearance order) of the character strings constituting the input sentence.

The output layer has a plurality of neurons that sequentially output each character string constituting the output sentence according to the order of appearance. Each neuron predicts the character string to be output based on the calculation result of the corresponding neuron in the middle layer.

As shown in FIG. 17, the neural network related to Seq2Seq is classified into a network that functions as an encoder and a network that functions as a Decoder. The Encoder accepts the input of each character string constituting the input sentence and performs the calculation in the intermediate layer. The Decoder sequentially outputs each character string constituting the output sentence based on a plurality of character strings input to the Encoder. Specifically, as shown in FIG. 17, the Decider first predicts the first character string, inputs the predicted first character string into the next LSTM block, and predicts the second character string. The Decider repeats the process, predicts the third and subsequent character strings in sequence, and finally predicts all the character strings.

Note that Seq2Seq is an example of the dialogue model 142A, and the server 1 may generate another neural network as the dialogue model 142A. Further, the dialogue model 142A is not limited to the neural network, and may be a model based on other learning algorithms such as a decision tree, a random forest, and an SVM (Support Vector Machine).

For example, the server 1 uses the dialogue history (combination of input sentences and output sentences) accumulated in the dialogue history DB 143 by the processing of the first embodiment as teacher data to generate the dialogue model 142A. The teacher data may be separately created manually. The server 1 divides the input sentence for the teacher into a character string of a predetermined unit, sequentially inputs it to the dialogue model 142A, and acquires the output sentence from the output layer. The server 1 compares the correct output sentence included in the teacher data with the output sentence output from the dialogue model 142A, and optimizes various parameters such as weights used for the calculation in the intermediate layer so that the two are approximated. To be. As a result, the server 1 generates the dialogue model 142A.

In the present embodiment, the server 1 predicts an output sentence using the above-mentioned dialogue model 142A, and outputs an output sentence indicating a countermeasure for the dangerous item to the input sentence input by the operator as a dangerous item. .. FIG. 18 is an explanatory diagram showing an example of the dialogue screen according to the third embodiment. For example, the worker terminal 2 accepts the input of an input sentence representing a dangerous matter in a free description format by the input field at the bottom of the dialogue screen of FIG. The worker terminal 2 outputs an input sentence to the server 1.

When the input sentence is acquired from the worker terminal 2, the server 1 inputs the input sentence into the dialogue model 142A and acquires the output sentence. Specifically, as in the case of learning, the server 1 divides the input sentence into character strings of a predetermined unit and sequentially inputs them into the dialogue model 142A, and sequentially acquires each character string constituting the output sentence. The server 1 outputs an output sentence consisting of each output character string to the worker terminal 2 and displays it on the dialogue screen.

In the above, the explanation was made assuming the case of inputting dangerous matters and outputting countermeasures for the dangerous matters, but other cases such as inputting / outputting texts related to past work cases (similar sites), etc. It goes without saying that the output sentence may be predicted by using the dialogue model 142A also in the dialogue scene of.

FIG. 19 is a flowchart showing the procedure of the generation process of the dialogue model 142A. Based on FIG. 19, the content of the process of performing machine learning to generate the dialogue model 142A will be described.
The server 1 acquires teacher data including an input sentence input from the worker and an output sentence to be output for the input sentence (step S301). The server 1 divides the input statement and the output statement into character strings of predetermined units, respectively (step S302).

The server 1 generates an interaction model 142A that takes an input sentence from an operator as an input and outputs an output sentence based on the character strings of the divided input sentence and the output sentence (step S303). For example, the dialogue model 142A is a neural network related to Seq2Seq, which is a neural network that takes each character string that constitutes an input sentence as an input and outputs each character string that constitutes an output sentence. The server 1 sequentially inputs each character string constituting the input sentence into the dialogue model 142A according to the order of appearance in the text, and sequentially outputs each character string constituting the output sentence according to the order of appearance. The server 1 compares the output character string with the character string of the correct output sentence included in the teacher data, optimizes various parameters of the intermediate layer so that they are similar to each other, and generates the dialogue model 142A. Server 1 ends a series of processes.

FIG. 20 is a flowchart showing the procedure of the dialogue processing according to the third embodiment. An interactive process using the interactive model 142A will be described with reference to FIG.
The worker terminal 2 displays a dialogue screen in which an input sentence input by the worker and an output sentence output to the input sentence are continuously arranged (step S321). The worker terminal 2 displays a predetermined question sentence (for example, a question sentence asking a dangerous matter) prompting the input of the input sentence on the dialogue screen (step S322).

The worker terminal 2 receives an input sentence (for example, an input sentence indicating a dangerous matter) via the dialogue screen and outputs it to the server 1 (step S323). The server 1 inputs an input sentence to the dialogue model 142A, acquires an output sentence (for example, an output sentence indicating a countermeasure against a dangerous matter), outputs it to the worker terminal 2, and displays it on the dialogue screen (step S324). Specifically, the server 1 divides the input sentence into character strings of a predetermined unit, sequentially inputs each character string to the dialogue model 142A according to the order of appearance in the text, and inputs each character string constituting the output sentence to the dialogue model. It is sequentially acquired from 142A and output to the worker terminal 2.

The worker terminal 2 receives an evaluation input for the output sentence displayed on the dialogue screen (step S325). The worker terminal 2 determines whether or not to end a series of processes depending on whether or not an operation for closing the dialogue screen has been accepted (step S326). When it is determined that the process is not completed (S326: NO), the worker terminal 2 displays the next question text on the dialogue screen (step S327), and returns the process to step S323. When it is determined to end (S326: YES), the worker terminal 2 stores the input and output statements input and output on the dialogue screen in the dialogue history DB 143 of the server 1 (step S328), and ends the series of processes. ..

For example, the server 1 may output an output statement by combining the rule-based dialogue table 142 described in the first embodiment and the machine learning dialogue model 142A described in the present embodiment. For example, the server 1 outputs an output statement using the dialogue model 142A generated by machine learning only when the input statement representing a dangerous matter is acquired, and outputs an output statement based on the dialogue table 142 in other cases. It may be. In this way, the server 1 may combine a rule-based model and a machine learning model.

From the above, according to the third embodiment, a more appropriate output sentence can be output by using the dialogue model 142A generated by machine learning.

The embodiments disclosed this time should be considered to be exemplary in all respects and not restrictive. The scope of the present invention is indicated by the scope of claims, not the above-mentioned meaning, and is intended to include all modifications within the meaning and scope equivalent to the scope of claims.

1 Server (information processing device)
11 Control unit 12 Main storage unit 13 Communication unit 14 Auxiliary storage unit P1 Program 141 Worker DB
142 Dialogue table 143 Dialogue history DB
2 Worker terminal 21 Control unit 22 Main storage unit 23 Communication unit 24 Display unit 25 Input unit 26 Imaging unit 27 Position acquisition unit 28 Auxiliary storage unit P2 Program 3 Administrator terminal 4 Goggles

Claims (10)

A dialogue screen for continuously arranging input and output sentences is displayed on the display.
On the dialogue screen, the operator accepts the input of the input sentence indicating the danger matter latent in the work site.
A program characterized in that, when an input of the input sentence representing the dangerous matter is received, a computer is made to execute a process of displaying the output sentence representing the countermeasure for the dangerous matter on the dialogue screen. A plurality of input sentence candidates representing the candidates for the dangerous matter are displayed on the dialogue screen.
Accepting a selection input from the worker to select one from the input sentence candidates,
The program according to claim 1, wherein the output sentence representing the countermeasure against the dangerous matter represented by the selected input sentence candidate is displayed. The program according to claim 2, wherein the display order of a plurality of the input sentence candidates is changed each time the input sentence candidates are displayed on the dialogue screen. On the dialogue screen, the worker accepts the input of the input sentence regarding the work site where the work is scheduled, and receives the input.
The program according to any one of claims 1 to 3, wherein when the input of the input sentence regarding the work site is received, the information about the similar site similar to the work site is displayed on the dialogue screen. .. Acquire work environment information about the work environment in which the worker is working,
The program according to any one of claims 1 to 4, wherein warning information for the worker is output according to the work environment information. A dialogue screen for continuously arranging input and output sentences is displayed on the display.
On the dialogue screen, the operator accepts the input of the input sentence indicating the danger matter latent in the work site.
An information processing method, characterized in that, when an input of the input sentence representing the dangerous matter is accepted, a computer is made to execute a process of displaying the output sentence representing the countermeasure against the dangerous matter on the dialogue screen. An acquisition unit that acquires input sentences that represent potential dangers at the work site and are input by the operator via a dialogue screen that continuously arranges input and output sentences.
Based on the dialogue model that defines the combination of the input sentence and the output sentence, it is characterized by including an output unit that outputs the output sentence representing the countermeasure to the dangerous matter represented by the acquired input sentence to the dialogue screen. Information processing device. A storage unit that stores the input history of the input sentence input from the worker, and
The information processing apparatus according to claim 7, further comprising an evaluation unit that evaluates the worker for risk prediction based on the input history. The acquisition unit sequentially acquires the input sentence from the terminal devices of each of the plurality of workers who display the dialogue screen.
The output unit sequentially outputs the output sentence to the terminal device of each of the plurality of workers.
The information according to claim 7 or 8, further comprising a second output unit that sequentially updates and outputs a list of the input sentences and output sentences input / output to the terminal device of each of the plurality of workers. Processing equipment. Claims 7 to 9 are characterized in that the output unit outputs the output sentence by using the dialogue model in which the input sentence and the output sentence to be output for the input sentence have been learned. The information processing apparatus according to any one of the above items.

Images