JP2014229180A - Apparatus, method and program for support of introspection, and device, method and program for interaction - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an apparatus that supports introspection of a user, and guides the user to awareness.SOLUTION: Problem specific means 11 acquires user's occupation, inquires a problem to the user, and acquires the problem answer data entered by the user with respect to the inquiry of the problems. Also, the problem of the user is acquired on the basis of the problem answer data and a related keyword associated with the problem corresponding to the user's occupation of attribute data 21. Problem solution support means 12 performs mirroring of the problem answer data, when the related keyword associated with the problem of the user is included in the problem answer data, and inquires detail of the problem. When the related keyword is not included in the problem answer data, the problem answer data is subjected to the mirroring.

Description

  The present invention relates to an introspection support apparatus, an introspection support method, and an introspection support program that support a user's introspection, and a dialog apparatus, a dialog method, and a dialog program used for these.

  In recent years, an increasing number of users have various problems and problems in various communities such as industry and campus. Users cannot linguize their own problems themselves, and it is difficult for them to deepen introspection and gain awareness by themselves.

  In light of this situation, there are workplaces where counselors (consultants) are placed in the community. However, since the problems and issues of people belonging to a community are specific to that community, there are few counselors that can handle that community. Further, since the counselor can handle 20 to 30 people per month, the situation is overwhelming.

  Therefore, a software agent that can act as a counselor is required. ELIZA is known as such a software agent (see, for example, Non-Patent Document 1). ELIZA described in Non-Patent Document 1 creates a response sentence that returns a text content input by the user. With the parrot return, the user feels that ELIZA is listening. In addition, the dialogue with parrot returns can be easily utilized because it does not require complicated and large amount of expertise.

Joseph Weisenbaum, Tadatoshi Akiba, “Computer Power: Artificial Intelligence and Human Reasons”, Simal Publishing, January 1979

  However, ELIZA described in Non-Patent Document 1 has a problem that problems and targets to be solved are not clarified because it returns only a parrot to information input by the user. In the response sentence by ELIZA, since it is a parrot return and does not maintain and utilize the context, it is difficult for the user to continue the dialogue. For this reason, there is a problem that it is difficult to continually deepen the user's introspection through dialogue, to clarify the problem, and to find out how to solve the problem.

  In addition, in order to solve a problem or problem, the user himself / herself needs to clarify the problem and find a solution. Therefore, the counselor needs to encourage introspection for the user to clarify the problem himself and find a solution.

  That is, as shown in Non-Patent Document 1, it is difficult to guide the user's awareness only by returning the parrot, and it is difficult to solve the problem.

  Therefore, an object of the present invention is to provide an introspection support apparatus, an introspection support method, and an introspection support program that can support and reflect on the user's introspection, as well as a dialog apparatus, a dialog method, and a dialog program used in these Is to provide.

  In order to solve the above problems, a first feature of the present invention relates to a reflection support apparatus that supports a user's reflection. The introspection support device according to the first feature of the present invention relates to a storage device that stores attribute data associating occupations with related keywords of the occupations and tasks, associating occupations with issues corresponding to occupations, and user issues And a problem solving support means for supporting the user's solution to the identified user problem. The problem identification means acquires the user's occupation, inquires the user about the problem, acquires the problem response data input by the user in response to the problem inquiry, and corresponds to the user's occupation of the problem response data and attribute data The user's task is acquired based on the related keyword associated with the task to be performed. The problem solving support means, when the related keyword associated with the user's problem is included in the problem answer data, the problem answer data is mirrored, the details of the problem are inquired, and the related keyword is included in the problem answer data. If not, mirror the answer data.

  Here, the attribute data further associates an occupation keyword and an emotion keyword indicating feelings with respect to the task, and the task solution support means further receives the task detail answer data input by the user in response to the query of the task details. If the related keyword or emotional keyword corresponding to the user's task data is included in the task detail response data, the task detail response data is mirrored, and further details of the task are inquired. When the related keyword and emotion keyword corresponding to the task data are not included, the task detail answer data may be mirrored.

  The storage device also stores template data including agent start greeting data for introducing a counselor represented by the introspection support device and agent experience data regarding the experience of the counselor, and the task specifying means stores the agent start greeting data. After displaying the agent experience data, inquiring the user about his / her occupation and obtaining the user's occupation, the user may be inquired about the experience corresponding to the occupation.

  A second feature of the present invention relates to a reflection support method for supporting a user's reflection. In the introspection support method according to the second aspect of the present invention, the computer stores attribute data associating occupations with tasks corresponding to occupations and associating related keywords of occupations and tasks. The introspection support method includes a step in which a computer identifies a user problem, and a computer assists the user in solving the identified user problem. The step of identifying the assignment includes a step in which the computer acquires the occupation of the user, a step in which the computer inquires about the assignment to the user, and a computer obtains assignment answer data input by the user in response to the inquiry about the assignment. And a step in which the computer acquires the user's task based on the task answer data and the related keyword associated with the task corresponding to the user's occupation in the attribute data. The step of supporting the solution by the user includes a step of mirroring the task answer data and inquiring about the details of the task, when the computer includes a related keyword associated with the user task, If the related keyword is not included in the task answer data, a step of mirroring the task answer data is provided.

  Here, the attribute data further associates an emotion keyword indicating emotion to the task with the occupation and the task, and the step of supporting the solution by the user is further input by the computer in response to the query of the task details. Obtaining detailed task response data and the computer mirrors the detailed task response data when the detailed task response data includes a related keyword or emotional keyword that corresponds to the user task data. The inquiry may include a step of mirroring the task detail answer data when the task detail answer data does not include a related keyword and an emotion keyword corresponding to the user task data.

  The computer stores template data including agent start greeting data for introducing a counselor represented by the computer and agent experience data related to the experience of the counselor. Displaying the agent start greeting data after the computer has displayed the agent start greeting data, the step of displaying the agent experience data and then inquiring the occupation to the user after the computer has displayed the agent experience, After acquiring the user's occupation, the method may further include a step of inquiring the user of an experience corresponding to the occupation.

  A third feature of the present invention relates to an introspection support program that supports introspection of a user. In the introspection support program according to the third feature of the present invention, the computer stores attribute data associating occupations with tasks corresponding to occupations and associating occupations and related keywords of tasks. The introspection support program causes a computer to execute a step of identifying a user problem and a step of assisting the user in solving the identified user problem. The step of identifying the assignment includes the steps of acquiring the occupation of the user, the step of inquiring the user about the assignment, the step of obtaining the assignment answer data input by the user in response to the inquiry of the assignment, the assignment answer data, and the attribute The step of acquiring a user's task based on a related keyword associated with the task corresponding to the user's occupation in the data is provided. The step of supporting the solution by the user includes a step of mirroring the task answer data when the task answer data includes a related keyword associated with the user's task, inquiring about the details of the task, and the task answer data, When the related keyword is not included, a step of mirroring the task answer data is provided.

  Here, the attribute data further associates an occupation keyword and an emotion keyword indicating emotions with respect to the task, and the step of supporting the solution by the user further includes a task detail answer input by the user in response to the task detail inquiry. If the related data or emotional keywords corresponding to the user's task data are included in the task detail response data, the task detail response data is mirrored, and further task details are inquired. When the response data does not include a related keyword and an emotion keyword corresponding to the user's task data, a step of mirroring the task detail response data may be provided.

  The computer stores template data including agent start greeting data for introducing a counselor represented by the computer and agent experience data related to the experience of the counselor, and the step of identifying a task displays the agent start greeting data. A step, displaying an agent start greeting, displaying agent experience data, displaying agent experience data, inquiring the user about his / her occupation, and after the computer obtains the user's occupation, The method may further include a step of inquiring about an experience corresponding to.

  A fourth feature of the present invention relates to an interactive apparatus that interacts with a user on behalf of a counselor. According to a fourth aspect of the present invention, there is provided a dialog device that stores agent start greeting data for introducing a counselor, template data including agent experience data relating to the experience of the counselor, and agent start greeting data. The system includes a trust relationship building means for displaying the agent experience data, inquiring the user about his / her occupation, obtaining the user's occupation and inquiring about the experience corresponding to the user's occupation.

  A fifth feature of the present invention relates to an interactive method in which a computer interacts with a user on behalf of a counselor. In the dialogue method according to the fifth aspect of the present invention, the computer stores template data including agent start greeting data for introducing a counselor and agent experience data regarding the experience of the counselor. The dialog method according to the fifth aspect of the present invention includes a step in which a computer displays agent start greeting data, a step in which the computer displays agent experience greeting data after displaying the agent start greeting data, After displaying the experience data of the agent, the method includes the steps of inquiring the user of the occupation, and the computer inquires the user of the experience corresponding to the occupation after obtaining the occupation of the user.

  A sixth feature of the present invention relates to an interactive program for interacting with a user on behalf of a counselor. In the dialogue program according to the sixth aspect of the present invention, the computer stores template data including agent start greeting data for introducing a counselor and agent experience data relating to the experience of the counselor. An interactive program according to a sixth aspect of the present invention includes a step of displaying agent start greeting data on a computer, a step of displaying agent experience data after displaying the agent start greeting data, and displaying agent experience data. Then, a step of inquiring the occupation to the user and a step of inquiring the user of the experience corresponding to the occupation after acquiring the occupation of the user are executed.

  According to the present invention, there are provided an introspection support device, an introspection support method and an introspection support program that can assist and inspire the user's introspection, and a dialog device, a dialog method, and a dialog program used therefor can do.

It is a flowchart explaining the outline of the reflection support method which concerns on embodiment of this invention. It is a figure explaining the hardware constitutions and functional block of the reflection support apparatus which concern on embodiment of this invention. It is a figure explaining an example of a data structure and data of attribute data concerning an embodiment of the invention. It is a flowchart explaining the introspection support method which concerns on embodiment of this invention. It is a figure explaining the example of a conversation in the introspection support method which concerns on embodiment of this invention. (Part 1) It is a figure explaining the example of a conversation in the introspection support method which concerns on embodiment of this invention. (Part 2) It is a figure explaining the example of a conversation in the introspection support method which concerns on embodiment of this invention. (Part 3) It is a figure explaining the example of a conversation in the introspection support method which concerns on embodiment of this invention. (Part 4) It is a figure explaining evaluation of the number of times of the dialog continuation by the reflection support method concerning an embodiment of the invention. It is a figure explaining the evaluation method of the trust relationship and awareness by the introspection support method which concerns on embodiment of this invention. It is a figure explaining evaluation by the reflection support method concerning an embodiment of the invention, and Drawing 11 (a) explains evaluation of trust relation and Drawing 11 (b) explains evaluation of notice. FIG. It is a figure explaining the hardware constitutions and functional block of the reflection support apparatus which concern on the 1st modification of this invention. It is a figure explaining the module structure of the reflection support apparatus which concerns on the 1st modification of this invention. It is a figure explaining the context orientation in the reflection support apparatus which concerns on the 1st modification of this invention. It is a part of counseling knowledge data of the reflection support apparatus according to the first modification of the present invention, and is a diagram for explaining an example of data referred to in order to organize work experience and experience knowledge. It is ontology dictionary data of the reflection support apparatus which concerns on the 1st modification of this invention, Comprising: It is a figure explaining an example of the data referred in order to specify a job type. It is ontology dictionary data of the reflection support apparatus which concerns on the 1st modification of this invention, Comprising: It is a figure explaining an example of the data referred in order to extract a task classification. In the introspection support apparatus according to the first modified example of the present invention, FIG. In the reflection support apparatus according to the first modified example of the present invention, it is a diagram for explaining the structure of the context object data, It is a figure explaining the hardware constitutions and functional block of the dialogue apparatus which concern on the 2nd modification of this invention.

  Next, embodiments of the present invention will be described with reference to the drawings. In the following description of the drawings, the same or similar parts are denoted by the same or similar reference numerals.

  In the embodiment of the present invention, the “user” is a target person who receives support from the introspection support apparatus according to the embodiment of the present invention, and may be referred to as a client.

  “Introspection” means deep reflection and reflection. The introspection support apparatus according to the embodiment of the present invention creates and outputs a response sentence that prompts the user to introspect according to the context.

  “Mirroring” is to create and output a message that responds to the parrot return of the text input by the other party. For example, consider a case where the user inputs “I want to become an SE but I am worried about getting a job”. By mirroring this, the computer outputs "You want to be SE, but you are worried about getting a job." Thus, in the mirroring process, “I” is replaced with “you”. Further, the computer may create a sentence to be confirmed while asking the user, for example, by adding an additional question sentence in the case of an English sentence, or by converting the ending of “~” in the case of a Japanese sentence.

(Outline of introspection support method)
The introspection support method according to the embodiment of the present invention is executed by the introspection support apparatus 1. The introspection support apparatus 1 is a general computer in which a predetermined introspection support program is installed. The introspection support apparatus 1 functions as an agent (agent) of a counselor that can respond to a problem of a community to which a user belongs.

  With reference to FIG. 1, the outline of the introspection support method according to the embodiment of the present invention will be described. First, when the user activates the introspection support device 1 or the introspection support program in step S1, the process proceeds to step S2.

  In step S2, the introspection support apparatus 1 performs a task identification phase. In the problem identification phase, the introspection support apparatus 1 plays an introductory role for the agent and the user to build a trust relationship while sharing the experience, clarify the outline of the problem, and promote introspection. Therefore, the context of the task identification phase is relatively clear. In the problem identification phase, the introspection support apparatus 1 narrows down occupations to which the user belongs, such as a member of society or a student, prior to the process of building a trust relationship. Then, the introspection support apparatus 1 asks questions about the experience up to the occupation and makes the user answer. In addition, students will reflect on the experiences and knowledge they have gained in each experience section to spend their lives.

  In this way, in the problem identification phase, the introspection support apparatus 1 proceeds with narrowing down occupations, sharing experiences and experience knowledge, and identifying problems. Specifically, in the introspection support device 1, the agent greets the user and provides the agent's own experience content when starting support. Furthermore, the introspection support apparatus 1 acquires the user's occupation, provides the user's own experience contents according to the user's occupation, and assists the user in organizing the experience. As a result, the agent and the user sympathize with each other and contribute to building a trust relationship. Furthermore, the introspection support apparatus 1 narrows down and identifies the user's task type and task from this occupation.

  In step S3, the introspection support apparatus 1 performs the problem solving phase. In the problem solving phase, the introspection support device 1 maintains the relationship of trust established in step S2 and deepens the dialogue with the user in a context corresponding to the problem specified in step S2. Create a response sentence for. The introspection support device 1 responds by using a context that is clearly verbalized by the user, and rephrasing it in accordance with the context or refining the task in order to solve the problem. Encourage ministry. By repeating such processing, the introspection support apparatus 1 supports the user so that he / she becomes aware.

  In step S3, the problem solving phase process is repeated, and in step S4, when the user stops the introspection support apparatus 1 or the introspection support program, the introspection support apparatus 1 ends the process.

  As described above, the introspection support method according to the embodiment of the present invention divides the problem into the problem identification phase and the problem resolution phase, thereby clarifying the framework of the problem. To keep the context and continue the conversation. As a result, the introspection support apparatus 1 can clarify a problem and find a solution itself by the user himself / herself in the continued conversation.

  In the introspection support method according to the embodiment of the present invention, a case where tasks are narrowed down by occupation will be described. At this time, in consideration of the user's experience acquired in step S111, the tasks may be narrowed down or responses may be made. You may create a sentence.

(Introspection support device)
With reference to FIG. 2, the introspection support apparatus 1 which concerns on embodiment of this invention is demonstrated. The introspection support device 1 is a general computer including a central processing control device 10, a storage device 20, an input device 30, an output device 40, and the like.

  The input device 30 is a keyboard or a mouse, and inputs a user instruction to the central processing control device 10. The output device 40 is a display device or the like, and outputs the processing result of the central processing control device 10 in a state that can be recognized by the user. In the embodiment of the present invention, the introspection support device 1 functions as a counselor agent. Therefore, the introspection support device 1 gives the user an experience of chatting with the agent via the input device 30 and the output device 40, and prompts the user for introspection.

  As another embodiment, the introspection support device 1 includes a communication control device (not shown), obtains a user instruction via an input device and an output device connected to a communication network, and transmits information to the user. Or may provide. In the embodiment of the present invention, a chat format using characters is assumed, but the present invention is not limited to this. For example, the introspection support apparatus 1 converts the text data of the agent's response sentence into voice data and outputs it, converts the user's voice into voice data, and converts the voice data into text data (not shown). May be realized in a voice conversation format.

  The storage device 20 stores an introspection support program and also stores attribute data 21 and template data 22.

  The attribute data 21 associates occupations with tasks corresponding to occupations, and associates keywords related to occupations and tasks. The attribute data 21 is an ontology that narrows down the user's issues and sets keywords that prompt the user to refine. The attribute data 21 may include three tables as shown in FIG. 3, for example.

  FIG. 3A shows the first attribute data 21a used for specifying the occupation of the user. The first attribute data 21a is data in which an occupation identifier is associated with a related keyword for determining the occupation. For example, when the introspection support apparatus 1 inquires of the user about his / her occupation, if “science” is included in the answer obtained from the user, the user's occupation is determined to be “science student”.

  FIGS. 3B and 3C show the second attribute data 21b and the third attribute data 21c used when narrowing down the user's tasks after the user's occupation is determined. In the embodiment of the present invention, the second attribute data 21b is used for narrowing down the type of assignment, and the third attribute data 21c is used for narrowing down the specified assignment type to a more detailed assignment. Although narrowing down in stages is assumed, it is not limited to this.

  The second attribute data 21b shown in FIG. 3B is data in which the task type is associated with the related keyword for determining the task type. Here, the task type is provided for each occupation. The second attribute data 21b shown in FIG. 3B lists only the task type related to the occupation “student”. For example, when the introspection support apparatus 1 inquires a user about a problem, if the answer obtained from the user includes “future career”, the user's problem type is determined to be “career development”. . Although not shown in FIG. 3, related keywords for determining the task type are also provided for other occupations such as “IT worker”.

  The third attribute data 21c shown in FIG. 3C is data in which an assignment identifier is associated with a related keyword and an emotion keyword for determining the assignment and encouraging reflection. This emotion keyword is a keyword that is a trigger for prompting the user to materialize the problem and assisting awareness, and specifically includes “worry” and “anxiety”. The third attribute data 21c shown in FIG. 3C lists only the tasks related to the occupation “student” and the task type “career”. Although not shown in FIG. 3, related keywords and emotion keywords for determining a task are also provided for other occupations and task types such as “career” of “IT worker”. The third attribute data 21c is used for narrowing down to occupations and task types from keywords and phrases included in text data input by the user. Thereby, the 3rd attribute data 21c specifies a user's problem.

  The template data 22 is data of an agent line that the introspection support apparatus 1 outputs. This dialogue is determined in advance assuming a counselor acting as the proxy for the introspection support device 1. For example, the template data 22 includes agent start greeting data for introducing a counselor and agent experience data regarding the counselor experience. The data of each line of the template data 22 is read by the central processing control device 10 and output to the output device 40 as appropriate.

  The central processing control device 10 includes a problem identification unit 11 and a problem solution support unit 12.

  The assignment specifying unit 11 specifies an assignment of the user. The process of the task specifying unit 11 corresponds to the process of step S2 in FIG.

  The task specifying means 11 displays the agent start greeting data of the template data 22, displays the agent experience data, inquires the user of the occupation, acquires the user's occupation, and inquires the user of the experience corresponding to the occupation. The assignment specifying unit 11 acquires the occupation of the user, inquires of the user about the assignment, and acquires assignment answer data input by the user in response to the inquiry of the assignment. The assignment specifying unit 11 further acquires the user's assignment based on the assignment answer data and the related keyword associated with the assignment corresponding to the occupation of the user in the attribute data 21.

  The problem solution support means 12 supports the user's solution for the identified user problem. The process of the problem solving support means 12 corresponds to the process of step S3 in FIG.

  The problem solving support means 12 mirrors the problem answer data and inquires about the details of the problem when the related keyword associated with the user's problem is included in the problem answer data input by the user. On the other hand, when the related keyword is not included in the task answer data, the task solution support means 12 mirrors the task answer data.

  Furthermore, the problem solving support means 12 acquires the problem detailed answer data input by the user in response to the problem details inquiry. When the problem detail answer data includes a related keyword or emotion keyword corresponding to the user's problem data, the problem solution support means 12 mirrors the problem detail answer data and inquires about further details of the problem. On the other hand, if the related keyword and emotion keyword corresponding to the user's task data are not included in the task detail response data, the task solution support means 12 mirrors the task detail response data.

(Introspection support method)
With reference to FIG. 4 thru | or 8, the introspection support method which concerns on embodiment of this invention, and its conversation example are demonstrated.

  First, when the user activates the introspection support apparatus 1 and inputs a start greeting, in step S101, the task identification unit 11 acquires user start greeting data generated based on an input signal by the user. Furthermore, the task specifying means 11 outputs the generated user-initiated greeting data to the output device 40 and displays it in a format that can be recognized by the user. For example, the output device 40 displays the sentence ST11 “I am the first time” in FIG.

  Next, in step S103, the task specifying means 11 refers to the template data 22 and extracts agent start greeting data and agent experience data. In step S <b> 104, the task specifying unit 11 outputs agent start greeting data to the output device 40. The agent start greeting data may be data obtained by mirroring the user start greeting data. Subsequently, in step S <b> 105, the task specifying unit 11 outputs the agent experience data to the output device 40. In step S106, a prompt for inquiring about the user's occupation is displayed to prompt the user to input the occupation.

  The output results from step S104 to step S106 correspond to the sentence ST12 in FIG. For example, as shown in FIG. 5, in response to the user inputting “I am the first time”, the task specifying means 11 is “this is the first time” as shown in the sentence ST12. A response sentence mirroring the sentence ST11 is generated and output. Furthermore, the task specifying means 11 outputs the counselor's business experience and experience knowledge as agent experience data to the output device 40, and in order to inquire about his occupation, "Please tell me your current occupation." Generate and output the following text.

  In step S <b> 107, the task specifying unit 11 acquires occupation answer data generated by user input in response to the prompt in step S <b> 106, and outputs it to the output device 40. This occupation response data corresponds to the sentence ST13 “I belong to an information department” in FIG. In step S108, the task identification unit 11 refers to the attribute data 21 and identifies the occupation data of the user from the occupation response data acquired in step S107. “Information” is included in the sentence ST13, and “information” is included in the related keyword of the occupation “science student” in the first attribute data 21a shown in FIG. Accordingly, the task specifying means 11 specifies that the user's occupation is “science student”.

  On the other hand, in step S109, the task specifying unit 11 outputs the occupation data of the user specified in step S108 to the output device 40. Here, the task specifying means 11 may mirror the job answer data generated based on the user input in step S107 and may output the job data specified in step S108. For example, as shown in FIG. 5, the assignment specifying unit 11 displays a sentence ST14 “You belong to an information department, that is, you are a science and engineering student.”

  Here, as shown in FIG. 6, it is assumed that the user inputs a sentence that affirms as the answer, the sentence ST21 “Yes, that is right”. In this case, in step S110, the task identification unit 11 displays a prompt for inquiring about the user's experience. The input of experience here depends on the occupation of the user. Specifically, the task specifying means 11 inquires about the experience corresponding to the occupation of the user with reference to data associating the experience value organizing items for each occupation as shown in FIG. Since the user is a student in the examples shown in FIG. 5 to FIG. 8, the assignment specifying means 11 corresponds to the occupation “student” in FIG. 15, the experience in high school, the purpose of entering the university, Inquire about experience items in sequence. Specifically, the task specifying unit 11 displays a sentence ST22 including a prompt for inquiring about the user's high school experience, which is the next context, and prompts the user to input the experience.

  Here, in step S <b> 111, the task specifying unit 11 acquires the experience answer data generated by the user input in response to the prompt in step S <b> 110, and outputs it to the output device 40. This experience answer data corresponds to the sentence ST23 "I was a club activity in high school." On the other hand, the task identification unit 11 outputs a sentence ST24 that is a mirror of the sentence ST23. Thereby, the task specifying unit 11 prompts the user to talk about the experience.

  The user further inputs experience, and the task specifying means 11 acquires the experience answer data further input by the user and outputs it to the output device 40. This experience answer data corresponds to the sentence ST25 “I was able to participate in the prefectural convention in the last summer of the third year” in FIG. The task specifying means 11 mirrors the sentence ST25 and, when there is no experience to be input, displays a sentence ST26 including a message for prompting selection of “no experience knowledge addition”.

  On the other hand, when the user selects “no experience knowledge addition”, the assignment specifying unit 11 displays a prompt for inquiring about the purpose of entering the university. In response to this, when the user answers the purpose of entering the university and selects “no experience knowledge addition”, the task specifying means 11 further displays a prompt for inquiring about an experience item at the university. In this way, by continuing the conversation shown in FIG. 6, the task specifying unit 11 acquires an experience suitable for the occupation of the user.

  As described above, when the user inputs the experience answer data in step S111, the task specifying unit 11 displays a prompt for inquiring the task type in step S112. The assignment specifying means 11 displays a sentence ST31 “Please tell us if you have any concerns or issues that come to mind here” shown in FIG. Here, in step S113, the task specifying unit 11 acquires task type input data generated by user input in response to the prompt in step S112, and outputs the acquired task type input data to the output device 40. The assignment type input data corresponds to the sentence ST32 “I am worried about whether or not I can learn IT technology steadily” in FIG.

  In step S114, the task specifying unit 11 refers to the attribute data 21, and specifies task type candidate data from the task type input data input in step S113 and the task type related to occupation data specified in step S108. Specifically, the assignment specifying unit 11 refers to the second attribute data 21b, and assigns the assignment types “career”, “relationship”, “school environment” and “school environment” related to the occupation “science student” of the user. Identify “self-anxiety”.

  Accordingly, the assignment specifying unit 11 displays a prompt for inquiring about the assignment type in step S115. Specifically, the task identification unit 11 mirrors the document input in the sentence ST32, and generates and displays a sentence ST33 that inquires the task type identified in step S114 as a candidate for the user task type. This sentence ST33 is, "I'm worried about whether you can learn IT technology steadily. Does it relate to career, human relations, school environment, self-anxiety?"

  On the other hand, when the user inputs task type data for selecting his / her task among “career”, “human relationship”, “school environment” and “self-anxiety” in step S116, the task specifying means 11 The task type selected by the user is displayed. Specifically, the task specifying means 11 displays the sentence ST34 “related to future carriers” based on the user input. The processing of step S115 and step S116 is processed in the conversation between the agent and the user, but the task specifying means 11 may be realized by displaying a radio button for selecting one from the task type candidates. .

  Furthermore, the task specifying unit 11 generates and displays a sentence ST35 that mirrors the sentence ST34 input by the user. The sentence ST35 is "What you are concerned about is related to your future career?" On the other hand, when the sentence ST36 “Yes.” Affirmed by the user is input, the assignment specifying unit 11 determines that the assignment type of the user is “future career” and further narrows down the determined assignment type.

  In step S117, the task specifying unit 11 displays a prompt for inquiring specifically about the task. Specifically, the assignment specifying means 11 generates and displays a sentence ST36 “You have a challenge for the future career. Please organize and discuss the problem specifically” in step S117. This sentence ST36 plays a role of confirming the user's problem "future career" and asking the user to verbalize this problem by asking "Please sort out the problem and talk specifically". .

  On the other hand, when the user inputs the task input data in step S118, the task solution support means 12 displays the task input data. The assignment input data is the sentence ST41 “I want to become an SE but I am worried about getting a job” in FIG.

  On the other hand, the problem solving support means 12 specifies the user's problem in step S119. Specifically, referring to the third attribute data 21c in FIG. 3C, the keyword “I want to be SE” included in the sentence ST41 and the related keyword “I want to be SE” corresponding to the task “job hunting” are shown. It is determined that they match. Thereby, the task specifying unit 11 determines that the user task is “job hunting”.

  Thus, when the user's problem is identified, the problem solution support means 12 supports the user's problem solution without departing from the context regarding the problem “job hunting” identified in this way. Specifically, the problem solving support means 12 determines whether or not there is a keyword related to the identified problem in the problem answer data input by the user in step S120. This keyword includes the related keyword and emotion keyword of the third attribute data 21c shown in FIG. If there is a keyword, the problem solving support means 12 generates and displays a prompt for details in step S121, and prompts the user for further input. At this time, the problem solving support means 12 may generate and display a sentence in which a character string obtained by mirroring text data input by the user and a character string that asks details are connected. On the other hand, if the keyword is not included, the problem solving support means 12 generates and displays a character string obtained by mirroring the text data input by the user in step S122.

  The conversation displayed on the output device 40 in steps S121 to S122 will be described with reference to FIG.

  In the sentence ST41 “I want to become SE but I'm worried about getting a job” generated by the user's input, the related keyword “I want to be SE” in the third attribute data 21c and the emotion keyword “Worry Is included. Therefore, as shown in step S121, the problem solving support means 12 outputs to the output device 40 the sentence ST42 “You want to become an SE, but you are worried about getting a job. In the sentence ST42, “You want to become an SE, but you are worried about getting a job?” Is a mirrored part of the sentence ST41. “Please tell us a little more.” Is a sentence that asks more details when the “anxiety” in the first sentence of the sentence ST42 matches the emotion keyword of the third attribute data 21c. In response to “Please tell us a little more in detail” in the sentence ST42, the problem solving support means 12 prompts the user to input “anxiety” in more detail.

  On the other hand, the sentence ST43 generated by the user's input is displayed on the output device 40 as “I am worried about getting a proper job due to age and ability”. In the sentence ST42, the user is prompted to input “anxiety” in more detail, so that the problem solving support means 12 assists the user in verbalizing the reason for the “anxiety”.

  On the other hand, the problem solving support means 12 outputs to the output device 40 the sentence ST44 “You are anxious about whether you can get a proper job because of your age and ability. Of the sentence ST44, “You are anxious about getting a proper job because of your age and ability” is a mirrored part of the sentence ST43. “Please tell me more in detail.” Is a sentence that asks further details when the “anxiety” in the sentence ST43 matches the emotion keyword of the third attribute data 21c. In response to “Please tell us a little more in detail” in the sentence ST44, the problem solving support means 12 prompts the user to input “anxiety” in more detail.

  On the other hand, the problem solving support means 12 outputs the sentence ST45 "I think it is difficult to get IT-related work in the future because it is slow to swallow IT" generated by the user's input. Output to 40. In the sentence ST44, by prompting the user to input “anxiety” in more detail, the problem solving support means 12 assists the user in easily converting the reason for the anxiety “age and ability” into verbalization.

  On the other hand, the problem solving support means 12 outputs the sentence ST46 “You are slow to swallow IT and think it is difficult to get IT related work in the future” to the output device 40. Sentence ST46, “You think IT is slow to swallow and it will be difficult to get IT-related work in the future.” Is a mirrored part of sentence ST45. In this sentence ST46, the problem solving support means 12 does not display a prompt for inquiring about the details, "Please tell me a little more detail" like the sentence ST42 and the sentence ST44. This is because the preceding sentence ST45 does not include keywords corresponding to emotion keywords and related keywords. In this case, the problem solving support means 12 waits for further input from the user after mirroring as shown in step S122 of FIG.

  Here, the user tries to further verbalize in order to continue the conversation while waiting. Thereby, the problem solving support means 12 outputs the sentence ST47 “I think I should spend more time to learn IT skills” generated to the user's input to the output device 40. On the other hand, the problem solving support means 12 outputs to the output device 40 the sentence ST48 “I think you should spend more time to learn IT skills. In sentence ST48, “You think you should spend more time to learn IT skills” is a mirrored part of sentence ST47. “Please tell us a little more.” Is a sentence that asks more details when “learn IT skills” in the sentence ST47 matches the related keyword in the third attribute data 21c. In response to “Please tell us a little more in detail” in the sentence ST48, the problem solving support means 12 prompts the user to input more details about “learning IT skills”.

  As described above, when the user inputs a sentence including a predetermined keyword with respect to the problem specified by the problem specifying means 11, the problem solving support means 12 prompts the verbalization of the user's idea by asking more details. . Further, when the user himself / herself is verbalized, the problem solving support means 12 can give notice to the user himself / herself and can assist the user himself / herself to verbalize the solving means.

  As described above, the introspection support apparatus 1 according to the embodiment of the present invention guides the user's awareness through the dialog shown in FIGS.

  In the conversation shown in FIG. 5, the introspection support device 1 has a chance to build a trust relationship through mutual experience, and further refers to the attribute data 21 to specify the occupation of the user. The subsequent conversation is advanced based on the occupation context of the user specified here.

  In the conversation shown in FIG. 6, the introspection support apparatus 1 proceeds with the organization of experiential knowledge that promotes reflection in the context narrowed down by the conversation shown in FIG. The introspection support device 1 prompts the user to reflect on his / her own experience and organize his / her own experience knowledge through dialogue. In this process, several patterns are prepared according to occupations such as students and members of society, and questions are repeated.

  In the conversation shown in FIG. 7, the introspection support device 1 prompts the user to verbalize the problems that the user has, and classifies the problems that are likely to be in the user's occupation into 5 to 6 categories (carrier relations, human relations, culture / environment). And the process of the task identification phase is finished. Thereby, the introspection support apparatus 1 can limit the context of the subsequent problem solving phase.

  In the conversation shown in FIG. 8, the introspection support apparatus 1 continues the response that deepens the introspection by matching a plurality of related keywords or a single keyword while interacting with the narrowed problem. Here, it is considered that the introspection support apparatus 1 can prompt the user to continue the introspection when the introspection support apparatus 1 makes an appropriate response for each of about 5 to 10 matching patterns.

  By the above method, the introspection support apparatus 1 according to the embodiment of the present invention can narrow down work experience and solve problems by applying limited pattern knowledge of 1000 or less and knowledge of tens of keywords in each pattern. Enables clarification of the framework. Furthermore, the vocabulary that cannot be grasped by narrowing down and problem framework can solve the problem of knowledge acquisition and explosive increase in search cost by the ELIZA method that randomly selects response sentences that promote dialogue.

(Evaluation)
With reference to FIG. 9 thru | or FIG. 11, the evaluation of the reflection support method which concerns on embodiment of this invention is demonstrated. The introspection support method according to the embodiment of the present invention evaluates with a goal as to how smoothly the user and the agent from the output of the introspection support apparatus 1 can proceed with the conversation. As the criteria, as shown in FIG. 9 to FIG. 11, the number of dialogue continuations, the degree of trust relationship construction, and the degree of awareness are evaluated as scales. Here, ELIZA shown in Non-Patent Document 1 as a prior art will be described in comparison with the present invention.

  As the evaluation in FIGS. 9 to 11, the user to be evaluated was a student in an information processing related department. Furthermore, since having awareness of problems affects the effectiveness of counseling, we selected six students who were aware of problems with their careers. In other words, it was aimed at people who have introspection ability above the standard, seriously worried about IT-related careers, and want counseling.

  With reference to FIG. 9, the evaluation based on the number of dialog continuations will be described. In the conventional invention, an average of 7.5 dialogs was made from the dialog start to the dialog completion. On the other hand, in the present invention, the task identification phase, that is, 13.3 times from the start of the dialogue to the organization of the experience and 10.8 times from the rearrangement of the experience to the completion of the dialogue was 24.1 times in total.

  With reference to FIG. 10, an evaluation method based on the degree of trust relationship construction and the degree of awareness will be described. This evaluation is based on questionnaire responses as shown in FIGS. 10A and 10B by the evaluator. As shown in FIG. 10A, it includes a 10-question questionnaire based on a trust relationship, and as shown in FIG. 10B, it includes a 12-question questionnaire regarding awareness. The answers are "7. I think so", "6. I think so", "5. I think so", "4. I can't say either", "3. I don't think so", "2. A method is adopted in which one of the seven evaluation values of “I don't think” and “1. I don't think so” is selected.

  As shown in FIG. 11A, the average evaluation value for the present invention was 4.39 for the degree of trust relationship construction, whereas the average evaluation value for the prior art was 2.8. there were. As shown in FIG. 11 (b), the average evaluation value for the present invention was 4.50 for the degree of awareness, whereas the average evaluation value for the conventional technique was 3.51. there were.

  The introspection support apparatus 1 according to the embodiment of the present invention supports the establishment of a trust relationship through an interactive process for organizing experience knowledge. At the stage where this process has passed, the number of dialogs of the conventional invention (ELIZA) that does not have this process is 7.5 times including the end of the dialog, whereas the present invention is 24.1 times. Thereby, it turns out that the introspection support apparatus 1 which concerns on embodiment of this application was able to assist so that a smoother dialogue could be made. Further, as shown in FIG. 11, it can be seen that the degree of construction of the trust relationship and the degree of awareness are more effective than the conventional invention. In addition, the introspection support device 1 according to the embodiment of the present invention is to follow the words of emotion, that is, to assist in verbalizing and refining the emotional state and the state of its change. Effective in improving and maintaining trust relationships.

  The introspection support apparatus 1 according to the embodiment of the present invention can promote the user's own reflection and build a sense of trust with the agent when organizing the user's own experience in the dialog with the user. . The introspection support device 1 uses the user's spoken and emotional keywords, and repeats the response while maintaining a thorough context, thereby encouraging the user's consciousness to be organized by the introspection and making improvements noticed. Realization is possible. Thus, the introspection support apparatus 1 can support the user's introspection.

(First modification)
With reference to FIG. 12, the introspection assistance apparatus 1a which concerns on a 1st modification is demonstrated. The introspection support apparatus 1a according to the first modification generates a context object and supports context reflection by inferring context orientation based on the generated context object.

  The storage device 20 stores ontology dictionary data 26, context object data 27, and counseling knowledge data 28. The central processing control device 10 includes a natural language dialogue processing means 16 and a context-oriented reasoning means 17.

  The natural language dialogue processing means 16 extracts vocabulary and structure using ontology dictionary data 26 and the like, and stores them in the context object data 27. The context-oriented reasoning means 17 infers using the stored information of the context object data 27 and the counseling knowledge data 28, which is a specialized knowledge model, and creates a response sentence.

  With reference to FIG. 13, the module configurations of the natural language dialogue processing means 16 and the context-oriented reasoning means 17 will be described.

  The natural language dialogue processing means 16 includes an initial termination processing unit 161, a dialogue text analysis unit 162, and a dialogue text output unit 165 that serve as an interface between the user and the reflection support apparatus 1a.

  The initial end processing unit 161 initializes the context object data 27 and various work information, and generates and outputs a dialog start message when the introspection support device 1a is started up. At the end, the initial end processing unit 161 saves saved information such as a word list, generates an end message, and outputs it.

  The dialog text analysis unit 162 includes a keyword extraction unit 163 and a context analysis unit 164, receives the dialog text, and performs the following processing. The keyword extraction unit 163 extracts keywords included in the text sentence input by the user. The context analysis unit 164 sets new context information according to the contents of the context object data 27 and ontology dictionary data 26.

  The dialog text output unit 165 outputs the response sentence created by the context-oriented reasoning means 17 to the output device 40.

  The context-oriented reasoning means 17 includes a context-based reasoning engine 171 that distributes contexts for task discovery and task realization, and a context-oriented counseling processing unit 172 that generates a response sentence with reference to the counseling knowledge data 28.

  The context-based reasoning engine 171 extracts the user's occupation, extracts the problems the user has by narrowing down, and stores them in the context object data 27.

  The context-oriented counseling processing unit 172 refers to the counseling knowledge data 28 and generates a paraphrase or refined question answer sentence using dialogue sentences and extracted keywords for organizing experience knowledge.

  With reference to FIG. 14, data used in the first modification will be described.

  The ontology dictionary data 26 stores knowledge for context extraction. The context object data 27 stores information for holding a context.

  The counseling knowledge data 28 holds data for context-oriented counseling. The counseling knowledge data 28 is described by the following context-oriented counseling terms and context-oriented counseling rules, for example, as knowledge for performing IT-related beginner level counseling.

  Explain context-oriented counseling terminology.

(1) Job type: IT professionals, science students, humanities students (2) Task types: career development, workplace intrinsic factors, organizational roles, relationships, organizational structure, personal characteristics (3) Issues: careers of science / humanities students For developmental problems: employment, aptitude, disagreement, feasibility, feelings, abilities, motivation, career development problems for IT professionals: desired occupation, occupational aptitude, salary, future occupation (4) Emotional expression: negative emotion (anxiety) , Worry, scary, worry), plus emotions Explain the context-oriented counseling rules. This context-oriented counseling rule corresponds to each AA in FIG. 14 referenced in each sub-context.

  As shown in FIG. 14, the context-oriented counseling rules include a rule related to a problem finding main context from the start of a dialogue to grasping the outline of the problem, and a rule related to a problem solving main context that clarifies the grasped problem and guides awareness. .

  The task discovery main context is a context that starts dialogue, identifies job types, organizes experience knowledge for building trust relationships, and identifies task types. The task finding main context includes an occupation grasping subcontext, a trust relationship subcontext, a task type identification subcontext, and an erroneous input corresponding subcontext.

  The occupation grasping sub-context identifies the occupation type by grasping the occupation term based on the occupation grasp AA. The occupation grasp AA is an execution rule group (AA: Action Agent) of a semantic network that identifies three occupation types expressed by ontology dictionary data 26b shown in FIG. The occupation grasp subcontext transitions to the trust relationship construction subcontext when the narrowing down of job types is completed.

  The trust relationship sub-context is a context in which occupational experience and experience knowledge in the identified job type are arranged based on the trust relationship building AA. The trust relationship construction AA is a group of execution rules of a semantic network that repeatedly asks the grasp contents designated according to occupation. The grasp contents designated in accordance with occupations depend on counseling knowledge data 28 shown in FIG. 15 to be described later, and have several items of experience knowledge for each occupation. The trust relationship sub-context transitions to a task type identification sub-context when the user inputs an instruction indicating the intention of “no experience knowledge addition”.

  The task type identification sub-context is a context for identifying the task type from the keyword representing the task that appears in the dialogue based on the task type identification AA. The task type identification AA clarifies which of the 5 to 10 tasks corresponds to each of the above task types from the task, the task subordinates, and the ontology keywords of emotion expression shown in FIG. When the identification of the task type is completed, a transition is made to the task solver context.

  The sub-context corresponding to erroneous input is a context in which the dialogue is continued even in the case of erroneous input based on the erroneous input corresponding AA. When the target vocabulary does not match in the target context, the erroneous input correspondence AA is regarded as an erroneous input, and one of several erroneous input correspondence response sentences is returned at random. The sub-context corresponding to erroneous input does not change.

  The task solver context is a context that deepens the task type identified by the task finder context and leads to awareness for resolution. The problem solving main context includes a problem clarification subcontext and a context nonconformity corresponding subcontext.

  Based on the detailed context reasoning AA and the global context reasoning AA, the task clarification sub-context is obtained from the task, task subordinate and emotion expression ontology dictionary data 26c shown in FIG. Clarify which of 10 issues falls under this category. The detailed context reasoning AA is a group of execution rules of a semantic network of a multi-item collation type such as a subject, predicate, objective, etc., for acquiring a detailed context corresponding to a task and for a response sentence. The global context inference AA is a group of execution rules of a single item matching type semantic network that creates a global context and creates a response sentence when some of the keywords of a task match. The task clarification subcontext transitions to a context incompatible subcontext when matching fails.

  The context nonconformity sub-context creates a response sentence based on the context nonconformity correspondence AA as it is, or selected randomly from a plurality of context nonconformity correspondence response sentences. The context incompatibility correspondence AA is an execution rule group of a semantic network that creates a response sentence for promoting dialogue when collation fails.

  The counseling knowledge data 28 shown in FIG. 15 is referred to in order to organize occupation and experience knowledge. The introspection support apparatus 1a refers to the counseling knowledge data 28, and can inquire the user about experiences according to the user's occupation.

  The ontology dictionary data 26 will be described with reference to FIGS. The ontology dictionary data 26 is used to extract context from keywords. The ontology dictionary data 26a shown in FIG. 16 is used for extracting job types from occupational terms. The ontology dictionary data 26b shown in FIG. 17 is used for extracting the task type. The ontology dictionary data 26b holds the task type and the task in association with each job type specified in FIG. The ontology dictionary data 26c shown in FIG. 18 is referred to in order to distribute the detailed context in the local inference.

  Referring to FIG. 19, the context object data 27 output from the context analysis unit 164 and referred to by the context-oriented reasoning means 17 will be described. The context object data 27 has items of experience, task type, task, and emotion as attribute names. The context analysis unit 164 sequentially sets each item set in the context object data 27 and sequentially updates the context object data 27 in the context of the interaction between the agent and the user. The context-oriented reasoning means 17 creates a response sentence necessary for prompting the user to reflect in accordance with each item set in the context object data 27.

  The introspection support apparatus 1a according to the first modification of the present invention uses CxBR rules. The CxBR rule (knowledge) has a main context and a sub-context (sub-context, sub-context, etc.). These context rules include transition rules and execution rules. A transition rule is a condition for transferring control to the next context. For example, the transition rule of the trust relationship construction subcontext of FIG. 14 is “if“ no additional learning ”and then transition to the task type identification subcontext”. An execution rule basically calls a rule in a lower context (the lowest is AA). For example, the sub-context rule in FIG. 14 is a function that calls AA. The detailed context reasoning AA creates a response sentence by a plurality of keyword matching. The global context inference AA creates a response sentence by single keyword matching. The context-adaptive adaptation AA continues and facilitates dialogue when keyword matching fails.

  Processing of the introspection support apparatus 1a according to the first modification will be described.

(1) When the introspection support apparatus 1a is called, the initial termination processing unit 161 is activated. The introspection support apparatus 1a creates a dialog start message and outputs a log.

(2) Upon receiving the input sentence, the introspection support apparatus 1a performs the processes (2a) and (2b) in the dialog text analysis unit 162. (2a) The keyword extraction unit 163 extracts a target keyword from the analyzed content. (2b) The context analysis unit 164 determines a new context state using the extracted keyword, ontology dictionary data 26, counseling knowledge data 28, and context object data 27, and stores it in the context object data 27.

(3) The context-oriented reasoning means 17 starts and processes the context-based reasoning engine 171. The context-based reasoning engine 171 performs the following processing.
(3a) Activate the initial (default) main context.
(3b) The following processing is repeated until the target is achieved or until it is determined that it cannot be achieved.
(3b-a) When the mission goal is reached, proceed to (3c) and finish.
(3b-b) When the other mission end criteria is reached, the process proceeds to (3d) and ends.
(3b-c) The condition part of all transition restrictions is evaluated, and if the situation should transition to the new main context, this is activated and the process returns to (3b-a).
(3b-d) If there is a control function, it is executed to control the action of the agent.
(3b-e) The condition part of all behavior restrictions is evaluated, and if the lower context is to be activated, activate it and execute (3b-d) therein.
Return to (3b-f) (3b-a).
(3c) Mission success.
(3d) Mission stop.

  The introspection support apparatus 1a according to the first modified example can promote the user's own introspection and build a sense of trust with the agent when organizing the user's own experience in the dialog with the user. The introspection support device 1a uses the user's spoken and emotional keywords and repeats the response while maintaining a thorough context, thereby organizing the user's awareness by introspection and recognizing improvement. Can be realized. Thus, the user's introspection can also be supported in the processing by the introspection support apparatus 1a according to the first modification.

(Second modification)
In the second modification, an interactive apparatus 2 that interacts with a user on behalf of a counselor will be described. The dialogue apparatus 2 according to the second modification smoothly facilitates the dialogue between the user and the agent among the reflection support apparatus 1 according to the embodiment of the present invention and the reflection support apparatus 1a according to the first modification. Realize functions related to trust building for recommendation. As shown in FIG. 20, the interactive device 2 includes a central processing control device 10, a storage device 20, an input device 30, and an output device 40, and is a general computer on which an interactive program is installed.

  The storage device 20 in the dialogue apparatus 2 stores template data 22 including agent start greeting data for introducing a counselor and agent experience data regarding the experience of the counselor.

  In addition, the dialogue apparatus 2 includes trust relationship building means 18 that assists in building a trust relationship. The trust relationship building means 18 displays the agent start greeting data of the template data 22 and displays the agent experience data. The trust relationship building means 18 obtains the user's occupation after inquiring about the occupation to the user, and inquires about the experience corresponding to the occupation of the user.

  At this time, the trust relationship building means 18 inquires about the experience corresponding to the user's occupation based on the arrangement items of the experience values for each occupation shown in FIG. Specifically, when the user is a science student, the trust relationship building means 18 sequentially inquires about high school experience items, university entrance purposes, and university experience items.

  As described above, the dialogue apparatus 2 according to the second modified example displays the agent greeting data, displays the agent experience data, and then inquires of the user about the occupation. The computer acts as an agent on behalf of a counselor, displays the greeting and experience of this agent, and then asks the user to make the user feel intimate with the computer and enter the user's own information Can be easier.

  Moreover, the dialogue apparatus 2 according to the second modified example can continue the dialogue according to the user attribute by inquiring about the experience according to the user's occupation. Further, by continuing such a dialogue, a situation is created in which the user can easily input his / her own information into the computer. As a result, as described in the embodiment of the present invention and the first modification, counseling according to the user's own attribute can be performed by the computer.

(Other embodiments)
As described above, the embodiments of the present invention and the first and second modifications thereof have been described. However, it should be understood that the description and drawings constituting a part of this disclosure limit the present invention. Absent. From this disclosure, various alternative embodiments, examples, and operational techniques will be apparent to those skilled in the art.

  For example, the introspection support apparatus described in the embodiment of the present invention may be configured on one piece of hardware as shown in FIG. 2, or on a plurality of pieces of hardware depending on the functions and the number of processes. It may be configured. Moreover, you may implement | achieve on the existing information processing apparatus.

  It goes without saying that the present invention includes various embodiments not described herein. Therefore, the technical scope of the present invention is defined only by the invention specifying matters according to the scope of claims reasonable from the above description.

DESCRIPTION OF SYMBOLS 1 Introspection support apparatus 2 Dialogue apparatus 10 Central processing control apparatus 11 Problem identification means 12 Problem solution support means 16 Natural language dialog processing means 17 Context-oriented reasoning means 18 Trust relationship construction means 20 Storage device 21 Attribute data 22 Template data 26 Ontology dictionary Data 27 Context object data 28 Counseling knowledge data 30 Input device 40 Output device 161 Initial termination processing unit 162 Dialogue text analysis unit 163 Keyword extraction unit 164 Context analysis unit 165 Dialogue text output unit 171 Context-based reasoning engine 172 Context-oriented counseling processing unit

Claims (12)

An introspection support device that supports user introspection,
A storage device that associates occupations with tasks corresponding to the occupations, and stores attribute data that associates the related keywords of the occupations and tasks;
A task identifying means for identifying the task of the user;
A problem solution support means for supporting a solution by the user for the identified problem of the user,
The task specifying means includes
Acquiring the user's profession, inquiring the user for an issue, obtaining the issue response data input by the user in response to the issue inquiry, the issue response data, and the occupation of the user of the attribute data Based on the related keywords associated with the problem corresponding to the user's problem,
The problem solving support means is:
If the assignment answer data includes a related keyword associated with the user's assignment, the assignment answer data is mirrored and the details of the assignment are inquired, and the assignment answer data includes the related keyword. If not, a reflection support device comprising mirroring the answer data. The attribute data further associates an emotion keyword indicating an emotion for the task with the occupation and the task,
The problem solving support means further includes
In response to an inquiry about the details of the task, the task details answer data input by the user is acquired, and when the related keyword or emotion keyword corresponding to the user task data is included in the task detail response data, the task In addition to mirroring detailed answer data, inquiries about details of further issues, and when the related keyword and emotion keywords corresponding to the user's task data are not included in the task detailed answer data, the task detailed answer data is mirrored. The introspection support apparatus according to claim 1, comprising: The storage device stores template data including agent start greeting data for introducing a counselor represented by the introspection support device, and agent experience data regarding the experience of the counselor,
The task specifying means includes
The agent start greeting data is displayed, the agent experience data is displayed, the user is inquired about the occupation, the occupation of the user is acquired, and then the user is inquired about the experience corresponding to the occupation. Item 3. The introspection support device according to item 1 or 2. A reflection support method for supporting user reflection,
The computer stores attribute data associating occupations with tasks corresponding to the occupations, and associating related keywords of the occupations and tasks,
The computer identifying a problem of the user;
The computer comprising the step of assisting the user in solving the identified problem of the user;
The step of identifying the task includes:
The computer obtaining the user's occupation;
The computer inquiring the user for an issue;
The computer obtaining task answer data input by the user in response to the task query;
The computer includes the step of acquiring the user's assignment based on the assignment answer data and a related keyword associated with the assignment corresponding to the user's occupation in the attribute data;
The step of supporting the solution by the user includes:
When the computer includes a related keyword associated with the user's task in the task response data, the computer mirrors the task response data and inquires about the details of the task;
The introspection support method, wherein the computer includes a step of mirroring the task answer data when the task answer data does not include the related keyword. The attribute data further associates an emotion keyword indicating an emotion for the task with the occupation and the task,
The step of supporting the solution by the user further includes:
The computer obtaining task detail answer data input by the user in response to an inquiry about the task details;
When the computer includes a related keyword or emotion keyword corresponding to the user's task data in the task detail response data, the computer mirrors the task detail response data and inquires about the details of further tasks;
The computer includes a step of mirroring the task detail response data when the task detail response data does not include a related keyword and an emotion keyword corresponding to the user task data. Introspection support method described in 1. The computer stores template data including agent start greeting data introducing a counselor represented by the computer, and agent experience data regarding the experience of the counselor,
The step of identifying the task includes:
The computer displaying the agent-initiated greeting data;
Displaying the agent experience data after the computer displays the agent start greeting data;
After the computer displays the agent experience data, inquiring the user for occupation;
The introspection support method according to claim 4, further comprising: a step of inquiring the user of an experience corresponding to the occupation after the computer acquires the occupation of the user. A reflection support program that supports user reflection,
The computer stores attribute data associating occupations with tasks corresponding to the occupations, and associating related keywords of the occupations and tasks,
In the computer,
Identifying the user problem;
A step of assisting the user in solving the identified problem of the user;
The step of identifying the task includes:
Obtaining the user's occupation;
Inquiring the user for an issue;
In response to the question inquiry, obtaining task answer data input by the user;
Obtaining the user's assignment based on the assignment answer data and a related keyword associated with the assignment corresponding to the user's occupation in the attribute data;
The step of supporting the solution by the user includes:
When the assignment answer data includes a related keyword associated with the user's assignment, mirroring the assignment answer data and inquiring about the details of the assignment;
A reflection support program comprising the step of mirroring the assignment answer data when the assignment answer data does not include the related keyword. The attribute data further associates an emotion keyword indicating an emotion for the task with the occupation and the task,
The step of supporting the solution by the user further includes:
In response to the inquiry about the details of the task, obtaining the task detail answer data input by the user;
When the related keyword or emotional keyword corresponding to the user's task data is included in the task detailed response data, the task detailed response data is mirrored and further details of the task are queried.
8. The method according to claim 7, further comprising: mirroring the task detail response data when the task detail response data does not include a related keyword and an emotion keyword corresponding to the user task data. 9. Ministry support program. The computer stores template data including agent start greeting data introducing a counselor represented by the computer, and agent experience data regarding the experience of the counselor,
The step of identifying the task includes:
Displaying the agent start greeting data;
Displaying the agent experience data after displaying the agent start greeting data;
After displaying the agent experience data, querying the user for occupation;
The introspection support program according to claim 7 or 8, further comprising the step of inquiring the user of an experience corresponding to the occupation after the computer acquires the occupation of the user. An interactive device that interacts with a user on behalf of a counselor,
A storage device for storing template data including agent start greeting data for introducing a counselor and agent experience data relating to the experience of the counselor;
A trust relationship construction means for displaying the agent start greeting data, displaying the agent experience data, inquiring the user for an occupation, obtaining the user's occupation, and inquiring an experience corresponding to the user's occupation; An interactive device characterized by that. A computer interaction method for interacting with a user on behalf of a counselor,
The computer stores template data including agent start greeting data introducing the counselor and agent experience data regarding the counselor's experience,
The computer displaying the agent-initiated greeting data;
Displaying the agent experience data after the computer displays the agent start greeting data;
After the computer displays the agent's experience data, inquiring the user for occupation;
After the said computer acquires the occupation of the said user, the step which inquires the said user for the experience corresponding to the said occupation is provided. The dialogue method characterized by the above-mentioned. An interactive program that interacts with users on behalf of a counselor,
The computer stores template data including agent start greeting data introducing the counselor and agent experience data regarding the counselor's experience,
In the computer,
Displaying the agent start greeting data;
Displaying the agent experience data after displaying the agent start greeting data;
Querying the user for occupation after displaying the agent experience data;
After acquiring the user's occupation, the interactive program is characterized by causing the user to inquire about an experience corresponding to the occupation.

Images