Question sentence generating method, question sentence generating apparatus, and question sentence generating program

Abstract

To provide a question sentence generation method, a question sentence generation device and a question sentence generation program capable of generating a question sentence conforming to a framework.SOLUTION: A question generation method comprising: a step of setting a question theme; a connecting phrase selection step of selecting a connecting phrase matching the set question theme; a question set selection step of selecting a question set matching the set question theme; and a question generation step of generating a question by combining the set question theme, the selected connecting phrase, and the selected question set.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to a question generation method, a question generation device, and a question generation program. [Background technology]

[0002] Launching new businesses is known as a way for companies to secure long-term advantages. Employee entrepreneurial spirit is considered important for launching and succeeding in new businesses. However, the emphasis on entrepreneurial spirit often overlooks whether the new businesses will generate profits. Large companies, in particular, often lack an internal environment conducive to launching new businesses, so the first step is to create an environment conducive to launching new businesses.

[0003] One method for analyzing past cases involves asking questions based on a specific framework to people who have experience with the case and analyzing the responses. In the example of creating an environment suitable for launching a new business mentioned above, questions based on a framework combining Herzberg's Motivation-Hygiene Theory (Herzberg theory) and the Job Demands-Resources Model (JDR model) could be asked of people who have experience launching new businesses. However, asking appropriate questions requires a thorough understanding of the framework, which is difficult to achieve. Furthermore, to obtain meaningful results, the questions must be carefully crafted, which takes time and effort.

[0004] One possible solution to this problem is to use a program that automatically generates questions. Patent Document 1 discloses an apparatus that generates questions using a language model. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Patent Publication No. 2017-027233 Summary of the Invention [Problem to be solved by the invention]

[0006] The question generation device disclosed in Patent Document 1 does not take frameworks into consideration, making it difficult to generate questions that conform to a framework. Furthermore, Patent Document 1 does not anticipate generating questions that conform to a framework that is composed of multiple questions, each of which contains different factors.

[0007] The present invention has been made in consideration of the above circumstances, and aims to provide a question generation method, a question generation device, and a question generation program that can generate questions that conform to a framework. [Means for solving the problem]

[0008] A question generation method according to an embodiment of the present invention includes: Steps to set a question theme, a conjunction phrase selection step of selecting a conjunction phrase that matches the set question sentence theme; a question set selection step of selecting a question set that matches the set question sentence theme; a question generation step of generating a question by combining the set question theme, the selected conjunction phrase, and the selected question set; Including, The connecting phrase selection step includes: outputting one or more of the stored plurality of conjunction phrases as conjunction phrase candidates; receiving an input to select a conjunction phrase that matches the set question sentence theme from the conjunction phrase candidates; and storing the set question theme and the selected connecting phrase in association with each other, The question set selection step includes: outputting one or more question sets from the stored plurality of question sets as candidate question sets; receiving an input to select a question set that matches the set question theme from the question set candidates; and storing the set question theme and the selected question set in association with each other.

[0009] In addition, in the question generation method according to one embodiment of the present invention, The question set may consist of one or more question parts, each of which includes a different factor.

[0010] In addition, in the question generation method according to one embodiment of the present invention, The question set is: A question part including an internal driving factor that combines an internal factor and a driving factor; a question part including an internal inhibiting factor that combines an internal factor and an inhibiting factor; a question part including an external driving factor that combines an external factor and a driving factor; and a question part including an external inhibiting factor that is a combination of an external factor and an inhibiting factor.

[0011] In addition, in the question generation method according to one embodiment of the present invention, The question set selection step further comprises: accepting input to modify wording of question parts included in the output question set; storing the question set including the modified question parts as a new question set; storing the set question theme and the revised question set in association with each other; may also include:

[0012] In addition, in the question generation method according to one embodiment of the present invention, In the question set selection step, when an input indicating that some question parts in the output question set are inappropriate is received, a question part candidate that is an alternative candidate for the inappropriate question part is output; When an input of an optimum question part candidate is received from the output question part candidates, the optimum question part candidate is replaced with the inappropriate question part to create a question set, which is stored as a new question set; and storing the set question theme and the replaced question set in association with each other; may also include:

[0013] In addition, in the question generation method according to one embodiment of the present invention, The method may further include a step of performing machine learning based on the results of the set question theme and the selected connecting phrase that are linked and stored.

[0014] In addition, in the question generation method according to one embodiment of the present invention, In the connecting phrase selection step, The order in which the conjunction phrase candidates are output may be determined to be highly likely to be linked to the set question theme based on the machine learning.

[0015] In addition, in the question generation method according to one embodiment of the present invention, The method may further include a step of performing machine learning based on the set question theme and the results of the selected question set that are linked and stored.

[0016] In addition, in the question generation method according to one embodiment of the present invention, In the question set selection step, The question set candidates may be output in order starting from those determined to be highly likely to be associated with the set question theme based on the machine learning.

[0017] In addition, in the question generation method according to one embodiment of the present invention, The question generated by the question generation method includes a situation setting phrase that sets a situation in which the question is asked by being linked to the question theme, The question generation method further includes: a situation setting phrase selection step of selecting a situation setting phrase that matches the set question sentence theme, The situation setting phrase selection step includes: outputting one or more situation setting phrases from the stored plurality of situation setting phrases as situation setting phrase candidates; receiving an input to select a situation setting phrase that matches the set question sentence theme from the situation setting phrase candidates; storing the set question theme and the selected situation setting phrase in association with each other; may also include:

[0018] Furthermore, the question generation device according to one embodiment of the present invention includes: A control unit and a storage unit are provided, The control unit Steps to set a question theme, a conjunction phrase selection step of selecting a conjunction phrase that matches the set question sentence theme; a question set selection step of selecting a question set that matches the set question sentence theme; a question generation step of generating a question by combining the set question theme, the selected conjunction phrase, and the selected question set; Perform a process including The connecting phrase selection step includes: outputting one or more conjunction phrases from among the plurality of conjunction phrases stored in the storage unit as conjunction phrase candidates; receiving an input to select a conjunction phrase that matches the set question sentence theme from the conjunction phrase candidates; and storing the set question theme and the selected conjunction phrase in the storage unit in association with each other, The question set selection step includes: outputting one or more question sets from the plurality of question sets stored in the storage unit as candidate question sets; receiving an input to select a question set that matches the set question theme from the question set candidates; and storing the set question theme and the selected question set in the storage unit in association with each other.

[0019] Furthermore, a question sentence program according to an embodiment of the present invention includes: On the computer, Steps to set a question theme, a conjunction phrase selection step of selecting a conjunction phrase that matches the set question sentence theme; a question set selection step of selecting a question set that matches the set question sentence theme; a question generation step of generating a question by combining the set question theme, the selected conjunction phrase, and the selected question set; Execute a process including The connecting phrase selection step includes: outputting one or more of the stored plurality of conjunction phrases as conjunction phrase candidates; receiving an input to select a conjunction phrase that matches the set question sentence theme from the conjunction phrase candidates; and storing the set question theme and the selected connecting phrase in association with each other, The question set selection step includes: outputting one or more question sets from the stored plurality of question sets as candidate question sets; receiving an input to select a question set that matches the set question theme from the question set candidates; and storing the set question theme and the selected question set in association with each other. [Effects of the Invention]

[0020] According to the question generation method, question generation device, and question generation program of the present disclosure, a question that conforms to a framework can be generated. [Brief explanation of the drawings]

[0021] [Figure 1] FIG. 1 is a block diagram illustrating a configuration of a question generation device according to an embodiment of the present disclosure. [Figure 2] FIG. 1 is a diagram illustrating an example of a question generated by a question generation device according to an embodiment of the present disclosure. [Figure 3] 10 is a flowchart illustrating an example of operation of the question generation device according to an embodiment of the present disclosure. [Figure 4] 10 is a flowchart illustrating an example of operation of the question generation device according to an embodiment of the present disclosure. [Figure 5] 10 is a flowchart illustrating an example of operation of the question generation device according to an embodiment of the present disclosure. [Figure 6] 10 is a flowchart illustrating an example of operation of the question generation device according to an embodiment of the present disclosure. [Figure 7] FIG. 1 is a diagram illustrating an example of a question generated by a question generation device according to an embodiment of the present disclosure. [Figure 8] 10 is a flowchart illustrating an example of operation of the question generation device according to an embodiment of the present disclosure. [Figure 9] 10 is a flowchart illustrating an example of operation of the question generation device according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0022] Hereinafter, an embodiment of the present disclosure will be described with reference to the drawings. In each drawing, parts having the same configuration or function are denoted by the same reference numerals. In the description of this embodiment, duplicated descriptions of the same parts may be omitted or simplified as appropriate.

[0023] The question generation device 1 is a device for generating question sentences that constitute structured descriptive questions. A structured descriptive question is a question composed of one or more question sentences using a specific framework and prompts respondents to respond in descriptive form. The purpose of structured descriptive questions is to carefully select the information obtained through the questions by designing each question to inquire about a specific factor and obtaining responses from a specific small number of respondents, thereby eliminating variability in answers and improving interpretability. Multiple-choice questions have the disadvantage of limiting the respondent's freedom and spontaneity. Open-ended questions also have the disadvantage of large variability, making them impossible to analyze and interpret. In contrast, employing structured descriptive questions can overcome the above-mentioned drawbacks of multiple-choice and open-ended questions. In one embodiment of the present invention, structured descriptive questions are explained using a framework that combines Herzberg's theory and the JDR model.

[0024] Herzberg's theory is a theory regarding the factors that cause job satisfaction and job dissatisfaction, and posits that motivational factors (promoting factors) that lead to job satisfaction and hygiene factors (hindrance factors) that lead to job dissatisfaction are different. In one embodiment of the present invention, this theory is applied to human behavior in general, and factors that motivate people to perform specific behavior are defined as promoting factors, and factors that cause people to avoid specific behavior are defined as hindrance factors.

[0025] The JDR model is a theory regarding factors that increase work engagement, and posits that individual psychological factors (individual resources, internal factors) and factors derived from the environment in which an individual finds themselves (work resources, external factors) interact with each other to increase work engagement. In one embodiment of the present invention, this theory is applied to human behavior in general, and factors derived from a specific individual themselves that relate to the decision to perform or avoid a specific behavior are considered internal factors, and factors derived from the environment surrounding that individual that relate to that decision are considered external factors.

[0026] The framework employed in one embodiment of the present invention consists of four questions, each containing four patterns of factors (internal facilitating factors, internal inhibiting factors, external facilitating factors, and external inhibiting factors) that combine either the facilitating factors or inhibiting factors defined in Herzberg's theory with either the internal factors or external factors defined in the JDR model.

[0027] 1 is a block diagram illustrating an example of a question generation device 1 according to an embodiment of the present disclosure. The question generation device 1 is, for example, a general-purpose computer such as a PC (Personal Computer), but may also be realized by a dedicated electronic device. The question generation device 1 may load a dedicated program to execute the processing of the present disclosure.

[0028] The question generation device 1 includes a control unit 10, a storage unit 20, an output unit 30, and an input unit 40.

[0029] The control unit 10 executes various processes related to the operation of the question generation device 1 and controls each unit of the question generation device 1. The control unit 10 includes at least one processor, at least one dedicated circuit, or a combination of these. The processor is a general-purpose processor such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit), or a dedicated processor specialized for a specific process. The control unit 10 may realize its control functions by executing a program stored in the storage unit 20.

[0030] The storage unit 20 includes at least one semiconductor memory, at least one magnetic memory, at least one optical memory, or a combination thereof. The storage unit 20 functions as, for example, a main storage device, an auxiliary storage device, or a cache memory. The storage unit 20 stores, for example, programs and data used in the operation of the question generation device 1, and data obtained by the operation of the question generation device 1.

[0031] The output unit 30 includes one or more output interfaces. The output interface is, for example, an external or built-in display that outputs information as an image or video, a speaker that outputs information as sound, or an interface for connecting to an external output device. The display is, for example, an LCD (Liquid Crystal Display) or an organic EL (Electro Luminescence) display. The output unit 30 outputs information obtained by the operation of the question generation device 1.

[0032] The input unit 40 includes one or more input interfaces. The input interface may be, for example, a physical key, a capacitive key, a pointing device, a touch screen integrated with a display, or a microphone that accepts voice input. The input interface may further include a camera that captures captured images or image codes, or an IC card reader. The input unit 40 accepts an operation to input information used in the operation of the question generation device 1 and sends the input information to the control unit 10.

[0033] The operation of the question generation device 1 is realized by the processor included in the control unit 10 executing a program stored in the storage unit 20. The program may be recorded on a computer-readable recording medium. Examples of the computer-readable recording medium include a magnetic recording device, an optical disk, a magneto-optical recording medium, or a semiconductor memory. The program is distributed in a state recorded on a portable recording medium such as a DVD or CD-ROM on which the program is recorded.

[0034] Next, the processing executed by the question generation device 1 will be described with reference to Fig. 2. Fig. 2 is a diagram showing an example of a question generated in an embodiment of the present disclosure. As shown in Fig. 2, a question generated in an embodiment of the present disclosure includes a question theme, a conjunction, and a question set. Furthermore, the question set includes one or more question parts.

[0035] (Question theme setting) The input unit 40 accepts input of a question theme from a user. In one embodiment of the present disclosure, the user refers to a person who intends to generate a question using the question generation device 1. The question theme is a word or phrase that is the subject of the question. Specifically, question themes such as "using a car sharing service" and "commercialization after launching a new business" can be used. The input unit 40 may accept input from the user in the form of free description. Upon acquiring the data accepted by the input unit 40, the control unit 10 sets the question theme.

[0036] (Selection of connecting phrases) When the control unit 10 sets a question theme, it outputs one or more connecting phrases stored in the storage unit 20 to the output unit 30 as connecting phrase candidates. A connecting phrase is a word, a collocation, or a phrase that connects the question theme and the question set. Examples of connecting phrases include, but are not limited to, "in," "about," "regarding," and "comes to mind." There is no limit to the number of connecting phrases that can be output as connecting phrase candidates. Furthermore, when multiple connecting phrase candidates are output, the order in which they are output may be in order of likelihood of matching with the question theme.

[0037] The input unit 40 accepts a selection from the user of a conjunction phrase that best suits the set question theme from among the output conjunction phrase candidates. The control unit 10 associates the conjunction phrase selected by the user with the question theme and stores them in the storage unit 20.

[0038] The control unit 10 may perform machine learning on the result of linking the conjunction phrases with the question theme. In this case, the conjunction phrase candidates may be output based on the result of the machine learning, and when output by the output unit 30, they may be output in the order determined by the machine learning to be most likely to match the question theme.

[0039] (Question Set Selection) After setting the connecting phrase, the control unit 10 outputs one or more question sets stored in the storage unit 20 to the output unit 30 as candidate question sets. A question set includes one or more question parts, as shown in FIG. 2 . A question part is a phrase or clause that can be combined with a question theme and a connecting phrase to form a question. Examples of question parts include, "What kind of internal environment, support, and atmosphere do you think are needed?", "What kind of internal environment and atmosphere do you think will be a barrier?", "What kind of people do you think are needed?", and "What kind of people do you think will be a barrier?". The number of question sets output as candidate question sets is not limited. When multiple candidate question sets are output, they may be output in order of likelihood of matching with the question theme.

[0040] The question parts that make up a question set may each contain different factors. For example, a question set may be composed of four question parts, each containing four patterns of factors (internal facilitators, internal impediments, external facilitators, and external impediments) that combine either the facilitators (motivators) or the impediments (hygiene factors) defined in Herzberg's theory with either the internal or external factors defined in the JDR model. For example, a question set may be composed of a question part that includes an internal facilitator, such as "What kind of people do you think are necessary?", a question part that includes an internal impediment, such as "What kind of people do you think will be a barrier?", a question part that includes an internal impediment, such as "What kind of internal environment, support, and atmosphere do you think are necessary?", a question part that includes an external facilitator, and a question part that includes an external impediment, such as "What kind of internal environment and atmosphere do you think will be a barrier?"

[0041] The input unit 40 accepts from the user a selection of a question set that best matches the set question theme from among the output question set candidates. If the control unit 10 accepts input from the user indicating that none of the output question set candidates are suitable, the control unit 10 may exclude previously output question sets and output question sets in descending order of their likelihood of matching the question theme. The control unit 10 associates the user-selected question set with the question theme and stores them in the storage unit 20.

[0042] The input unit 40 may receive from the user an evaluation of each question part included in the output candidate question set. Specifically, the user evaluates the question parts using three levels: A, B, and C. An A rating is received if the question part is suitable for the question theme; a B rating is received if the question part is inappropriate as is but suitable with some wording correction; and a C rating is received if the question part is inappropriate. If the question set includes a question part rated C, the control unit 10 determines that the question set is inappropriate as a question theme. If the question set includes a question part rated B, the control unit 10 outputs a screen to the output unit 30 for correcting the wording of the question part. When the control unit 10 receives the corrected wording of the question part via the input unit 40, the control unit 10 stores the question set including the corrected question part as a new question set in the storage unit 20.

[0043] When the control unit 10 receives an input from the user indicating that some question parts of the output candidate question set have received a C rating, the control unit 10 may output a phrase that can replace the question part that received the C rating as a candidate question part to the output unit 30 instead of determining the question set including the question part as inappropriate. The output phrase contains the same factors as the question part that received the C rating. The output phrase may be generated by the control unit 10 or may be input by the user in a free-form writing format. When the control unit 10 receives a selection of a suitable candidate question part or input of a phrase from the user, the control unit 10 replaces the question part that received the C rating with the candidate question part or phrase, and stores the question set including the question part as a new question set in the storage unit 20.

[0044] The control unit 10 may perform machine learning on the result of linking the question set with the question theme, and in this case, the candidate question set may be output based on the result of the machine learning.

[0045] (First Example) A first example of a question generation method of the question generation device 1 according to this embodiment will be described with reference to Figures 3 to 6. Figure 3 is a flowchart showing the process by which the question generation device 1 generates a question.

[0046] The input unit 40 accepts a question theme from the user in a free description format. When the control unit 10 acquires a wording as the question theme from the input unit 40, the control unit 10 sets the wording as the question theme (step S110).

[0047] Next, the control unit 10 selects a conjunction that matches the question theme (step S120). The operation of the question generation device 1 in step S120 will be specifically described below with reference to Fig. 4. Fig. 4 is a flowchart showing the process of the question generation device 1 selecting a conjunction phrase.

[0048] The control unit 10 outputs one or more conjunction phrases that are considered to fit the question theme set in step S110, from among the multiple conjunction phrases stored in the memory unit 20, to the output unit 30 as conjunction phrase candidates (step S121).

[0049] The input unit 40 receives an input from the user as to whether or not the outputted conjunction phrase candidates include a conjunction phrase that matches the theme of the question sentence (step S122).

[0050] If there is no matching conjunction phrase (step S122: No), the control unit 10 excludes the conjunction phrases that have already been output, and then performs the procedure of step S121 again.

[0051] If there is a matching conjunction phrase (step S122: Yes), the control unit 10 determines the conjunction phrase candidate input as the matching conjunction phrase as the conjunction phrase (step S123).

[0052] Thereafter, the control unit 10 associates the conjunction determined in step S123 with the question theme set in step S110 and stores it in the storage unit 20 (step S124).

[0053] Next, the control unit 10 selects a question set that matches the question theme (step S130). The question set is composed of four question parts, each of which includes four patterns of factors (internal facilitating factors, internal inhibiting factors, external facilitating factors, and external inhibiting factors) that combine Herzberg's theory and the JDR model. The operation of the question generation device 1 in step S130 will be specifically described below with reference to FIG. 5. FIG. 5 is a flowchart showing the process by which the question generation device 1 selects a question set.

[0054] The control unit 10 outputs one or more question sets that are considered to fit the question theme set in step S110 from among the multiple question sets stored in the memory unit 20 to the output unit 30 as question set candidates (step S131).

[0055] The input unit 40 receives an input from the user as to whether or not any of the output question set candidates is suitable for the question statement theme (step S132).

[0056] If there is no matching question set (step S132: No), the control unit 10 excludes the question sets that have already been output, and then performs the procedure of step S131 again.

[0057] If there is a matching question set (step S132: Yes), the control unit 10 determines the question set candidate input as the matching question set as the question set (step S133).

[0058] Thereafter, the control unit 10 associates the question set determined in step S133 with the question theme set in step S110 and stores the result in the storage unit 20 (step S134).

[0059] Step S130 can also be performed by the following procedure instead of the above-mentioned steps S131 to S134. Another example of the operation of the question generation device 1 in step S130 will be specifically described below with reference to Fig. 6. Fig. 6 is a flowchart showing another example of the process by which the question generation device 1 selects a question set.

[0060] The control unit 10 outputs one or more question sets that are considered to fit the question theme set in step S110 from among the multiple question sets stored in the memory unit 20 to the output unit 30 as question set candidates (step S201).

[0061] The input unit 40 receives an input of the result of the user's evaluation on three levels of A, B, and C as to whether the question parts included in the output question set candidate are suitable for the question theme (step S202).

[0062] The control unit 10 determines whether the evaluations received in step S202 include a C evaluation (step S203).

[0063] If any question part is evaluated as C (step S203: Yes), the control unit 10 determines that the question part evaluated as C is inappropriate in relation to the question theme, and accepts an alternative question part (step S204). As a method for accepting the alternative question part, the user may input the wording in a free description format, or the control unit 10 may output one or more alternative candidate words and accept the user's selection of the most appropriate wording.

[0064] If no C rating is included (step S203: No) or when step S204 is completed, the control unit 10 checks whether the ratings received in step S202 include a B rating (step S205).

[0065] If a B-rating is included (step S205: Yes), the control unit 10 determines that the question part with the B-rating is inappropriate in relation to the question theme, and accepts correction of the wording of the question part (step S206).

[0066] If no C rating is included (step S205: No) or when step S206 is completed, the control unit 10 determines the question set candidate made up of the current question parts as the question set (step S207).

[0067] Thereafter, the control unit 10 associates the question set determined in step S207 with the question theme set in step S110 and stores it in the storage unit 20 (step S208). Note that if step S204 or S206, or both steps, are executed, the control unit 10 stores the question set determined in step S207 as a new question set in the storage unit 20.

[0068] When step S130 is completed, the control unit 10 generates a question by combining the question theme set in step S110, the conjunction selected in step S120, and the question set selected in step S130 (step S140).

[0069] By executing the method for generating questions according to the first example, the question generation device 1 according to this embodiment can easily generate questions that take the framework into consideration, thereby increasing the time spent on analyzing answers given to questions that take the framework into consideration.

[0070] Answers obtained from questions using a framework combining Herzberg theory and the JDR model, as described in one embodiment of the present disclosure, can be used to create a co-occurrence network. A co-occurrence network is a graph in which lines connect words with similar occurrence patterns. Interpreting a co-occurrence network can quickly identify unexpected factors with a certain degree of validity for a specific topic based on a specific small number of responses. Interpretation of a co-occurrence network may be performed by focusing on subgraphs. A subgraph is a collection of nodes discovered by community detection that are clustered within the network. Interpretation of a co-occurrence network may also be performed based on triangulation in qualitative research. Triangulation refers to the use of multiple sources or procedures to repeatedly establish identifiable patterns.

[0071] According to one embodiment of the present disclosure, even for a topic for which related text such as papers or materials does not exist on the Internet or in public documents, text can be generated based on answers obtained from questions using the framework, and text mining can be performed to analyze the topic using techniques such as co-occurrence networks. For example, text can be generated even for topics for which text is not released outside the company due to confidentiality obligations, such as the launch of a new business, or topics such as a new service or new product currently under development that does not exist in the world.

[0072] (Second Example) A method for generating a question using a projective method by the question generation device 1 will be described with reference to Fig. 7. Projective method is a technique for eliciting difficult-to-express feelings, thoughts, opinions, etc. through something else or someone, such as a method of having a respondent respond by imagining what it would be like in a specific situation. Fig. 7 is a diagram showing an example of a question generated in one embodiment of the present disclosure.

[0073] In the second example, as shown in Figure 7, a situation setting sentence is placed before a sentence that includes a question theme, a connecting phrase, and a question set. A situation setting sentence sets up a specific situation and asks the respondent to answer by imagining what would happen if they were in that situation. An example of a situation setting sentence is, "Imagine using a meat substitute."

[0074] A situation setting sentence includes a question theme and a situation setting phrase. A situation setting phrase is a phrase or clause that, when combined with a question theme, sets up a specific situation and asks the respondent to respond by imagining what would happen if they were in that situation. Examples of situation setting phrases include "Imagine someone who uses ," "Imagine a case where ," and "Imagine what would happen if were used."

[0075] Next, the operation of the question generation device 1 in the second embodiment will be described with reference to Figures 8 and 9. Figure 8 is a flowchart showing the process by which the question generation device 1 generates a question using the projection method.

[0076] The input unit 40 accepts a question theme from the user in a free description format. When the control unit 10 acquires a wording as the question theme from the input unit 40, the control unit 10 sets the wording as the question theme (step S310).

[0077] Next, the control unit 10 selects a situation-setting phrase that matches the question theme (step S320). The operation of the question generation device 1 in step S320 will be specifically described below with reference to Fig. 9. Fig. 9 is a flowchart showing the process of the question generation device 1 selecting a situation-setting phrase.

[0078] The control unit 10 outputs one or more situation setting phrases that are considered to fit the question theme set in step S310 from among the multiple situation setting phrases stored in the memory unit 20 to the output unit 30 as situation setting phrase candidates (step S321).

[0079] The input unit 40 receives an input from the user as to whether or not the output situation setting phrase candidates include a situation setting phrase that matches the theme of the question sentence (step S322).

[0080] If there is no matching situation-defining phrase (step S322: No), the control unit 10 excludes the situation-defining phrases that have already been output, and then performs the procedure of step S321 again.

[0081] If there is a matching situation setting phrase (step S322: Yes), the control unit 10 determines the situation setting phrase candidate input as the matching situation setting phrase as the situation setting phrase (step S323).

[0082] Thereafter, the control unit 10 associates the situation-defining phrase determined in step S323 with the question theme set in step S310 and stores the phrase in the storage unit 20 (step S324).

[0083] Next, the control unit 10 selects a conjunction that matches the theme of the question sentence (step S330). The specific procedure is the same as steps S121 to S124 in the first embodiment.

[0084] Next, the control unit 10 selects a question set that matches the question theme (step S340). The specific procedure is the same as either steps S131 to S134 or steps S201 to S208 in the first embodiment.

[0085] When step S340 is completed, the control unit 10 generates a situation setting sentence from the question theme set in step S310 and the situation setting phrase selected in step S320. Thereafter, the control unit 10 generates a question by combining the situation setting sentence, the question theme set in step S310, the connecting phrase selected in step S120, and the question set selected in step S130 (step S350).

[0086] By executing the method for generating a question according to the second embodiment, the question generation device 1 according to this embodiment can easily generate a question that takes into account the framework and utilizes the projection method.

[0087] The present disclosure is not limited to the above-described embodiments. For example, multiple blocks shown in the block diagrams may be integrated, or one block may be divided. Instead of executing multiple steps shown in the flowcharts in chronological order as described, the steps may be executed in parallel or in a different order depending on the processing capabilities of the device executing each step, or as needed. Other modifications are possible within the scope of the present disclosure.

[0088] Furthermore, although the present disclosure has been described using the example of launching a new business, the scope of application of the present disclosure is not limited to this. For example, the present disclosure can be generally applied to themes for which there are no reference texts available on the Internet or in public documents, such as the development of new services or new products. [Explanation of symbols]

[0089] 1 Question generation device 10 Control Unit 20 Memory section 30 Output section 40 Input section

Claims

1. A question generation method executed by a question generation device, comprising: Steps to set a question theme, a conjunction phrase selection step of selecting a conjunction phrase that matches the set question sentence theme; a question set selection step of selecting a question set that matches the set question sentence theme; a question generation step of generating a question by combining the set question theme, the selected conjunction phrase, and the selected question set; Including, The connecting phrase selection step includes: outputting one or more of the stored plurality of conjunction phrases as conjunction phrase candidates; receiving an input to select a conjunction phrase that matches the set question sentence theme from the conjunction phrase candidates; and storing the set question theme and the selected connecting phrase in association with each other, The question set selection step includes: outputting one or more question sets from the stored plurality of question sets as candidate question sets; receiving an input to select a question set that matches the set question theme from the question set candidates; and storing the set question theme and the selected question set in association with each other.

2. The question set is composed of one or more question parts, each of which includes a different factor. The method for generating a question according to claim 1 .

3. The question set is: A question part including an internal driving factor that combines an internal factor and a driving factor; a question part including an internal inhibiting factor that combines an internal factor and an inhibiting factor; a question part including an external driving factor that combines an external factor and a driving factor; and a question part including an external inhibiting factor that combines an external factor and an inhibiting factor. The method for generating a question according to claim 2 .

4. The question set selection step further comprises: accepting input to modify wording of question parts included in the output question set; storing the question set including the modified question parts as a new question set; storing the set question theme and the revised question set in association with each other; The method of claim 2 , further comprising:

5. In the question set selection step, when an input indicating that some question parts in the output question set are inappropriate is received, a question part candidate that is an alternative candidate for the inappropriate question part is output; When an input of an optimum question part candidate is received from the output question part candidates, the optimum question part candidate is replaced with the inappropriate question part to create a question set, which is stored as a new question set; and storing the set question theme and the replaced question set in association with each other; The method of claim 2 , further comprising:

6. The method further includes a step of performing machine learning based on the set question theme and the selected connective phrase that are stored in association with each other. The method for generating a question according to claim 1 .

7. In the connecting phrase selection step, the output order of the conjunction phrase candidates is determined based on the machine learning to be highly likely to be linked to the set question sentence theme, and the conjunction phrase candidates are output in that order. The method for generating a question according to claim 6.

8. The method further includes a step of performing machine learning based on the set question theme and the results of the selected question set that are linked and stored. The method for generating a question according to claim 1 .

9. In the question set selection step, the output order of the question set candidates is determined based on the machine learning to be highly likely to be associated with the set question theme, and the output order is determined to be in ascending order of likelihood. The method for generating a question according to claim 8.

10. The question generated by the question generation method includes a situation setting phrase that sets a situation in which the question is asked by being linked to the question theme, The question generation method further includes: a situation setting phrase selection step of selecting a situation setting phrase that matches the set question sentence theme, The situation setting phrase selection step includes: outputting one or more situation setting phrases from the stored plurality of situation setting phrases as situation setting phrase candidates; receiving an input to select a situation setting phrase that matches the set question sentence theme from the situation setting phrase candidates; storing the set question theme and the selected situation setting phrase in association with each other; The method of claim 1 , further comprising:

11. A question generation device including a control unit and a storage unit, The control unit Steps to set a question theme, a conjunction phrase selection step of selecting a conjunction phrase that matches the set question sentence theme; a question set selection step of selecting a question set that matches the set question sentence theme; a question generation step of generating a question by combining the set question theme, the selected conjunction phrase, and the selected question set; Perform a process including The connecting phrase selection step includes: outputting one or more conjunction phrases from among the plurality of conjunction phrases stored in the storage unit as conjunction phrase candidates; receiving an input to select a conjunction phrase that matches the set question sentence theme from the conjunction phrase candidates; and storing the set question theme and the selected conjunction phrase in the storage unit in association with each other, The question set selection step includes: outputting one or more question sets from the plurality of question sets stored in the storage unit as candidate question sets; receiving an input to select a question set that matches the set question theme from the question set candidates; and storing the set question theme and the selected question set in the storage unit in association with each other.

12. On the computer, Steps to set a question theme, a conjunction phrase selection step of selecting a conjunction phrase that matches the set question sentence theme; a question set selection step of selecting a question set that matches the set question sentence theme; a question generation step of generating a question by combining the set question theme, the selected conjunction phrase, and the selected question set; Execute a process including The connecting phrase selection step includes: outputting one or more of the stored plurality of conjunction phrases as conjunction phrase candidates; receiving an input to select a conjunction phrase that matches the set question sentence theme from the conjunction phrase candidates; and storing the set question theme and the selected connecting phrase in association with each other, The question set selection step includes: outputting one or more question sets from the stored plurality of question sets as candidate question sets; receiving an input to select a question set that matches the set question theme from the question set candidates; and storing the set question theme and the selected question set in association with each other.

Images