Inference device, inference method, and inference program

Abstract

An inference device according to the present invention infers a skill, an attribute, and an answer motivation level of each contributor from posts in a chat log DB, and stores those in an attribute DB. Then, when the inference device has received a question from a questioner, the inference device extracts, from the chat log DB, a chat log which is highly relevant to the question. The inference device infers, from posts in the extracted chat log and attribute information of contributors, whether or not a contributor of the posts can answer the question from the questioner. Then, for the contributor (user) who has been inferred to be able to answer the question, the inference device ranks and presents the degree of suitableness as an answerer on the basis of the personal relationship with the questioner and the attribute (for example, the answer motivation level).

Description

Estimation device, estimation method, and estimation program

[0001] The present invention relates to an estimation device, an estimation method, and an estimation program for estimating an appropriate person to whom a user should ask a question about a problem that the user wants to solve.

[0002] Traditionally, information needed to carry out business operations (hereinafter referred to as "business information") has been obtained by searching documents scattered throughout the company or by searching through archives of frequently asked questions (FAQs). Recently, technologies have emerged that acquire business information using interactive document search AI that utilizes large-scale language models known as LLMs (Large Language Models). Furthermore, there are also technologies that estimate the knowledge held by chat participants (e.g., keyword centrality betweenness) and their relationships (e.g., intimacy networks) from chat data, and then use these estimation results to recommend people who are connected to the desired knowledge.

[0003] Llamaindex, [online], [Retrieved November 30, 2023], Internet, <URL: https: / / www.llamaindex.ai / > Pham Thanh Quang, Junichi Yamamoto, Study on Knowledge Management Systems Considering Human Relationships, 7F-02, 82nd National Convention of Information Processing Society of Japan, 2020

[0004] All of the above technologies treat information as explicit knowledge. However, in reality, much of this information is implicit knowledge, which is highly personal. As a result, there continue to be problems that cannot be solved without obtaining tacit knowledge from specific individuals. Furthermore, since information about who possesses tacit knowledge is also a type of tacit knowledge, it is difficult to find individuals who possess the tacit knowledge that can lead to problem solving.

[0005] Therefore, an object of the present invention is to solve the above-mentioned problems and to estimate an appropriate person to whom a user should ask a question.

[0006] In order to solve the above-mentioned problems, the present invention is characterized by comprising an attribute analysis unit that analyzes the skills and attributes of each user from each user's posts and stores the results of the analysis in an attribute information storage unit as attribute information for each user; an input receiving unit that receives input of a question from a questioner; an extraction unit that extracts a series of posts from a post information storage unit that stores the posts of each user, the series of posts having a relevance to the input question of a predetermined value or more; and an answerability estimation unit that, if the similarity between the content of the question and the content of the questions in the extracted series of posts is a predetermined value or more, estimates whether the poster can answer the question based on whether the poster ultimately solved the problem in the series of posts, and, if the similarity between the content of the question in the extracted series of posts and the content of the questions is less than a predetermined value, estimates whether the poster can answer the question based on the skills and attributes of the poster indicated in the attribute information.

[0007] According to the present invention, it is possible to estimate an appropriate person to whom a user should ask a question.

[0008] FIG. 1A is a diagram for explaining an overview of an estimation device. FIG. 1B is a diagram showing an example of estimation of users who own tacit knowledge by the estimation device. FIG. 2 is a diagram showing an example of the configuration of an estimation device. FIG. 3A is a diagram for explaining processing executed by an attribute analysis unit of the estimation device. FIG. 3B is a diagram showing an example of a chat log. FIG. 4 is a diagram for explaining processing executed by an extraction unit of the estimation device. FIG. 5 is a diagram for explaining processing executed by an answer possibility estimation unit of the estimation device. FIG. 6 is a diagram for explaining processing executed by a question destination estimation unit of the estimation device. FIG. 7 is a flowchart showing an example of processing procedures executed by the estimation device. FIG. 8 is a diagram for explaining an application example of the estimation device. FIG. 9 is a diagram for explaining an application example of the estimation device. FIG. 10 is a diagram showing a computer that executes a program.

[0009] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a description will be given of an embodiment of the present invention with reference to the drawings, but the present invention is not limited to the embodiment.

[0010] First, an outline of the estimating device of this embodiment will be described. The estimating device estimates appropriate question recipients in response to a question from a questioner and presents the estimation results (candidates for question recipients).

[0011] For example, for a user who has information they want that is not documented (tacit knowledge) and does not know who to ask when they are in trouble (tacit knowledge of the knowledge owner), the estimation device estimates and presents to them who they should ask to find an answer.

[0012] In the following description, an example will be described in which the estimating device estimates the destination of a user's question based on a post in a chat system, but the estimation may also be made based on a post other than a chat system.

[0013] First, as shown in FIG. 1A , for example, the estimation device estimates each poster’s skills (including, for example, knowledge, abilities, interests, etc.), attributes, and level of motivation to answer questions from chat logs in a chat log DB (database), and stores the estimation results in the attribute DB as poster (user) attribute information (see reference numeral 101) ((1)).

[0014] Thereafter, when the estimation device receives an input of a question from the user (e.g., questions 1 and 2 shown in FIG. 1A ), it extracts chat logs highly relevant to the question from the chat log DB ((2)). The estimation device extracts, for example, chat logs 1, 2, and 3 shown in FIG. 1B as chat logs highly relevant to questions 1 and 2.

[0015] Next, as shown in FIG. 1A, the estimation device estimates users who possess the tacit knowledge (users who can answer the question) from the extracted chat log posts ((3)).

[0016] For example, if there is a post in the extracted chat log that is highly similar to a user's question, the estimating device determines whether the problem asked in the post has been solved. If the estimating device determines that the problem asked in the post has been solved, it estimates that the poster who has answered the post is a "user who can answer the question."

[0017] On the other hand, if the estimation device determines that the problem asked in the post has not been resolved, it estimates that the poster who is answering the post is a “user who cannot answer the question.” Furthermore, if there is not a high degree of similarity between the content of the question asked in the extracted chat log post and the content of the question from the user, the estimation device also combines the attribute information of the poster estimated in (1) to estimate whether the poster can answer the question.

[0018] For example, in chat log 1 shown in FIG. 1B, User B is unable to solve the problem in in-house system A. Therefore, the estimation device estimates that "User B cannot answer questions 1 and 2." Furthermore, in chat log 2, User C is able to solve the password problem in in-house system A. Therefore, the estimation device estimates that "User C can answer question 1, but it is unclear whether he can answer question 2, so it is impossible to determine."

[0019] Furthermore, in chat log 3, User D did not solve the problem with ERROR2345 in in-house system A, but solved a problem with a different error code in the same system. Furthermore, User D's attribute information reveals that "User D independently answered advanced questions from multiple users, and is highly proficient with the system and proactive." Therefore, the estimation device estimates that "User D can answer questions 1 and 2."

[0020] Then, as shown in FIG. 1A, the estimation device weights posters estimated to be able to answer the user's question based on their personal relationships and attributes (e.g., their motivation to answer) ((4)).

[0021] For example, the estimation device ranks posters (UserC, UserD) estimated to be able to answer the question in order of ease with which the asker (UserX) can ask a question and likelihood of them actively answering the question (see reference numeral 102), based on their personal relationships with the asker (UserX), their motivation to answer (attributes of the posters indicated by reference numeral 101), etc. This allows the asker to know who to ask a question to and, among them, who should be given priority in asking questions.

[0022] [Configuration Example] Next, a configuration example of the estimation device 10 will be described with reference to Fig. 2. The estimation device 10 includes, for example, an input / output unit 11, a storage unit 12, and a control unit 13.

[0023] The input / output unit 11 is an interface that controls the input and output of various data. The input / output unit 11 receives, for example, a question input from a user. The input / output unit 11 also outputs an estimation result of who the user should ask the question to (candidates of appropriate people to whom the user should ask the question).

[0024] The storage unit 12 stores data, programs, etc. that are referenced when the control unit 13 executes various processes. The storage unit 12 is realized by a semiconductor memory element such as a random access memory (RAM) or a flash memory, or a storage device such as a hard disk or an optical disk.

[0025] For example, the storage unit 12 includes a chat log DB (posted information storage unit) and an attribute DB (attribute information storage unit) that accumulates attribute information of each user created by the control unit 13.

[0026] The control unit 13 is responsible for overall control of the estimation device 10. The functions of the control unit 13 are realized, for example, by a central processing unit (CPU) executing a program stored in the storage unit 12.

[0027] The control unit 13 includes, for example, an attribute analysis unit 131 , an input reception unit 132 , an extraction unit 133 , an answer possibility estimation unit 134 , and a question recipient estimation unit (question recipient output unit) 135 .

[0028] The attribute analysis unit 131 analyzes the skills and attributes of each user from the posts of each user and stores the analysis results in the attribute DB as attribute information for each user. For example, the attribute analysis unit 131 creates attribute information for each user from the posts of each user in the chat log DB and stores the information in the attribute DB.

[0029] For example, as shown in FIG. 3A, the attribute analysis unit 131 extracts main keywords from the chat post content in the chat log DB (see chat logs 1, 2, 3, and 4 shown in FIG. 3B), classifies the nature of the post (whether it is a question or a general post), the topic, etc., and analyzes the relationship between the keywords and the poster.

[0030] Then, through the above analysis, the attribute analysis unit 131 obtains analysis results such as the nature of the chat log post, keywords, topics, technical level, questioner, answerer, the answerer's proactiveness, whether or not the problem was ultimately successfully solved (results), etc. Then, the attribute analysis unit 131 creates attribute information of the poster from the analysis results of the post content and stores it in the attribute DB.

[0031] Returning to the description of Fig. 2, the input receiving unit 132 receives a question input from a user. The extraction unit 133 extracts, from the chat log DB, a series of posts whose relevance to the input user question is equal to or greater than a predetermined value.

[0032] For example, the estimation device 10 converts each piece of text in the chat system into a multidimensional vector in advance, as shown in Fig. 4, and stores the vectorized text in a chat log DB. For example, GPT (Generative Pretrained Transformer), BERT (Bidirectional Encoder Representations from Transformers), or the like is used to vectorize the text.

[0033] The extraction unit 133 then vectorizes the input Question 1 of User X and searches the chat log DB for logs that are semantically similar to Question 1 (logs with a short vector distance). As a result, the extraction unit 133 extracts, for example, a series of chat logs (see reference numeral 401) whose relevance to Question 1 is equal to or greater than a predetermined value.

[0034] Returning to the description of Fig. 2, the answer possibility estimation unit 134 estimates, based on the series of posts (chat logs) extracted by the extraction unit 133, whether the poster of the post can answer the user's question.

[0035] For example, the answerability estimation unit 134 estimates whether the poster of a series of chat logs can answer the user's question based on the natural language pairing the user's question with the series of chat logs and the attribute information of the chat poster.

[0036] For example, if the answerability estimation unit 134 determines that the similarity between a question in the extracted series of posts and the user's question is equal to or greater than a predetermined value, it estimates whether the poster can answer the user's question based on whether the poster was ultimately able to solve the problem in the series of posts.On the other hand, if the answerability estimation unit 134 determines that the similarity between a question in the extracted series of posts and the user's question is less than a predetermined value, it estimates whether the poster can answer the question based on the poster's skills and attributes indicated in the attribute information.

[0037] For example, the answer possibility estimation unit 134 estimates whether or not a poster of a series of chat logs can answer a user's question based on the following viewpoints.

[0038] - The similarity of the questions in the series of posts to the user's questions - Whether the problem was ultimately solved in the series of posts - The accuracy of the poster's answers in the series of posts (for example, how many exchanges it took to answer the question) - Who answered in the series of posts

[0039] For example, if the similarity between a question in a series of posts extracted by the extraction unit 133 and a user's question is equal to or greater than a predetermined value, the answerability estimation unit 134 estimates whether the poster can answer the user's question based on whether the problem was ultimately solved in the series of posts, who the poster who ultimately answered was, and whether the poster's answer was accurate.

[0040] For example, consider a case where the series of posts (logs) extracted by the extraction unit 133 are the posts indicated by reference numerals 501 and 502 in Fig. 5. In this case, the questions in these posts are similar to the user's question, "I don't know the password for in-house system A." Therefore, the answerability estimation unit 134 estimates whether the poster who answered the series of posts can answer the question based on whether they ultimately solved the problem.

[0041] For example, in the case of the series of posts indicated by reference numeral 501, User B, who answered the question, was ultimately unable to solve the problem. Therefore, the answerability estimation unit 134 estimates that "User B cannot answer." On the other hand, in the case of the series of posts indicated by reference numeral 502, User C, who answered the question, was ultimately able to solve the problem. Therefore, the answerability estimation unit 134 estimates that "User C can answer."

[0042] When the answer possibility estimation unit 134 estimates whether the poster is able to answer, it may take into consideration attribute information of the poster (for example, whether the poster is knowledgeable about the in-house system A, whether the poster is highly motivated to answer, etc.).

[0043] Also, consider a case where the series of posts (logs) extracted by the extraction unit 133 is the post indicated by the reference numeral 503. In this case, the questions in the series of posts are about "ERROR 1234 in in-house system A," and therefore, the similarity to the user's question "Error 2345 has appeared in in-house system A and I can't proceed" is considered to be low.

[0044] In such a case, the answerability estimation unit 134 estimates whether UserD can answer the question based on the attribute information of UserD who answered the question. For example, if the answerability estimation unit 134 obtains information from UserD's attribute information that "UserD is a former developer of System A, uses it daily, is knowledgeable about technical aspects, and is kind to people in trouble," the answerability estimation unit 134 estimates that "UserD can answer." Note that the answerability estimation unit 134 may also estimate, based on a series of posts by UserD in addition to the above attribute information of UserD, that "Although there is no history of ERROR2345, UserD has experience dealing with other errors. UserD's skills are also such that UserD can answer," etc.

[0045] Returning to the explanation of Fig. 2, the question recipient estimation unit 135 selects an appropriate question recipient from among the users estimated to be able to answer the question by the answer ability estimation unit 134. For example, from among the users estimated to be able to answer the question, the question recipient estimation unit 135 weights users to whom the asker is likely to ask questions and who are estimated to be likely to actively answer the question, and outputs the weighted users as question recipient candidates.

[0046] For example, based on the attribute information of each user who is estimated to be able to answer the question, the question recipient estimation unit 135 outputs users who have a close relationship with the questioner (e.g., User X) and who are highly proactive in answering the question as candidates for question recipients.

[0047] For example, the question recipient estimation unit 135 acquires the personal relationships between the users (e.g., User C and User D) and the questioner (e.g., User X) analyzed from the chat log, as shown in Fig. 6. The question recipient estimation unit 135 also acquires information on the department to which each user (User X, User C, User D) belongs from an external DB such as personnel data.

[0048] The question recipient estimation unit 135 then estimates the level of relationship between UserX and each user (UserC, UserD) from the acquired information. For example, the question recipient estimation unit 135 estimates that a user who has a close relationship with UserX and belongs to the same department as UserX is a user who has a strong relationship with UserX. The question recipient estimation unit 135 also acquires the level of answering motivation (proactivity) of each user (UserC, UserD) from the attribute information of each user (UserC, UserD).

[0049] Then, based on these estimation results and the acquired information, the question recipient estimation unit 135 weights users who are estimated to have a strong relationship with the questioner and who are likely to actively answer questions, and outputs them as candidates for question recipients.

[0050] For example, as shown in Figure 6, consider the case where User C has medium proactiveness, high relationship with User X, question 1: yes (can answer), question 2: no (cannot answer), and User D has medium proactiveness, medium relationship with User X, question 1: yes (can answer), question 2: yes (can answer).

[0051] In this case, the question recipient estimation unit 135 outputs information that the most suitable candidate for question 1 is User C, the next most suitable candidate is User D, and the candidate for question 2 is User D.

[0052] By having the question recipient estimation unit 135 output the above information, the questioner (UserX) can know to whom to ask a question and, among them, to whom the question should be given priority.

[0053] [Example of Processing Procedure] Next, an example of processing procedure executed by the estimation device 10 will be described with reference to Fig. 7. First, when the attribute analysis unit 131 of the estimation device 10 acquires a new chat post (S11), it extracts changes in the knowledge and skills of the user (poster) from the post (S12), and updates the attribute information of the user in the attribute DB based on the extracted changes (S13). The attribute analysis unit 131 executes the above process every time it acquires a new chat post or at predetermined intervals.

[0054] The input receiving unit 132 of the estimation device 10 receives a question from a questioner (S21). Thereafter, the extraction unit 133 extracts a plurality of chat logs (chat log groups) that are highly relevant to the question received in S21 (S22). Next, the answer availability estimation unit 134 selects one group from the chat log groups extracted in S22 (S23) and estimates whether the poster of the selected chat log group can answer the question (S24).

[0055] If the reply possibility estimation unit 134 estimates in S24 that the poster can reply (Yes in S24), it adds the chat log group to the reply possible list (S25). On the other hand, if the reply possibility estimation unit 134 estimates that the poster cannot reply (No in S24), it proceeds to S26.

[0056] If there are still chat log groups that have not been selected in S26 (Yes in S26), the process returns to S23. On the other hand, if all chat log groups have been selected (No in S26), the process proceeds to S34. The question destination estimation unit 135 then extracts attribute information of the posters of the chat log groups registered in the answerable list from the attribute DB (S34). Thereafter, the question destination estimation unit 135 determines and presents a question destination from among the posters of the chat log groups registered in the answerable list based on the attribute information extracted in S34 (S35).

[0057] By having the estimating device 10 execute the above process, the questioner can know to whom to ask a question.

[0058] [Application Example] Next, a description will be given of an application example of the estimating device 10. For example, the estimating device 10 may determine and output a possibility of solving a user's question based on a chat log in the chat log DB and attribute information of each user in the attribute DB.

[0059] For example, as shown in Fig. 8, the estimation device 10 analyzes the skills and knowledge of each user based on the chat logs in the chat log DB ((1)). For example, based on the chat log of User D, the estimation device 10 analyzes that "User D is a former developer of System A. He uses it daily and is knowledgeable about the technology, so he is quite knowledgeable. He is kind to people in need." The estimation device 10 then registers the analysis results in the attribute information of User D.

[0060] After that, when the estimation device 10 receives a question from a user, it extracts chat logs similar to the question from the chat log DB ((2)). Then, based on the extracted chat logs and the analysis results of the skills and knowledge of the chat poster registered in the attribute information, the estimation device 10 calculates the possibility that the poster can solve the content of the question.

[0061] For example, the estimating device 10 determines that "the solvability is 100%, and error 1234 has been addressed in the past" for question 1 shown in Fig. 8 and outputs the determination result. Also, the estimating device 10 determines that "the solvability is 70%, and error 2345 has never been addressed, but error 1234 has been addressed in the past. In addition, the system's proficiency is high" for question 2 and outputs the determination result.

[0062] In this way, the estimation device 10 can determine and output the possibility of solving the problem even if it is not exactly the same case as the user's question, based on the similarity with past cases and the user's skills and knowledge, etc.

[0063] However, the conventional technology does not take into account the willingness of each user to answer questions or the difficulty of the questions, so there is a possibility that the conventional technology may suggest people who cannot answer questions or who can answer but are not willing (unwilling) to answer questions.

[0064] However, the estimating device 10 suggests a person to whom a question should be asked while taking into consideration each user's technical proficiency, willingness to answer questions, relationships with other users, and whether or not the user can answer each question. Thus, for example, for a question with low expertise, the estimating device 10 can suggest a person who has a high relationship with the questioner (a close personal relationship) as a person to whom a question should be asked, and for a question with high expertise, the estimating device 10 can suggest a person who has a low relationship with the questioner but is willing to answer as a person to whom a question should be asked.

[0065] For example, if the attribute information of each user is not taken into consideration, as shown in Figure 9, questions may be directed to people who are less proactive or who have a weak relationship with the questioner, or questions may be concentrated on people with high technical proficiency.

[0066] However, by taking into account the attribute information of each user (UserB, UserC, UserD) like the estimation device 10, it is possible to suggest UserC, who has a "medium" level of proficiency but a "high" proactiveness in answering and a "high" relationship with the questioner, as a person to whom a question should be asked for Question 1 (a question with low expertise).Furthermore, for Question 2 (a question with high expertise), the estimation device 10 can suggest UserD, who has a "high" level of proficiency, a "medium" proactiveness in answering, and a "medium" relationship with the questioner, as a person to whom a question should be asked.

[0067] As a result, the estimation device 10 can suggest people who are highly proactive in answering questions and people who have a strong relationship with the questioner as question recipients. Generally, it is considered that the lower the difficulty of a question, the greater the number of people who can answer the question. Therefore, by having the estimation device 10 suggest people who can answer the question and have a strong relationship with the questioner as question recipients, it is possible to reduce the concentration of questions on people with high technical proficiency.

[0068] [System Configuration, etc.] The components of each unit shown in the figure are conceptual functional units and do not necessarily have to be physically configured as shown. In other words, the specific form of distribution and integration of each device is not limited to that shown, and all or part of them can be functionally or physically distributed and integrated in any unit depending on various loads, usage conditions, etc. Furthermore, all or any part of the processing functions performed by each device can be realized by a CPU and a program executed by the CPU, or can be realized as hardware using wired logic.

[0069] Furthermore, among the processes described in the above embodiments, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically using a known method.In addition, the information including the processing procedures, control procedures, specific names, various data and parameters shown in the above documents and drawings can be changed as desired unless otherwise specified.

[0070] [Program] The above-described estimation device 10 can be implemented by installing a program (estimation program) as package software or online software on a desired computer. For example, by executing the above-described program on an information processing device, the information processing device can function as the estimation device 10. The information processing device referred to here includes mobile communication terminals such as smartphones, mobile phones, and PHS (Personal Handyphone Systems), as well as terminals such as PDAs (Personal Digital Assistants).

[0071] 10 is a diagram showing an example of a computer that executes an estimation program. The computer 1000 includes, for example, a memory 1010 and a CPU 1020. The computer 1000 also includes a hard disk drive interface 1030, a disk drive interface 1040, a serial port interface 1050, a video adapter 1060, and a network interface 1070. These components are connected by a bus 1080.

[0072] The memory 1010 includes a read-only memory (ROM) 1011 and a random access memory (RAM) 1012. The ROM 1011 stores a boot program such as a basic input / output system (BIOS). The hard disk drive interface 1030 is connected to a hard disk drive 1090. The disk drive interface 1040 is connected to a disk drive 1100. A removable storage medium such as a magnetic disk or optical disk is inserted into the disk drive 1100. The serial port interface 1050 is connected to a mouse 1110 and a keyboard 1120, for example. The video adapter 1060 is connected to a display 1130, for example.

[0073] The hard disk drive 1090 stores, for example, an OS 1091, an application program 1092, a program module 1093, and program data 1094. That is, the programs that define the processes executed by the above-described estimation device 10 are implemented as program modules 1093 in which computer-executable code is written. The program modules 1093 are stored, for example, in the hard disk drive 1090. For example, the program modules 1093 for executing processes similar to those of the functional configuration of the estimation device 10 are stored in the hard disk drive 1090. Note that the hard disk drive 1090 may be replaced with an SSD (Solid State Drive).

[0074] Data used in the processing of the above-described embodiment is stored as program data 1094, for example, in the memory 1010 or the hard disk drive 1090. The CPU 1020 then reads the program module 1093 or the program data 1094 stored in the memory 1010 or the hard disk drive 1090 into the RAM 1012 as necessary and executes them.

[0075] The program module 1093 and program data 1094 may not necessarily be stored in the hard disk drive 1090, but may also be stored in a removable storage medium and read by the CPU 1020 via the disk drive 1100 or the like. Alternatively, the program module 1093 and program data 1094 may be stored in another computer connected via a network (such as a local area network (LAN) or a wide area network (WAN)). The program module 1093 and program data 1094 may then be read by the CPU 1020 from the other computer via the network interface 1070.

[0076] REFERENCE SIGNS LIST 10 Estimation device 11 Input / output unit 12 Storage unit 13 Control unit 131 Attribute analysis unit 132 Input reception unit 133 Extraction unit 134 Answer possibility estimation unit 135 Question recipient estimation unit

Claims

1. An attribute analysis unit that analyzes the skills and attributes of each user from the posts of each user and stores the results of the analysis in an attribute information storage unit as attribute information for each user; an input reception unit that receives an input of a question from a questioner; an extraction unit that extracts a series of posts from a post information storage unit that stores the posts of each user, the relevance of which to the input question is equal to or greater than a predetermined value; and an answerability estimation unit that estimates whether the poster can answer the question based on whether the poster finally solved the problem in the series of posts when the similarity between the content of the question and the content of the question in the extracted series of posts is equal to or greater than a predetermined value, and estimates whether the poster can answer the question based on the skills and attributes of the poster indicated in the attribute information when the similarity between the content of the question and the content of the question in the extracted series of posts is less than a predetermined value. A presumption device characterized by comprising:

2. The attribute information further includes the result of the analysis of the closeness of the human relationship with other users and the positivity of the answer to the question analyzed from the posts of each user. The estimation device further includes a question destination output unit that weights and outputs the user who is estimated to be able to answer the question and is estimated to have a close human relationship with the questioner and to answer the question positively based on the attribute information of the user. The estimation device according to claim 1, characterized by comprising:

3. A method of estimation executed by an estimation device, the method comprising: analyzing skills and attributes for each user from posts of each user, and storing the results of the analysis in an attribute information storage unit as attribute information for each user; receiving an input of a question from a questioner; extracting a series of posts having a relevance degree equal to or higher than a predetermined value with respect to the input question from a post information storage unit that stores the posts of each user; when a similarity between the content of the question and the content of the question in the extracted series of posts is equal to or higher than a predetermined value, estimating whether the poster can answer the question based on whether the poster finally solved the problem in the series of posts, and when the similarity between the content of the question and the content of the question in the extracted series of posts is less than a predetermined value, estimating whether the poster can answer the question based on the skills and attributes of the poster indicated by the attribute information.

4. An estimation program for causing a computer to execute: analyzing skills and attributes for each user from posts of each user, and storing the results of the analysis in an attribute information storage unit as attribute information for each user; receiving an input of a question from a questioner; extracting a series of posts having a relevance degree equal to or higher than a predetermined value with respect to the input question from a post information storage unit that stores the posts of each user; when a similarity between the content of the question and the content of the question in the extracted series of posts is equal to or higher than a predetermined value, estimating whether the poster can answer the question based on whether the poster finally solved the problem in the series of posts, and when the similarity between the content of the question and the content of the question in the extracted series of posts is less than a predetermined value, estimating whether the poster can answer the question based on the skills and attributes of the poster indicated by the attribute information.

Images