Consultation support control device, method and program

Abstract

One aspect of the present invention involves: acquiring first text information in which experience of each of a plurality of first users is reflected; creating, by parsing the acquired first text information into a plurality of character strings by a predetermined number of characters, index information that is for the first text information and that includes information indicating the appearance positions of the plurality of character strings; extracting, on the basis of the created index information, a frequently appearing word by combining a plurality of character strings having a co-occurrence relationship; and registering the extracted frequently appearing word in a morphological analysis dictionary. In this state, second text information in which the present situation of a second user is reflected is acquired, and terms including a special term are extracted from the second text information through a morphological analysis of the acquired second text information by using the dictionary.

Description

[Technical field]

[0001] One aspect of the present invention relates to a consultation support control device, method, and program having a function of pairing a person who is having trouble at work with an appropriate person to consult with. [Background technology]

[0002] In recent years, for example, due to the diversification of work styles and infectious disease prevention measures, an increasing number of companies are adopting remote work. In remote work, it is difficult for employees to have natural conversations such as casual conversations between employees compared to face-to-face work, and there are no opportunities to see or hear other employees talking to each other. For this reason, even if an employee has a problem at work, it becomes difficult for that employee to consult with other employees casually, and it is also difficult to grasp who has what knowledge, which may lead to a decline in work efficiency and quality.

[0003] Therefore, as disclosed in Non-Patent Document 1, for example, a technology has been proposed that uses morphological analysis to identify words related to work-related problems and words related to the work situation when the problem occurred from information handled by an employee on a terminal, and pairs the employee who entered the words related to the problem with a consultant who has experience in solving the problem. With this technology, for example, when an employee signals the occurrence of a problem with a button or the like, a consultant who has experience in solving the problem is automatically selected and notified to the employee. This makes it possible for the employee to promptly consult with the consultant about the problem. [Prior art documents] [Non-patent literature]

[0004] [Non-Patent Document 1] "Consultant Pairing Technology by Matching Work Experiences", NTT, R6-9, Tsukuba Forum 2020 ONLINE, October 29, 2020 - October 30, 2020 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the technology disclosed in Non-Patent Document 1 uses morphological analysis to search for terms related to problems. Morphological analysis generally performs syntax analysis on a target document and breaks it down into words that each have a meaning. For this reason, when a target document contains an unknown special term that has not been generalized, such as an in-house term, a coined word, or an abbreviation, the special term is broken down into generalized word units. Then, the broken down words are used as entry words to compare with, for example, the work experience information of other employees, or the frequency of use of the broken down words is calculated, and as a result, comparison with the special term or calculation of the frequency of use is not performed. For this reason, when an employee handles a special term such as the above-mentioned in-house term, it is not possible to extract the special term as a single coherent character string from the information handled in the work and compare it with the work experience information of other people without breaking it down.

[0006] This invention has been made in light of the above circumstances, and aims to provide a technique for extracting specialized terms from information describing the current situation without breaking them down, and for comparing them with other people's work experience information, or for searching other people's work experience information for information containing those specialized terms. [Means for solving the problem]

[0007] In order to solve the above problem, one aspect of the consultation support control device or consultation support control method according to the present invention is to obtain first text information reflecting the experiences of each of a plurality of first users, to break down the obtained first text information into a plurality of character strings for each predetermined number of characters, to create index information for the first text information including information representing the appearance positions of the plurality of character strings, to combine a plurality of character strings having a co-occurrence relationship based on the created index information to extract frequent words, and to register the extracted frequent words in a dictionary for morphological analysis. Then, in this state, obtain second text information reflecting the current situation of a second user, and to extract terms including special terms from the second text information by performing morphological analysis on the obtained second text information using the dictionary.

[0008] According to one aspect of the present invention, the experience information of a plurality of users is decomposed into a plurality of character strings for each unit character number to create index information including occurrence position information, and frequently occurring words consisting of a plurality of character strings having a co-occurrence relationship are extracted based on the created index information, and the extracted frequently occurring words are registered in a dictionary for morphological analysis. Therefore, by performing morphological analysis on information reflecting the user's current situation in this state using the dictionary, special terms are extracted from the information representing the current situation as a single coherent character string without being decomposed, and this makes it possible to compare it with other people's work experience information or to search for other people's work experience information that includes the special term. Effect of the Invention

[0009] In other words, according to one aspect of the present invention, it is possible to provide technology that can extract specialized terms from information describing the current situation without breaking them down, and compare them with other people's work experience information, or search for other people's work experience information that contains those specialized terms. [Brief description of the drawings]

[0010] [Figure 1]FIG. 1 is a diagram showing a schematic configuration of a consultation support system according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a block diagram showing an example of a hardware configuration of a server device having a function as a consultation support control device according to an embodiment of the present invention. [Diagram 3] FIG. 3 is a block diagram showing an example of a software configuration of a server device having a function as a consultation support control device according to an embodiment of the present invention. [Figure 4] FIG. 4 is a flowchart showing an example of a processing procedure and processing contents of the dictionary registration processing executed by the server device shown in FIG. [Diagram 5] FIG. 5 is a flowchart illustrating an example of a procedure and content of the pairing process executed by the server device illustrated in FIG. [Figure 6] FIG. 6 is a diagram showing an example of a process of decomposing a special term using N-gram in the dictionary creation process shown in FIG. [Figure 7] FIG. 7 is a diagram illustrating an example of the pairing process illustrated in FIG. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0012] [One embodiment] (Configuration example) (1) System FIG. 1 is a diagram showing an example of the overall configuration of a consultation support system according to an embodiment of the present invention. This system enables the transmission of information data via a network NW between a number of user terminals UT1 to UTn, each of which is used by an employee of a company, and a server device SV that functions as a consultation support control device.

[0013] The user terminals UT1 to UTn are, for example, personal computers, and in addition to normal information processing and communication functions, have a function for saving the contents displayed on the terminal screen as a terminal log, and an application program used in remote operations. By using this application program, for example, it is possible to hold online conferences with other user terminals UT1 to UTn and access the server device SV. Note that other types of user terminals UT1 to UTn, such as smartphones and tablet terminals, can also be used.

[0014] The network NW includes a wide area network with the Internet at its core, and an access network for accessing the wide area network. The access network may be, for example, a public communication network using wired or wireless communication, a local area network (LAN) using wired or wireless communication, or a cable television (CATV) network.

[0015] (2) Server device SV 2 and 3 are block diagrams showing examples of the hardware and software configurations of the server device SV, respectively.

[0016] The server device SV is, for example, a server computer arranged on the cloud or the Web, and includes a control unit 1 using a hardware processor such as a central processing unit (CPU). A storage unit having a program storage unit 2 and a data storage unit 3, and a communication I / F unit 4 are connected to the control unit 1 via a bus 5.

[0017] Under the control of the control unit 1, the communication I / F unit 4 transmits and receives information data to and from each of the user terminals UT1 to UTn using a communication protocol defined by the network NW.

[0018] The program storage unit 2 is configured, for example, by combining a non-volatile memory such as a HDD (Hard Disk Drive) or SSD (Solid State Drive) as a storage medium that can be written to and read from at any time, and a non-volatile memory such as a ROM (Read Only Memory), and stores application programs necessary for executing various control processes related to one embodiment of the present invention, in addition to middleware such as an OS (Operating System).

[0019] The data storage unit 3 is, for example, a combination of a non-volatile memory such as an HDD or SSD as a storage medium that can be written to and read from at any time, and a volatile memory such as a RAM (Random Access Memory), and includes, as storage areas according to one embodiment, a work experience information storage unit 31, an index information storage unit 32, a frequently used word storage unit 33, a dictionary storage unit 34, a problem information storage unit 35, and a pairing information storage unit 36.

[0020] The work experience information storage unit 31 is used for storing information (work experience information) indicating the work experience of a user, which is acquired as terminal logs from each of the multiple user terminals UT1 to UTn, in association with a user ID.

[0021] The index information storage unit 32 is used to store index information including a plurality of character strings obtained by decomposing the above-mentioned work experience information using N-grams, information indicating the positions at which the character strings appear, and an index assigned to each character string.

[0022] The frequent word storage unit 33 is used to store the frequent words extracted by the control unit 1 based on the above-mentioned multiple character strings.

[0023] The dictionary storage unit 34 stores a dictionary for morphological analysis.

[0024] The problem information storage unit 35 is used to store information about a problem and peripheral information about the work situation at the time the problem occurred, which are extracted from information that represents the current work situation handled by the user.

[0025] The pairing information storage unit 36 ​​is used to store information indicating the pairing results between a user who has encountered a problem and other users who are candidates for consultation.

[0026] The control unit 1 includes, as processing functions according to an embodiment of the present invention, a work experience information acquisition processing unit 11, an index information creation processing unit 12, a frequent word extraction processing unit 13, a dictionary management processing unit 14, a work situation information acquisition processing unit 15, a problem information extraction processing unit 16, a pairing processing unit 17, and a pairing result notification processing unit 18. All of these processing units 11 to 18 are realized by causing the hardware processor of the control unit 1 to execute application programs stored in the program storage unit 2.

[0027] The work experience information acquisition processing unit 11 acquires the terminal logs from each of the multiple user terminals UT1 to UTn via the communication I / F unit 4. Then, the acquired terminal logs are stored in the work experience information storage unit 31 in association with the user ID as work experience information of the user who uses the user terminal UT1 to UTn. The terminal logs to be acquired include text information handled during the user's work, such as chats, e-mail messages, and documents.

[0028] The index information creation processing unit 12 reads text information included in the business experience information from the business experience information storage unit 31, decomposes this text information into a plurality of character strings in units of N characters using N-grams, and creates index information with each decomposed character string as a headword. At this time, the index information creation processing unit 12 generates information indicating the appearance position of each decomposed character string and includes it in the index information. The index information creation processing unit 12 then stores each of the created index information in the index information storage unit 32.

[0029] The frequent word extraction processing unit 13 reads the index information from the index information storage unit 32. Then, based on each piece of index information that has been read, it extracts frequent words by connecting multiple character strings that appear consecutively and co-occur beyond a preset threshold, and performs processing to store the extracted frequent words in the frequent word storage unit 33.

[0030] The dictionary management processing unit 14 performs a process of additionally registering the frequently occurring words stored in the frequently occurring word storage unit 33 in a dictionary for morphological analysis stored in a dictionary storage unit .

[0031] The task status information acquisition processing unit 15 performs processing to acquire, via the communication I / F unit 4, text information input by a user from the user terminal UT1 to UTn during a task as information indicating the current task status of the user.

[0032] The problem information extraction processing unit 16 performs morphological analysis on the acquired text information based on the dictionary stored in the dictionary storage unit 34, thereby extracting terms including special terms. Examples of special terms include in-house terminology, technical terms, coined words, abbreviations, new words, and names of people that are not generalized. The problem information extraction processing unit 16 then determines whether the extracted term including the special term corresponds to information on the user's problem and peripheral information on the business situation when the problem occurred, and if it is information on the problem and peripheral information on the business situation when the problem occurred, extracts the term as problem information and stores it in the problem information storage unit 35. Note that it is preferable that the extraction target is both information on the user's problem and peripheral information about the problem, but it may be either one of them.

[0033] When the problem information is extracted from the information representing the current work status of the user who is working, the pairing processing unit 17 selects from among the other users a user who will be a candidate for the user who has encountered the problem. The selection of the candidate for the problem is performed, for example, by referring to data representing the handling status of the problem information or related information up to that point, which is created for each of the other users based on their work experience information. Then, the pairing processing unit 17 stores the selected candidate for the problem in the pairing information storage unit 36 ​​in association with the user ID of the user who has encountered the problem.

[0034] The pairing result notification processing unit 18 reads out information representing the candidate consultation destinations from the pairing information storage unit 36, and transmits the read out information on the candidate consultation destinations from the communication I / F unit 4 to the user terminal used by the user who encountered the problem.

[0035] (Example of operation) Next, an example of the operation of the server device SV configured as above will be described.

[0036] (1) Dictionary registration process Prior to pairing between users, the server device SV performs a process of extracting frequently occurring words from the work experience information of a plurality of users and registering the words in a dictionary for morphological analysis.

[0037] FIG. 4 is a flowchart showing the procedure and contents of the dictionary registration process executed by the control unit 1 of the server device SV.

[0038] (1-1) Acquisition of work experience information In step S10, the control unit 1 of the server device SV monitors the reception of terminal logs from each of the user terminals UT1 to UTn under the control of the business experience information acquisition processing unit 11. In this state, when a terminal log is received from the user terminals UT1 to UTn, in step S11, the business experience information acquisition processing unit 11 associates the text information included in the received terminal log with the user ID of the sender, and stores this in the business experience information storage unit 31 as the business experience information of each user.

[0039] (1-2) Index information creation process When the above process of acquiring the business experience information is completed, in step S12, the control unit 1 of the server device SV reads the text information of the business experience information from the business experience information storage unit 31 under the control of the index information creation processing unit 12, and decomposes this text information into a plurality of character strings in units of N characters using N-grams. Then, index information is created with each decomposed character string as a heading word.

[0040] For example, when 2-grams are used as N-grams, the index information creation processing unit 12 breaks down the text information into a plurality of character strings in units of two characters, and assigns an index to each of the broken down character strings.

[0041] Fig. 6 shows an example of the index information creation process when using this 2-gram. In this example, the word "consultant pairing technology is" is broken down into two characters, and indexes are assigned to each character string in sequence, such as "01", "02", .... Note that the value of N in N-gram is not limited to "2", and may be other values ​​such as "3".

[0042] When creating the index information, the index information creation processing unit 12 includes information indicating the appearance position of each decomposed character string in the index. Then, each character string to which the index is assigned is stored in the index information storage unit 32.

[0043] (1-3) Extraction of frequent words In step S13, the control unit 1 of the server device SV judges whether or not it is time to register the information in the dictionary. If it is not time to register the information, the process returns to step S10 and the process of acquiring the business experience information and creating the index information is repeated. In response to this, for example, when a predetermined amount of work experience information has been accumulated and it is time to register the information in the dictionary, the control unit 1 of the server device SV, under the control of the frequent word extraction processing unit 13, sequentially reads the accumulated work experience information of each user in step S14, and executes a process of extracting frequently occurring words based on each character string, for example as follows.

[0044] That is, for each character string decomposed by the index information creation processing unit 12, the frequent word extraction processing unit 13 calculates a value representing the degree of co-occurrence between character strings that appear adjacently (consecutively). For example, the frequent word extraction processing unit 13 calculates a value representing the degree of co-occurrence between character strings that appear adjacently (consecutively) for each character string decomposed from all the work experience information stored in the work experience information storage unit 31 by counting the number of times the corresponding character strings co-occur. The frequent word extraction processing unit 13 then compares the calculated number of times the co-occurrence occurs with a preset threshold, and if the number of times the co-occurrence occurs exceeds the threshold, it joins the adjacent character strings together to form one word. Thereafter, in the same manner, it calculates the number of times co-occurrence occurs between adjacent character strings in each of the character strings, and joins the character strings whose number of times the co-occurrence occurs exceeds the threshold to form one word.

[0045] The above process will be explained using the index information shown in Fig. 6 as an example. That is, the frequent word extraction processing unit 13 first finds the number of times that the character string "consultation" corresponding to index "01" and the character string "consultant" corresponding to index "02" co-occur, and if the number of times this co-occurrence exceeds a threshold, it connects these character strings to generate a single character string "consultant".

[0046] Next, the frequent word extraction processing unit 13 finds the number of times that the character string "consultant" co-occurs with the character string "person Pe" corresponding to the following index "03", and if the number of times that this co-occurrence occurs exceeds a threshold, it joins the character string "consultant" and "person Pe" to generate the character string "consultant Pe".

[0047] Thereafter, the frequent word extraction processing unit 13 similarly determines the number of times the concatenated character string and the following character string co-occur in index order, and if the number of times the co-occurrence exceeds a threshold, it sequentially concatenates the above character strings to generate a character string.

[0048] As a result, if the number of co-occurrences of adjacent character strings exceeds a threshold, all character strings from index "01" to "09" are connected in order to generate a "consultant pairing technology," which becomes a single frequently occurring word.

[0049] The frequent word extraction processing unit 13 stores the frequent words extracted as described above in the frequent word storage unit 33.

[0050] (1-4) Registering in the dictionary Next, in step S15, the control unit 1 of the server device SV, under the control of the dictionary management processing unit 14, reads out the frequent words stored in the frequent word storage unit 33, and additionally registers the read out frequent words in the dictionary for morphological analysis stored in the dictionary storage unit 34. As a result, for example, the "consultant pairing technology" is registered as an entry word in the dictionary.

[0051] (2) Extraction of problem information The server device SV performs a process of extracting information indicating a problem in business and peripheral information related to the business situation at the time the problem occurred, based on the terminal log sent from the user terminal UT1 to UTn for each user.

[0052] FIG. 5 is a flowchart showing the processing steps and contents of a process executed by the control unit 1 of the server device SV to extract problem information and peripheral information related to the work situation at the time the problem occurred, and a process using the results of the process to pair the user who has encountered the problem with other users who are candidates for consultation.

[0053] (2-1) Acquisition of business status information In step S20, the control unit 1 of the server device SV monitors the reception of terminal logs from each of the user terminals UT1 to UTn under the control of the business status information acquisition processing unit 15. In this state, when a terminal log is received from the user terminals UT1 to UTn, the business status information acquisition processing unit 15 passes the text information included in the received terminal log to the problem information extraction processing unit 16 as information representing the user's current business status.

[0054] The terminal log acquired as the information representing the business status is stored in the business experience information storage unit 31 by the business experience information acquisition processing unit 11 as information representing business experience.

[0055] (2-2) Extraction of problem information When the information representing the business situation is acquired, the control unit 1 of the server device SV, under the control of the problem information extraction processing unit 16, executes an extraction process of information representing the user's business problem and peripheral information related to the business situation when the problem occurred, as follows.

[0056] That is, in step S21, the problem information extraction processing unit 16 first performs morphological analysis on the acquired text information based on the dictionary stored in the dictionary storage unit 34, thereby breaking down the text information into words. At this time, words including special terms are registered in advance as entry words in the dictionary by the dictionary registration process described above. Therefore, even if the text information of the information expressing the business situation includes special terms that are difficult to break down by normal morphological analysis, such as in-house terms, technical terms, coined words, abbreviations, new words, and people's names, the special terms can be extracted with a high probability by the morphological analysis.

[0057] Next, in step S22, the problem information extraction processing unit 16 judges whether or not the extracted term including the special term corresponds to information on the user's problem, and if the term is information on the problem and peripheral information on the business situation when the problem occurred, extracts the term as problem information and stores it in the problem information storage unit 35. The judgment of whether or not the term corresponds to a problem is made based on, for example, the frequency of appearance of the term in the business experience information of the user who encountered the problem.

[0058] The problem information extraction processing unit 16 may transmit the extracted terms to the user terminals UT1 to UTn that sent them each time, display them, and determine whether the terms correspond to a problem by having the user respond as to whether the term is a problem.

[0059] (3) Pairing process Next, in step S23, the control unit 1 of the server device SV, under the control of the pairing processing unit 17, executes a process of pairing the user who has encountered a problem with other users who are candidates for consultation, as follows.

[0060] That is, the pairing processing unit 17 compares the vocabulary distribution including the problem information of the user who has encountered the problem with the vocabulary distributions created in advance for each of the other users based on their work experience information. Then, the other users who have a high frequency of the problem information of the user who has encountered the problem are regarded as users who have experience in solving the problem, and the other users are selected as candidates for consultation.

[0061] FIG. 7 is a diagram showing an example of the pairing process. In the diagram, Ua indicates the vocabulary distribution of the user who has encountered a problem, and Ub and Uc indicate the vocabulary distributions of other users. The vocabulary distribution indicates the distribution of terms extracted from the terminal log, and in FIG. 7, the frequency of occurrence of each term is indicated by shading. In this example, the frequency of occurrence is expressed by using diagonal lines, shading, and blank spaces in order from high to low shading. Also, E in FIG. 7 indicates the distribution range of terms extracted in the current business situation, and Ga, Gb, and Gc indicate the distribution range of terms that correspond to problems.

[0062] In the example shown in Fig. 7, the pairing processing unit 17 identifies a distribution range Ga of terms corresponding to the problem from a distribution range E of terms corresponding to the current business situation in the vocabulary distribution Ua of the user who has encountered a problem. Then, the pairing processing unit 17 compares the identified distribution range Ga of terms corresponding to the problem with distribution ranges Gb and Gc of terms corresponding to the corresponding problems in the vocabulary distributions Ub and Uc of other users, respectively. As a result of this comparison, among the other users, a user having a high concentration of the distribution ranges Gb and Gc of terms corresponding to the problem is regarded as a user who has a lot of experience in solving the same problem, and this user is selected as a candidate for consultation.

[0063] The pairing processing unit 17 stores information representing the selected users who are candidates for consultation in the pairing information storage unit 36 ​​in association with information representing the user who has encountered a problem.

[0064] (4) Pairing result notification When the pairing result is obtained, the control unit 1 of the server device SV, under the control of the pairing result notification processing unit 18, reads out the pairing result from the pairing information storage unit 36 ​​in step S24. Then, information indicating users who are candidates for consultation, which is indicated by the pairing result, is transmitted from the communication I / F unit 4 to the user terminal of the user who has encountered the problem.

[0065] Therefore, a user who has encountered a problem can quickly consult with the notified candidate users about the problem by contacting them via email or SNS, for example.

[0066] (Action and effect) As described above, in one embodiment, the business experience information of multiple users is decomposed into multiple character strings for each character number unit using N-gram to create index information including information on the appearance position of each character string, and frequently occurring words are extracted based on the created index information, and these frequently occurring words are additionally registered in a dictionary for morphological analysis.In this state, information representing the user's current business situation is obtained, and the information representing the business situation is decomposed into multiple words by morphological analysis using the dictionary, and information representing the user's trouble and peripheral information related to the business situation when the trouble occurred are extracted based on the decomposed words, and other users who are candidates for consultation are selected based on the information representing the trouble, and information related to the selected other users is notified to the user who has encountered the trouble.

[0067] Therefore, frequently occurring words including special terms are registered in advance as entry words in the dictionary for morphological analysis. Therefore, by performing morphological analysis on information reflecting the user's current situation using the dictionary, it becomes possible to extract the special terms from the information expressing the current situation as a single coherent character string without breaking them down, and to compare it with other people's work experience information, or to search for other people's work experience information that includes the special terms.

[0068] [Other embodiments] (1) The process of extracting frequent words does not necessarily have to use a method for determining the co-occurrence relationship of character strings decomposed by N-gram, and other methods such as mining character strings in frequent parts may also be used. This method is described in detail in, for example, the following literature: Yuta Tsuboi et al., "Mining frequent substrings," Institute of Electronics, Information and Communication Engineers, November 2003.

[0069] (2) In the above embodiment, a series of processing functions from acquiring business experience information to registering frequently used words in a dictionary, and from acquiring business information to pairing processing and notification processing of the results, are executed in the server device SV. However, the present invention is not limited to this, and the above series of processing functions may be provided in the user terminals UT1 to UTn.

[0070] (3) In the above embodiment, the present invention is described as being applied to a business entity such as a company. However, the present invention can also be applied to educational institutions such as schools and cram schools, seminars and exhibitions, and other situations where a student or participant needs to choose a person to consult with among other students or participants when they have a problem.

[0071] (4) In addition, the use, the type and configuration of the consultation support control device, the processing procedure and the processing content can be modified in various ways without departing from the gist of the present invention.

[0072] Although the embodiment of the present invention has been described in detail above, the above description is merely an example of the present invention in every respect. It goes without saying that various improvements and modifications can be made without departing from the scope of the present invention. In other words, in implementing the present invention, specific configurations according to the embodiment may be appropriately adopted.

[0073] In short, this invention is not limited to the above-mentioned embodiment as it is, and in the implementation stage, the components can be modified and embodied without departing from the gist of the invention. In addition, various inventions can be formed by appropriately combining multiple components disclosed in the above-mentioned embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, components from different embodiments may be appropriately combined. [Explanation of symbols]

[0074] SV: Server device UT1~UTn: User terminal NW…Network 1...Control section 2...Program memory section 3. Data storage section 4. Communication I / F section 5. Bus 11...Work experience information acquisition processing section 12...Index information creation processing unit 13...Frequent word extraction processing section 14...Dictionary management processing unit 15...Business status information acquisition processing unit 16...Problem information extraction processing section 17...Pairing processing unit 18... Pairing result notification processing unit 31...Work experience information storage section 32...Index information storage unit 33…Frequent word memory section 34…Dictionary memory section 35…Problem information storage section 36…Pairing information storage section

Claims

1. an acquisition processing unit that acquires first text information reflecting the experiences of each of a plurality of first users; an index information creation processing unit that creates index information for the first text information by breaking down the acquired first text information into a plurality of character strings for each predetermined number of characters, the index information including information indicating appearance positions of the plurality of character strings; a first extraction processing unit that extracts frequent words by combining the plurality of character strings having a co-occurrence relationship based on the created index information; a registration processing unit that registers the extracted frequent words in a dictionary for morphological analysis; a second extraction processing unit that acquires second text information reflecting a current situation of a second user, and extracts terms including special terms from the second text information by performing morphological analysis of the acquired second text information using the dictionary; A consultation assistance control device comprising:

2. a third extraction processing unit that extracts information related to the problem of the second user or peripheral information related to a business situation when the problem occurred based on the term extracted by the second extraction processing unit; a pairing processing unit that selects a candidate who will be a consultation partner for the second user from among the plurality of first users by referring to the information related to the trouble or the peripheral information that is generated in advance for each of the first users; The consultation assistance control device according to claim 1 , further comprising:

3. The consultation support control device according to claim 1, wherein the index information creation processing unit creates the index information of the first text information, the index information including information representing the appearance positions of the plurality of character strings, by decomposing the first text information into the plurality of character strings every N characters using N-grams.

4. The consultation support control device according to claim 1 , wherein the first extraction processing unit extracts the frequent words by combining the plurality of character strings that appear consecutively and co-occur more than a preset threshold value.

5. A consultation support control method executed by an information processing device, comprising: acquiring first text information reflecting the experience of each of a plurality of first users; creating index information for the first text information, the index information including information indicating appearance positions of the plurality of character strings by breaking down the first text information into a plurality of character strings each having a predetermined number of characters; extracting frequent words by combining the plurality of character strings having a co-occurrence relationship based on the created index information; a step of registering the extracted frequent words in a dictionary for morphological analysis; acquiring second text information reflecting a current situation of a second user, and extracting terms including special terms from the second text information by performing morphological analysis on the acquired second text information using the dictionary; A consultation support control method comprising the steps of:

6. 5. A program for causing a processor included in the consultation support control device to execute processing by each of the processing units included in the consultation support control device according to claim 1.

Images