JP2023008461A - Answer creation support program, answer creation support method and answer creation support device - Google Patents

Abstract

To provide an answer creation support program that supports answer rate improvement of an answer for a question while understanding intention of the question.SOLUTION: An answer creation support program performs processing of: acquiring a question; determining a noun for comparison that contrasts with a category of an answer given to the answer on the basis of an appearance frequency of the noun appearing in a question mark added sentence or an interrogative sentence equivalent when the question includes a question mark or is the interrogative sentence equivalent and cannot acquire the answer according to the question; identifying the category matching the noun for comparison as a question category of the question; narrowing down electronic mails on the basis of a comparison result between the question category and a specific word that is selected from answer sentences of the electronic mail to be associated with the electronic mail; presenting the narrowed down answer sentence of the electronic mail as the answer according to the question; and notifying a terminal of an answer creator of the narrowed down answer sentence of the electronic mail.SELECTED DRAWING: Figure 14

Description

The present invention relates to an answer creation support program, an answer creation support method, and an answer creation support device.

A chatbot is known as a system that automatically responds to inquiries (see Patent Document 1, for example). In addition, when creating FAQ (Frequently Asked Questions) to support automatic responses from dialogue logs, Q candidates (question candidates) and A candidates (answer candidates) to be added to FAQ are selected from a huge amount of dialogue logs. , techniques for efficient extraction using supervised learning have been proposed. As a result, the creation of FAQs is labor-saving (see Patent Document 2, for example).

In addition, there is also known a system in which a question from an inquirer is input as a chat message, and an answer creator on the site provider side manually inputs an answer to the question as a chat message. In this system, candidate answers are extracted and presented to the answer creator using a trained model generated by machine learning using combinations of past questions and answers as teacher data. This reduces the burden on the answer creator while maintaining the quality of the content of the answer (see Patent Document 3, for example).

JP 2019-139746 A JP 2020-071690 A JP 2020-071679 A

By the way, after the introduction of the chatbot, the answers are maintained for the purpose of improving the answer rate of questions and maintaining the answer rate above a certain level. Preparation of answers is often done for frequently asked questions. The frequency can be classified, for example, by performing ABC analysis on various questions.

However, there is not always a correlation between frequently asked questions and the questioner's intentions. Therefore, even if answers are prepared for frequently asked questions, answers that reflect the questioner's intentions are not necessarily prepared. Therefore, it is required to improve the answer rate to the question while understanding the intention of the questioner's question.

Therefore, in one aspect, it is an object to provide an answer creation support program, an answer creation support method, and an answer creation support device that help improve the response rate of answers to questions while grasping the intent of the question.

In one embodiment, the answer creation support program obtains a question, and if the question contains a question mark or is equivalent to an interrogative sentence, and if an answer according to the question cannot be obtained, the sentence with the question mark or the Based on the frequency of appearance of nouns appearing in interrogative sentences, a comparative noun to be compared with the category of the answer assigned to the answer is determined, and the category matching the comparative noun is set as the question category of the question. identifying and narrowing down the e-mail based on a comparison result between the question category and specific words selected from the e-mail answer text and associated with the e-mail; and is presented as the answer corresponding to the question, and the answer text of the narrowed down e-mail is notified to the terminal of the answer creator.

It is possible to support the improvement of the response rate of the answer to the question while grasping the intention of the question.

FIG. 1 is a diagram for explaining an example of a dialogue system. FIG. 2 shows an example of the hardware configuration of the dialog server. FIG. 3 is an example of a block diagram of a dialogue server. FIG. 4 is an example of a scenario DB. FIG. 5(a) is an example of a question screen. FIG. 5(b) is another example of the question screen. FIG. 6 is an example of a mail DB. FIG. 7 is an example of a question DB. FIG. 8 is a flowchart showing an example of dialogue-related processing. FIG. 9 is an example of an answer DB. FIG. 10 is a flow chart showing an example of the first narrowing down process. FIG. 11 is an example of the category table DB. FIG. 12 is a flow chart showing an example of the second narrowing down process. FIG. 13 is a flowchart showing an example of notification processing. FIG. 14 is an example of an answer preparation support screen.

Hereinafter, the form for carrying out this case will be described with reference to the drawings.

As shown in FIG. 1, the dialogue system ST includes a user terminal 100, an administrator terminal 150, and a dialogue server 200. FIG. Dialogue server 200 is an example of an answer preparation support device. Although one user terminal 100 and one administrator terminal 150 are shown in FIG. 1, one or both of the user terminal 100 and administrator terminal 150 may be plural. Also, although FIG. 1 shows a PC (Personal Computer) as an example of the user terminal 100, it may be a smart device such as a smart phone or a tablet terminal. The administrator terminal 150 is also similar to the user terminal 100 .

User terminal 100 includes input device 110 , display device 120 , and control device 130 . Both the input device 110 and the display device 120 are connected to the control device 130 . When a user (for example, a questioner) using the dialogue system ST operates the input device 110, the control device 130 controls the display content of the display device 120 based on instructions according to the operation content. For example, when the user operates the input device 110 to access the dialogue server 200 , the control device 130 displays the question screen SC on the display device 120 based on the processing of the dialogue server 200 .

In addition, although the details will be described later, when the user operates the input device 110 to input and transmit a question in sentences such as natural sentences on the question screen SC, the control device 130 asks the input question. Display within the screen SC. Instead of a sentence, a sentence ending with a word such as a noun or a pronoun may be input and transmitted as a question. Control device 130 transmits the input question to dialogue server 200 . When the user operates the input device 110 to select a predetermined selection button appearing on the question screen SC, the control device 130 transmits options corresponding to the selection button to the dialogue server 200 . In addition to the selection buttons, the question screen SC includes character strings that cannot be selected.

The administrator terminal 150 is operated by an administrator who manages the operation of the interaction server 200 . Since the configuration of the administrator terminal 150 is basically the same as that of the user terminal 100, detailed description thereof will be omitted. The administrator, as an answer creator, operates administrator terminal 150 to create FAQs for supporting automatic responses and add them to dialogue server 200 . Specifically, guidance questions (for example, questions and answers) and answer candidates to be added to the FAQ are created from a huge amount of dialogue logs between the questioner and the dialogue server 200 and registered in the dialogue server 200 . Further, when the administrator receives an inquiry by e-mail via the dialogue server 200, the administrator replies with an e-mail reply to the inquiry via the dialogue server 200. FIG.

The user terminal 100, administrator terminal 150 and dialogue server 200 are connected to each other via a communication network NW. As the communication network NW, for example, there is a LAN (Local Area Network). In this embodiment, the operation form of the interaction server 200 is described as being on-premise, but the operation form of the interaction server 200 may be cloud. A cloud may be a public cloud or a private cloud. In this case, the Internet may be used as the communication network NW.

When the dialogue server 200 receives a question or an option transmitted from the user terminal 100, it executes various processes. For example, the dialogue server 200 determines questions and options, and returns screen information including answers and guiding questions according to the questions and options to the user terminal 100 . Depending on the option, the dialogue server 200 returns screen information requesting an inquiry by e-mail to the user terminal 100 . This allows the user to make an inquiry by e-mail. In this way, the dialogue server 200 returns guidance questions and answers according to the user's questions and options. Therefore, the dialog system ST or the dialog server 200 may be called a chatbot that can automatically answer questions. Further, the dialog system ST or the dialog server 200 may be called a scenario type chatbot that interacts with a user according to a predetermined scenario regarding a specific topic.

Next, with reference to FIG. 2, the hardware configuration of the dialogue server 200 that executes the answer preparation support method will be described. Since the control unit 130 and the control unit of the administrator terminal 150 described above have basically the same hardware configuration as that of the dialogue server 200, the description thereof will be omitted.

As shown in FIG. 2, the dialogue server 200 includes a CPU (Central Processing Unit) 200A as a processor, and RAM (Random Access Memory) 200B and ROM (Read Only Memory) 200C as memories. The dialogue server 200 also includes a network I/F (interface) 200D and a HDD (Hard Disk Drive) 200E. An SSD (Solid State Drive) may be employed instead of the HDD 200E.

Furthermore, the dialog server 200 may include at least one of an input I/F 200F, an output I/F 200G, an input/output I/F 200H, and a drive device 200I. The CPU 200A to the drive device 200I are interconnected by an internal bus 200J. That is, the dialogue server 200 can be implemented by a computer.

An input device 710 is connected to the input I/F 200F. The input device 710 includes, for example, a keyboard and a mouse. Note that the input device 110 described above is similar to the input device 710 . A display device 720 is connected to the output I/F 200G. The display device 720 is, for example, a liquid crystal display. Note that the display device 120 described above is similar to the display device 720 . A semiconductor memory 730 is connected to the input/output I/F 200H. Examples of the semiconductor memory 730 include USB (Universal Serial Bus) memory and flash memory. The input/output I/F 200H reads the answer creation support program stored in the semiconductor memory 730 . The input I/F 200F and the input/output I/F 200H are provided with USB ports, for example. The output I/F 200G has, for example, a display port.

A portable recording medium 740 is inserted into the drive device 200I. Examples of the portable recording medium 740 include removable discs such as CD (Compact Disc)-ROM and DVD (Digital Versatile Disc). Drive device 200I reads the answer preparation support program recorded in portable recording medium 740 . The network I/F 200D has, for example, a LAN port and a communication circuit. Network I/F 200D is connected to communication network NW.

In the RAM 200B described above, the CPU 200A temporarily stores the answer preparation support program stored in the ROM 200C, the HDD 200E, and the semiconductor memory 730. FIG. The CPU 200A temporarily stores the answer preparation support program recorded in the portable recording medium 740 in the RAM 200B. As the CPU 200A executes the stored answer preparation support program, the CPU 200A realizes various functions described later and also executes various processes described later. Incidentally, the answer preparation support program may be adapted to correspond to the flow chart described later.

Next, the functional configuration of the interaction server 200 will be described with reference to FIG. Note that FIG. 3 shows the essential functions of the interaction server 200 .

As shown in FIG. 3, the dialogue server 200 comprises a storage section 210, a processing section 220, and a communication section 230. FIG. The storage unit 210 can be realized by one or both of the RAM 200B and HDD 200E described above. The processing unit 220 can be realized by the CPU 200A described above. The communication unit 230 can be implemented by the network I/F 200D described above. Therefore, the storage unit 210, the processing unit 220, and the communication unit 230 are connected to each other.

Here, the storage unit 210 includes a scenario DB (Database) 211, a question DB 212, a mail DB 213, a category table DB 214, and an answer DB 215. On the other hand, the processing section 220 includes a dialogue processing section 221 , an acquisition section 222 , a noun determination section 223 and a question category identification section 224 . The processing section 220 also includes a narrowing section 225 , an answer presentation section 226 and a notification section 227 .

For example, the acquisition unit 222 accesses the question DB 212 and acquires question information. Although the details will be described later, the question DB 212 stores question information input to the question screen SC and sent from the user terminal 100 . Also, for example, the answer presenting unit 226 accesses the mail DB 213 and acquires an e-mail. Although the details will be described later, the mail DB 213 stores e-mails that contain only inquiries sent by users using e-mails. In other words, the mail DB 213 stores e-mails before replies by the administrator. The mail DB 213 also stores e-mails containing both this inquiry and the administrator's response to this inquiry. In other words, the mail DB 213 also stores e-mails after responses by the administrator.

Here, the timing at which the acquisition unit 222 acquires the question information and the timing at which the answer presentation unit 226 acquires the e-mail are different. For example, the acquisition unit 222 monitors the question DB 212, and when the dialogue processing unit 221 stores question information in the question DB 212, the acquisition unit 222 immediately acquires the question information. That is, the acquisition unit 222 can acquire question information in real time. On the other hand, although the details will be described later, the answer presentation unit 226 acquires the e-mail after the narrowing down unit 225 has narrowed down the e-mails. That is, unlike the acquisition of question information, the answer presenting unit 226 does not acquire e-mails in real time, and there is a time lag between acquisition of question information.

Further, while the acquisition unit 222 acquires the question information received by the dialogue processing unit 221 during the first period, the answer presenting unit 226 receives the e-mail received by the dialogue processing unit 221 during the second period before the first period. get. That is, the answer presenting unit 226 acquires the e-mail received by the interactive processing unit 221 during the second period past the first period when the interactive processing unit 221 received the question information. Details of the remaining DB elements such as the scenario DB 211 included in the storage unit 210 and the remaining functional units such as the noun determination unit 223 included in the processing unit 220 will be described later.

Next, operations of the interaction server 200 will be described with reference to FIGS. 4 to 14. FIG.

First, when the user operates the input device 110 to access the dialogue server 200, the dialogue processing unit 221 refers to the scenario DB 211 based on the access and acquires the first guidance question information. Here, as shown in FIG. 4, the scenario DB 211 stores a plurality of guiding question information 11, 12, 13, 14 in which the order of guiding questions is defined as scenarios. In FIG. 4, the order of the guiding questions is defined by the guiding ID "Y001" etc. assigned to each of the guiding question information 11, 12, 13 and 14. In FIG. The guidance ID is an identifier that identifies guidance question information. In the present embodiment, the leading question information 11 including the leading question "What is your question..." corresponds to the first leading question information.

When the initial guidance question information is acquired, the dialogue processing unit 221 transmits the screen information of the question screen SC including the guidance question information to the control device 130 . When receiving the screen information, the control device 130 displays the question screen SC on the display device 120 based on the received screen information. As a result, the question screen SC appears on the display device 120, as shown in FIG. 5(a). The question screen SC includes a guide sentence 20 that guides the user on the operation method of the question screen SC, a plurality of selection buttons 21, 22, and 23, an input field 24 for inputting a question, a send button 25, and the like.

When the user operates the input device 110 and presses the selection button 22 with the pointer PT, for example, the control device 130 transmits the option “product system” associated with the selection button 22 to the dialogue server 200 . As a result, the dialogue processing unit 221 acquires the next guiding question information 12 following the guiding question information 11 from the scenario DB 211 and transmits the screen information of the next question screen SC including the guiding question information 12 to the control device 130 .

Further, for example, when the selection button 23 is pressed, the control device 130 displays a predetermined screen (not shown) on the display device 120 for supporting creation of an e-mail and transmission/reception of the e-mail. This predetermined screen contains the mail address of the user terminal 100 representing the source of the e-mail. This predetermined screen also includes the mail address of the dialogue server 200 as the destination of the electronic mail. This allows the user to send an inquiry to the interaction server 200 by e-mail. When the user operates input device 110 to instruct transmission of an e-mail containing an inquiry, control device 130 transmits the e-mail to dialogue server 200 . As a result, the interactive processing unit 221 receives the e-mail. Upon receiving the e-mail, the interactive processing unit 221 stores the e-mail in the mail DB 213 and transfers the received e-mail to the administrator terminal 150 . This allows the administrator to create an answer sentence to the inquiry.

When the administrator operates the input device of administrator terminal 150 to instruct transmission of an e-mail containing a reply, the control unit of administrator terminal 150 transmits this e-mail to dialogue server 200 . As a result, the interactive processing unit 221 receives the e-mail. Upon receiving the e-mail, interactive processing unit 221 stores the e-mail in mail DB 213 . As a result, as shown in FIG. 6, the mail DB 213 stores e-mails 31 and the like including both the inquiry and the answer. The e-mail 31 is associated with a keyword selected based on a known technique such as TF-IDF (Term Frequency-Inverse Document Frequency) for evaluating the degree of importance of words. The keyword is hereinafter referred to as an email selection keyword. In addition to the e-mail 31, the e-mail DB 213 stores a plurality of e-mails 32 and 33 each including an inquiry alone. Upon receiving the e-mail containing the reply text, the dialogue processing unit 221 transmits the received e-mail to the mail address of the user terminal 100 . This allows the user to confirm the answer to the inquiry.

By the way, as shown in FIG. 5A, when the user operates the input device 110 to input a question in the input field 24 and presses the send button 25, the control device 130 sends the question input in the input field 24. The question information included is transmitted to the dialogue server 200 . In this embodiment, the control device 130 transmits question information including the question "I want to know about product A. How much does product A cost?" Upon receiving the question information, the dialogue processing unit 221 adds a question ID for identifying the question to the received question information, and stores it in the question DB 212 together with the date and time of reception. Thereby, as shown in FIG. 7, question DB212 memorize|stores question information.

As described above, the acquisition unit 222 monitors the question DB 212, and when the dialogue processing unit 221 stores question information in the question DB 212, acquires the question as shown in FIG. 8 (step S1). For example, as shown in FIG. 7, if a plurality of pieces of question information with different reception dates are stored in the question DB 212, the acquisition unit 222 acquires the question with the latest reception date. In this embodiment, the acquisition unit 222 acquires the question "I want to know about product A. How much does product A cost?"

When the acquisition unit 222 acquires the question, as shown in FIG. 8, the interaction processing unit 221 determines whether or not the question can be answered (step S2). More specifically, the interactive processing unit 221 refers to the answer DB 215 based on the question and determines whether or not the question can be answered. As shown in FIG. 9, the answer DB 215 stores, in addition to the answers corresponding to the questions, answer information including a plurality of transition states combining answer IDs for identifying answers, guidance IDs, and options in association with each other. ing. Also, each answer is assigned a category of the answer.

The dialogue processing unit 221 refers to the answer DB 215, and if it determines that there is an answer corresponding to the acquired question, it determines that the answer is possible (step S2: YES). In this case, the dialogue processing unit 221 terminates the dialogue-related processing. On the other hand, if it is determined that there is no answer corresponding to the acquired question, it is determined that the answer is impossible (step S2: NO). In this case, the interactive processing unit 221 determines whether or not the number of determinations as to whether an answer is possible is equal to or greater than a predetermined number (step S3). The predetermined number of times represents a threshold for this determination, and can be several times such as 3 to 5 times, for example.

If the number of determinations as to whether an answer is possible is less than the predetermined number (step S3: NO), the dialogue processing unit 221 presents a guiding question (step S4). More specifically, the dialogue processing unit 221 acquires guidance question information corresponding to the question from the scenario DB 211 and transmits screen information of the question screen SC including the guidance question information to the control device 130 . As a result, a new question screen (not shown) appears on the display device 120 . For example, when the user operates the input device 110 to press a selection button, the control device 130 transmits options associated with the selection button to the dialogue server 200 as a response to the guiding question. Thereby, the obtaining unit 222 obtains a reply (step S5). When the acquisition unit 222 acquires the reply, the dialogue processing unit 221 again determines whether or not the reply is possible by the process of step S2.

If the number of determinations as to whether an answer is possible is equal to or greater than the predetermined number (step S3: YES), the narrowing down unit 225 executes first narrowing down processing (step S6). The first narrowing-down process is a process of specifying sentences with question marks from the acquired questions or answers, performing morphological analysis on the sentences, classifying them into categories, and narrowing down e-mails in the mail DB 213 based on the result.

Details of the first narrowing process will now be described with reference to FIG. 10 . When the narrowing-down unit 225 starts executing the first narrowing-down process, first, sentences with question marks are specified (step S11). The question mark is specifically a question mark "?". When the acquisition unit 222 acquires the question "I want to know about product A. How much is product A?" do.

After identifying the sentences with question marks, the narrowing unit 225 then performs morphological analysis on the sentences with question marks (step S12). The narrowing-down unit 225 may use the morphological-analyzing function of another server different from the dialogue server 200 instead of providing the narrowing-down unit 225 with the morphological analysis function. For example, when specifying a sentence with a question mark "How much is A product?", the morpheme analysis decomposes it into the morpheme "A product", the morpheme "ha", the morpheme "how much", and the morpheme "?".

After executing the morphological analysis, the narrowing unit 225 performs category classification (step S13). More specifically, the narrowing-down unit 225 refers to the category table DB 214 and performs category classification. As shown in FIG. 11, the category table DB 214 stores a category table in which a plurality of characters and categories to which the plurality of characters belong are associated in advance. Several characters and categories are pre-associated. The narrowing-down unit 225 acquires a plurality (or all) of characters such as the character “how much” and the character “performance” registered in the category table, compares the acquired plurality of characters with morphemes one by one, and finds matching characters. Identify. For example, when the morpheme "A product", the morpheme "ha", the morpheme "how much", and the morpheme "?"

Once the characters are identified, narrowing down unit 225 determines the category. More specifically, the narrowing down unit 225 refers to the category table DB 214 and determines the category to which the specified character belongs. For example, when the character "how much" is specified, as shown in FIG. 11, the narrowing unit 225 determines the category "cost" to which the character "how much" belongs.

After completing the category classification, the narrowing down unit 225 narrows down the e-mails (step S14). More specifically, the narrowing down unit 225 refers to the mail DB 213 and narrows down the e-mails based on the determined category. As described above, for example, when the category "cost" is determined, the narrowing unit 225 refers to the mail DB 213 and narrows down the e-mails including the category "cost" from various e-mails stored in the mail DB 213. For example, as shown in FIG. 6, e-mails 31, 32, and 33 are stored in the mail DB 213, and if the e-mails 32 and 33 do not include the category "cost", the narrowing unit 225 selects the category "cost". The e-mails 31 containing . As a result, the narrowing-down unit 225 ends the first narrowing-down process.

After completing the first narrowing-down process, as shown in FIG. 8, the narrowing-down unit 225 determines whether or not the e-mails have been narrowed down (step S7). For example, the result of narrowing down e-mails may be a huge number such as 1000. In this case, the narrowing down unit 225 determines that the e-mails have not been narrowed down (step S7: NO). On the other hand, the results of narrowing down the e-mails may range from several to a dozen or so. In this case, the narrowing down unit 225 determines that the e-mails have been narrowed down (step S7: YES). The threshold number of e-mails for judging whether or not the e-mails have been narrowed down can be appropriately determined according to settings, design, and the like.

If the e-mails have not been narrowed down, the narrowing down unit 225 cooperates with the noun determining unit 223 and the question category specifying unit 224 to execute a second narrowing process (step S8). If the e-mails are narrowed down, the narrowing-down unit 225 skips the process of step S8.

Now, with reference to FIG. 12, the details of the second narrowing down process will be described. When the narrowing-down unit 225 starts to execute the second narrowing-down process, first, it identifies an interrogative sentence (step S21). When the acquisition unit 222 acquires the question "I want to know about product A. How much does product A cost?" Then it can be estimated. In such a case, the narrowing-down unit 225 identifies this question as corresponding to an interrogative sentence. In addition to the word "want to know", the word "want" or the word "tell me" may be used. In addition to these words, the acquisition unit 222 may also acquire a non-verbal ending such as the question "price of product A". Even in the case of such an implied ending, the narrowing-down unit 225 identifies this question as corresponding to an interrogative sentence.

After specifying the interrogative sentence equivalent, the narrowing unit 225 then performs morphological analysis on the interrogative sentence equivalent (step S22). For example, when specifying the question equivalent to "I want to know about A product. How much is A product?" etc.

After executing the morphological analysis, the noun determination unit 223 then determines a comparison keyword (step S23). More specifically, the noun determination unit 223 first identifies nouns that appear in interrogative sentences and the frequency of appearance of the nouns. As described above, when the morphological analysis is performed on the interrogative equivalent "I want to know about product A. How much is product A?", the noun morpheme "A product" appears twice. Identify the noun "A product" and the occurrence frequency "twice". On the other hand, since the remaining morphemes are parts of speech other than nouns, the noun determining unit 223 cannot specify nouns and frequency of occurrence.

After specifying the noun and the frequency of appearance, the noun determination unit 223 determines whether the frequency of appearance of the noun is equal to or higher than the threshold frequency. is determined as a comparison keyword to be compared with. For example, if a threshold frequency of "2 times" is set, the noun "A product" has an appearance frequency of "2 times", so it can be determined that the frequency is equal to or higher than the threshold frequency. In this case, the noun determining unit 223 determines this noun "A product" as the comparison keyword "A product". If the appearance frequency of nouns is less than the threshold frequency, the noun determination unit 223 terminates the second narrowing process without determining the comparison keyword.

After determining the comparison keyword, the question category identifying unit 224 identifies the question category as shown in FIG. 12 (step S24). More specifically, first, the question category specifying unit 224 refers to the answer DB 215 and determines whether the determined comparison keyword matches the answer category stored in the answer DB 215 . Here, as shown in FIG. 9, the answer DB 215 includes the category "A product". Therefore, the question category identifying unit 224 determines that the comparison keyword “A product” determined by the noun determining unit 223 matches the answer category “A product” stored in the answer DB 215 . In this case, the question category identifying unit 224 identifies the category "A product" that matches the comparison keyword "A product" as the question category of the question. In this way, when the comparison keyword matches the answer category, it can be recognized that the comparison keyword is an effective keyword. Therefore, if no category that matches the comparison keyword is stored in answer DB 215, noun determination unit 223 terminates the second narrowing process without specifying the question category.

After identifying the question category, the narrowing down unit 225 compares the question category and the email selection keyword (step S25), and narrows down the emails (step S26). More specifically, the narrowing down unit 225 compares the question category specified by the question category specifying unit 224 with the email selection keywords associated with the emails stored in the email DB 213 to narrow down the emails. For example, when the question category identifying unit 224 identifies the question category “A product”, the narrowing unit 225 matches the question category “A product” among the plurality of e-mails 31, 32, and 33 stored in the mail DB 213. Filter to emails associated with email selection keywords that you want. In this embodiment, as shown in FIG. 6, the e-mail 31 is associated with the e-mail selection keyword "A product", so the narrowing-down unit 225 selects at least e-mails from among the plurality of e-mails 31, 32, and 33. Narrow down to 31. After narrowing down the e-mails, the narrowing-down unit 225 ends the second narrowing-down process.

When the first narrowing-down process or the second narrowing-down process ends, as shown in FIG. 8, the answer presentation unit 226 and the notification unit 227 cooperate to execute the notification process (step S9), and the process ends. The notification process is a process of presenting the e-mail answers narrowed down by the narrowing unit 225 as answers to the questions and notifying the administrator terminal 150 of the narrowed-down e-mail answers. In this embodiment, the notification process is executed for either e-mails narrowed down by the category "cost" or e-mails narrowed down by the relationship between the question category and the e-mail selection keyword.

Details of the notification process will now be described with reference to FIG. 13 . First, the answer presenting unit 226 gives points to the e-mail (step S31). More specifically, the answer presenting section 226 gives scores to the e-mails narrowed down by the narrowing section 225 based on predetermined conditions. As the predetermined condition, for example, a mail update date when the e-mail is updated by an answer to an inquiry by e-mail can be used. When an e-mail containing an answer is returned to an e-mail containing only an inquiry, the reply date of the reply e-mail becomes the mail update date. The answer presenting unit 226 gives a higher score to an e-mail whose mail update date is closer to the latest.

In addition, an answer evaluation given to an answer based on e-mail may be employed as a predetermined condition. In this case, the answer presenting unit 226 gives a higher score to the e-mail for which the answer evaluation is higher. Also, the response frequency of responses based on e-mail may be employed as a predetermined condition. In this case, the answer presenting unit 226 gives a higher score to an e-mail with a higher frequency of replies presented to the user by e-mail. Furthermore, a specific word (for example, "failure" or "precaution") included in the reply text of the e-mail and having a high degree of failure-inducing influence may be adopted as the predetermined condition. In this case, the answer presenting unit 226 gives a higher score to the e-mail containing these words. Any one of these predetermined conditions may be used, or two or more of them may be used in combination.

After giving points, the answer presenting unit 226 identifies high-scoring e-mails (step S32). For example, the answer presenting unit 226 identifies several e-mails higher than a threshold score according to the design or a preset threshold score as high-score e-mails. The number of e-mails specified by the answer presenting unit 226 can also be determined according to the design, settings, and the like. Although the details will be described later, if the number of e-mails is set to more than 10, the confirmation burden on the reply creator increases, so it is desirable that the number of e-mails is several (for example, about 3 to 5).

When the e-mail is specified, the reply presenting unit 226 presents the reply text of the e-mail (step S33). More specifically, the answer presenting unit 226 acquires the specified e-mails from the mail DB 213 and presents to the user terminal 100 the answer contained in the e-mail with the highest score among the acquired e-mails. For example, if the e-mail 31 (see FIG. 6) has the highest score, the answer presenting unit 226 presents the answer included in the e-mail 31 to the user terminal 100 . At this time, the reply presenting unit 226 transmits to the control device 130 extracted information obtained by extracting the entire text or a part of the text from the specified e-mail as a reply text. When receiving the extracted information, the control device 130 updates the question screen SC displayed on the display device 120 based on the received extracted information.

Thereby, the question screen SC is updated on the display device 120 . 5 ( As shown in b), the answer appears in the question screen SC. When the answer presenting unit 226 extracts a part of the text from the e-mail, for example, the remaining part after removing the part including the person's name, the part including the greeting, and the part including the quotation marks from the entire text is taken as part of the text. good. In the case of the e-mail 31, the portions including the person's name "Mr. Taro Yamada" and "Responsible for company XX: Hanako Fuji" are removed. Similarly, the part containing the greeting "... thank you." is removed, and the part containing the quotation mark ">" is also removed.

In this way, the user is presented with the answer contained in the e-mail with the highest score almost in real time without withholding the answer from the interaction server 200 . In particular, since the answer sentence of the e-mail corresponding to the question is presented as the answer to the question, it can be said that the possibility of understanding the intention of the question is extremely high. In this way, since an answer sentence with relatively high answer accuracy is presented to the user, it is possible to avoid negative answer evaluation by the user. When there are a plurality of e-mails with the highest score, the answer presenting unit 226 may present the reply sentences included in each of the plurality of e-mails, or any one of the e-mails selected intentionally or at random. You may present the reply text contained in the said e-mail.

When the e-mail response text is presented, the notification unit 227 notifies the e-mail response text (step S34), and terminates the notification process. More specifically, the notification unit 227 acquires high-scoring e-mails specified by the answer presentation unit 226, and notifies the control unit of the administrator terminal 150 of the answer sentences included in the acquired e-mails as answer preparation support information. When the control unit of the administrator terminal 150 is notified of the answer creation support information, it displays an answer creation support screen as shown in FIG. 14 based on the notified answer creation support information.

The answer creation support screen includes items such as date of question, time of question, content of question, and a plurality of answer candidates as items. Answer sentences included in high-scoring e-mails specified by the answer presentation unit 226 are displayed in the multiple answer candidate items. If there are several e-mails with high scores, each reply text included in the e-mails is displayed. As shown in the answer candidate "Answer candidate #n", answer materials such as materials and information (specifically, document information and image information) for answers may be displayed in addition to answer sentences. A display column in which an answer sentence is displayed on the answer preparation support screen can be selected by the pointer PT. For example, when the answer displayed in the answer candidate "candidate answer #1" is selected by the pointer PT, and then the add button 51 is pressed, the controller of the administrator terminal 150 displays the selected answer and the question. are associated with each other and transmitted to the dialogue server 200 .

On the other hand, when the download button 52 is pressed, the controller of the administrator terminal 150 downloads the multiple items displayed on the answer preparation support screen from the dialogue server 200 . After downloading the plurality of items, the control unit activates, for example, spreadsheet software and arranges the plurality of items in a state consistent with the display format of the answer preparation support screen. As a result, the administrator can smoothly proceed with the answer preparation work.

For example, upon receiving an answer sentence and a question, the interaction processing unit 221 of the interaction server 200 assigns an answer ID to the received answer sentence, and associates and manages the question ID and answer ID of the question information including the question. As a result, when the dialogue processing unit 221 receives a question similar to the question with the question ID “Q002” (see FIG. 7) at a different time, the dialogue processing unit 221 receives the answer ID associated with the question ID “Q002”. An answer sentence can be presented to the user. In this way, with the answer preparation support screen according to the present embodiment, it is possible to understand the intent of the question and assist in improving the response rate of the answer to the question. In particular, since such an answer preparation support screen is presented to the administrator, it is possible to improve the efficiency of answer maintenance such as addition of answers and maintenance by the administrator. In addition, if the number of displayed e-mail response sentences is several (for example, about 3 to 5), it is possible to avoid an increase in the confirmation burden on the response creator and the response creation burden.

Although the preferred embodiments of the present invention have been described in detail above, the present invention is not limited to the specific embodiments, and various modifications can be made within the spirit and scope of the present invention described in the claims.・Changes are possible. The various servers described above may be realized by physical servers or by virtual servers.

Further, in the embodiment described above, it has been explained that sentences with question marks are specified in the process of step S11, and interrogative sentences are specified in the process of step S21. However, it is not limited to such identification, and for example, these processes may be reversed, or sentences with question marks and interrogative sentences may be identified in both the processes of steps S11 and S21. Also, in the above-described embodiment, it has been explained that the relationship between the question category and the email selection keyword is one-to-one. However, if the relationship between the question category and the email selection keyword is 1:2 or more, or if the relationship is 2 or more:1, the e-mail should be narrowed down.

ST dialogue system 100 user terminal 150 administrator terminal 200 dialogue server 210 storage unit 220 processing unit 221 dialogue processing unit 222 acquisition unit 223 noun determination unit 224 question category identification unit 225 narrowing unit 226 answer presentation unit 227 notification unit 230 communication unit

Claims (6)

get questions,
If the question contains a question mark or is equivalent to an interrogative sentence and an answer corresponding to the question cannot be obtained, the answer is based on the appearance frequency of nouns that appear in the sentence with the question mark or the interrogative sentence. Determining a contrasting noun to contrast with the category of the given answer;
identifying the category that matches the contrasting noun as a question category of the question;
narrowing down the e-mail based on a comparison result of the question category and a specific word selected from the reply text of the e-mail and associated with the e-mail;
Presenting the answer sentence of the narrowed down e-mail as the answer corresponding to the question,
Notifying the terminal of the reply creator of the reply text of the narrowed down e-mail;
An answer creation support program for executing processing. In the process of determining the noun for comparison, based on morphological analysis of the sentence or the interrogative sentence, nouns that appear in the sentence or the interrogative sentence are specified, and based on the appearance frequency of the specified noun, determining the contrast noun;
2. The answer creation support program according to claim 1, characterized by: The process of notifying the reply includes: the mail update date when the e-mail was updated by the reply in response to the inquiry by the e-mail, the evaluation given to the reply based on the e-mail, and the e-mail giving a score to the narrowed down e-mail based on at least one of the answer frequency of the answer based on, and at least one of the specific words contained in the answer sentence of the e-mail,
Notifying the terminal of the response text of the e-mail whose score is higher than a threshold score;
3. An answer creation support program according to claim 1 or 2, characterized by: The answer creation support program is a program used for a scenario-type chatbot that interacts with the questioner according to a predetermined scenario about a specific topic.
4. The answer creation support program according to any one of claims 1 to 3, characterized by: get questions,
If the question contains a question mark or is equivalent to an interrogative sentence and an answer corresponding to the question cannot be obtained, the answer is based on the appearance frequency of nouns that appear in the sentence with the question mark or the interrogative sentence. Determining a contrasting noun to contrast with the category of the given answer;
identifying the category that matches the contrasting noun as a question category of the question;
narrowing down the e-mail based on a comparison result of the question category and a specific word selected from the reply text of the e-mail and associated with the e-mail;
Presenting the answer sentence of the narrowed down e-mail as the answer corresponding to the question,
Notifying the terminal of the reply creator of the reply text of the narrowed down e-mail;
An answer creation support method that performs processing. get questions,
If the question contains a question mark or is equivalent to an interrogative sentence and an answer corresponding to the question cannot be obtained, the answer is based on the appearance frequency of nouns that appear in the sentence with the question mark or the interrogative sentence. Determining a contrasting noun to contrast with the category of the given answer;
identifying the category that matches the contrasting noun as a question category of the question;
narrowing down the e-mail based on a comparison result of the question category and a specific word selected from the reply text of the e-mail and associated with the e-mail;
Presenting the answer sentence of the narrowed down e-mail as the answer corresponding to the question,
Notifying the terminal of the reply creator of the reply text of the narrowed down e-mail;
An answer preparation support device having a processing unit.

Images