JP2023176691A - Information processor, program, information processing system, and information processing method - Google Patents

Abstract

To provide an information processor capable of utilizing log information more efficiently on messages as knowledge.SOLUTION: An information processor includes: extraction means for extracting log information on messages written by a plurality of users for communication; classification means for classifying messages included in the log information into at least question messages and answer messages, using predetermined conditions to perform the classification on the messages; assignment means for assigning the answer message to the question message on the basis of a classification result; and output means for outputting the assigned question message and the answer message as a pair of messages.SELECTED DRAWING: Figure 3

Description

The present invention relates to an information processing device, a program, an information processing system, and an information processing method.

Chat systems in which multiple users exchange information and consult via messages have been known for a long time. In a chat system, messages exchanged between multiple users are stored as a chat log or the like.

For example, in Patent Document 1, when receiving a question from a user, a knowledge database is created in a chatbot system that generates an answer to the question using a knowledge database that stores the question and its answer in advance. A technique for doing so has been disclosed.

Currently, questions and answers such as information exchange and consultation have to be carried out by knowledgeable individuals. It is conceivable to utilize chat logs in such a chat system as a knowledge database. In addition, in this case, there was a problem in that chat logs stored in conventional chat systems do not store question and answer messages separately, resulting in an extremely large amount of man-hours.

Note that Patent Document 1 is based on the premise that character data by the questioner and character data by the answerer can be acquired separately, and does not solve the above problem.

An embodiment of the present invention aims to provide an information processing device that can use message log information as knowledge more efficiently.

In order to achieve the above-mentioned problems, one embodiment of the present invention includes an extraction means for extracting log information of messages written by a plurality of users for communication, and classification of the messages using predetermined conditions. a classification means for classifying at least a question message and an answer message from the messages included in the log information; an assignment means for allocating the answer message to the question message based on the classification result; and output means for outputting the message of the question and the message of the answer as a pair of messages.

According to an embodiment of the present invention, message log information can be used as knowledge more efficiently.

FIG. 1 is a configuration diagram of an example of an information processing system according to an embodiment. FIG. 1 is a hardware configuration diagram of an example of a computer according to the present embodiment. FIG. 1 is a functional configuration diagram of an example of an information processing system according to the present embodiment. 3 is a flowchart of an example of a processing procedure of the information processing apparatus according to the present embodiment. It is an explanatory diagram of an example of a chat log. FIG. 3 is an explanatory diagram of an example of messages included in a chat log. 3 is a flowchart of an example of a processing procedure for message classification. FIG. 3 is an explanatory diagram of an example of message classification processing. FIG. 3 is an explanatory diagram of an example of a message after message classification; 3 is a flowchart of an example of a processing procedure for processing a message. FIG. 3 is an explanatory diagram of an example of a message after personal information has been removed. FIG. 7 is an explanatory diagram of an example of a message after removing extra text. 3 is a flowchart of an example of a processing procedure for message allocation. 12 is a flowchart of an example of the processing procedure of step S62. FIG. 2 is an explanatory diagram of an example of a semantic space. 12 is a flowchart of an example of the processing procedure of step S62. FIG. 1 is a configuration diagram of an example of an information processing system according to an embodiment. FIG. 1 is a functional configuration diagram of an example of an information processing system according to the present embodiment.

Hereinafter, each embodiment of the present invention will be described with reference to the accompanying drawings.

[First embodiment]
<System configuration>
FIG. 1 is a configuration diagram of an example of an information processing system according to this embodiment. The information processing system 1 according to the present embodiment has a configuration in which an information processing device 10, an administrator terminal 12, and a user terminal 14 are communicably connected via a network 18 such as the Internet or a LAN (Local Area Network). be.

The user terminal 14 is a computer operated by a user who uses a chat function. The information processing device 10 provides a chat function to a plurality of users who operate the user terminals 14. The information processing device 10 stores a chat log of a plurality of users who operate the user terminal 14 (in this embodiment, this is called a chat log for convenience, but it is information that stores the exchange of messages such as questions and answers, and the name is limited to this). (It may also be referred to as historical information, dialogue information, etc.). A chat log is an example of log information of messages exchanged between multiple users for communication.

The administrator terminal 12 is a computer operated by an administrator. The information processing device 10 provides the administrator who operates the administrator terminal 12 with functions described below for more efficiently using chat logs as knowledge.

When the information processing device 10 receives a message specifying a chat group from a user terminal 14, the information processing device 10 distributes the message to a plurality of user terminals 14 participating in the chat group. Provide chat functionality. The information processing device 10 stores messages exchanged in a certain chat group as a chat log.

By transmitting and receiving data to and from the administrator terminal 12, the information processing device 10 executes information processing according to the administrator's operation received by the administrator terminal 12, and provides the execution results to the administrator terminal 12. . The administrator terminal 12 displays the execution results provided from the information processing device 10.

Note that the configuration of the information processing system 1 shown in FIG. 1 is an example. The information processing device 10 may be implemented using a single computer or multiple computers, or may be implemented using a cloud service. Further, the information processing device 10, the administrator terminal 12, and the user terminal 14 include, for example, a projector, a display device having an electronic whiteboard function, an output device such as a digital signage, a HUD (Head Up Display) device, an industrial machine, an imaging device, Whether it's a sound collection device, medical equipment, network appliances, connected cars, personal computers, mobile phones, smartphones, tablets, game consoles, PDAs, digital cameras, wearable PCs, or desktop PCs, etc. good.

<Hardware configuration>
"Computer"
The information processing device 10, administrator terminal 12, and user terminal 14 in FIG. 1 are realized, for example, by a computer 500 having a hardware configuration shown in FIG. 2. FIG. 2 is a hardware configuration diagram of an example of a computer according to this embodiment.

The computer 500 includes a CPU (Central Processing Unit) 501, a ROM (Read Only Memory) 502, a RAM (Random Access Memory) 503, an HD 504, an HDD (Hard Disk Drive) controller 505, a display 506, and an external device connection I/F (Interface). ) 508, a network I/F 509, a data bus 510, a keyboard 511, a pointing device 512, a DVD-RW (Digital Versatile Disk Rewritable) drive 514, and a media I/F 516.

Among these, the CPU 501 controls the overall operation of the computer 500 according to a program. The ROM 502 stores programs used to drive the CPU 501 such as IPL. RAM 503 is used as a work area for CPU 501. The HD 504 stores various data such as programs. The HDD controller 505 controls reading and writing of various data to the HD 504 under the control of the CPU 501.

The display 506 displays various information such as a cursor, menu, window, characters, or images. External device connection I/F 508 is an interface for connecting various external devices. The external device in this case is, for example, a USB (Universal Serial Bus) memory. The network I/F 509 is an interface for data communication using the network 18. The data bus 510 is an address bus, a data bus, etc. for electrically connecting each component such as the CPU 501.

The keyboard 511 is a type of input means that includes a plurality of keys for inputting characters, numbers, various instructions, and the like. The pointing device 512 is a type of input means for selecting and executing various instructions, selecting a processing target, moving a cursor, and the like. The DVD-RW drive 514 controls reading and writing of various data on a DVD-RW 513, which is an example of a removable recording medium. Note that it is not limited to DVD-RW, but may be DVD-R or the like. The media I/F 516 controls reading or writing (storage) of data to a recording medium 515 such as a flash memory.

Note that the hardware configuration shown in FIG. 2 is an example, and it is not necessary to include all the components shown in FIG. 2, or it may include components other than those shown in FIG.

<Functional configuration>
FIG. 3 is a functional configuration diagram of an example of the information processing system according to this embodiment. In the following, illustration of the user terminal 14 is omitted in order to explain processing after a chat log of messages exchanged by a plurality of users operating the user terminal 14 is stored in the information processing device 10.

Further, the functional configuration shown in FIG. 3 is an example. For example, at least a part of the functional configuration of the information processing device 10 shown in FIG. 3 may be provided in the administrator terminal 12 or another information processing device to perform processing in cooperation with each other. Note that, in the functional configuration diagram of FIG. 3, configurations unnecessary for explaining the present embodiment are omitted as appropriate.

For example, the information processing device 10 realizes the functional configuration shown in FIG. 3 by executing an OS and programs. The information processing device 10 in FIG. 3 includes a chat processing section 30, a log information storage section 32, an initial setting section 34, an extraction section 36, a classification section 38, a processing section 40, an allocation section 42, an output section 44, and an allocation result storage section 46. , an operation receiving section 48, and an allocation result updating section 50. Further, the administrator terminal 12 implements the functional configuration shown in FIG. 3 by executing the OS and programs. The administrator terminal 12 has a display section 60 and an operation reception section 62.

The chat processing unit 30 provides a chat function to a plurality of users who operate the user terminal 14, and allows the plurality of users to exchange messages written in text or the like. The log information storage unit 32 stores messages exchanged via text or the like between multiple users as a chat log.

The initial setting unit 34 receives settings for chat logs to be monitored from, for example, an administrator operating the administrator terminal 12. The chat log to be monitored may be set, for example, by specifying a chat group. Further, it is assumed that the chat log to be monitored includes messages from the questioner and messages from the answerer.

The extraction unit 36 extracts the chat log to be monitored from the log information storage unit 32. For example, the extraction unit 36 extracts the chat log so that the speaker of the message and the order of message exchange can be identified. The extraction unit 36 may extract the chat log in such a way that parent-child relationships of messages can be identified.

The classification unit 38 analyzes and classifies the messages included in the chat log extracted by the extraction unit 36. For example, the classification unit 38 classifies at least question messages and answer messages from the messages included in the chat log using predetermined conditions, that is, a machine-learned classification model described below. Details of message classification by the classification unit 38 will be described later.

The processing unit 40 performs processing for removing personal information, removing unnecessary sentences, and forming sentences for messages included in the extracted chat log. For example, the processing unit 40 uses various machine-learned models to perform processing such as removing personal information, removing unnecessary sentences, and forming sentences from messages included in the extracted chat log.

The assignment unit 42 assigns answer messages to question messages based on the results of classification by the classification unit 38. For example, if the chat log is extracted so that the parent-child relationship of the message can be specified, the allocation unit 42 allocates the answer message to the question message based on the parent-child relationship information. The process of assigning an answer message to a question message may be expressed as matching the question message and the answer message.

Furthermore, the assignment unit 42 may assign answer messages to question messages based on the order of message exchanges, as long as the chat logs are extracted so that the order of message exchanges can be specified. The assignment unit 42 may assign answer messages that are close in position on the semantic space, as described below, to the question message based on message similarity.

In addition, even if the chat log is extracted so that the parent-child relationship of messages can be identified, the assignment unit 42 assigns questions based on message similarity, giving priority to messages whose answers are close in the semantic space as described below. may be assigned to the message. The allocation unit 42 may allocate answer messages to question messages based on one or more of the parent-child relationship of messages, the order of exchange, and the similarity of messages.

The output unit 44 outputs the question message and answer message assigned by the assignment unit 42 as a pair of messages. The output unit 44 transmits the message data of the question message and answer message pair to the administrator terminal 12, and causes the administrator terminal 12 to display the question message and answer message pair.

The display unit 60 of the administrator terminal 12 displays a pair of messages, a question message and an answer message, output by the information processing device 10. Therefore, the administrator can easily check the paired messages of the question message and the answer message included in the chat log to be monitored.

For example, the administrator can use a pair of messages, a question message and an answer message, included in a chat log to be monitored, as knowledge. The Q&A (Question and Answer) format, in which a question message and an answer message are paired, can be used for FAQ (Frequently Asked Questions), which is a collection of frequently asked questions and their answers.

Furthermore, the administrator may utilize the paired messages, the question message and the answer message, as learning data for an application program such as a chatbot. Chatbot is a term that combines chat and bot, and is an automatic conversation program that uses artificial intelligence. The pair of messages, a question message and an answer message, included in the monitored chat log can be used, for example, as learning data necessary for a chatbot to learn to have a conversation.

Further, the operation reception unit 62 of the administrator terminal 12 receives an operation from the administrator to update the paired message of the question message and the answer message displayed on the display unit 60. Further, if the administrator needs to update, such as modifying, the pair of messages of the question message and the answer message displayed on the display section 60, the administrator performs an operation from the operation reception section 62. The operation reception unit 62 transmits the operation details received from the administrator to the information processing device 10.

The operation receiving unit 48 of the information processing device 10 receives the operation details that the administrator terminal 12 receives from the administrator. The assignment result storage unit 46 stores the paired messages of the question message and the answer message assigned by the assignment unit 42 as assignment results. The assignment result update unit 50 updates the pair of messages of a question message and an answer message stored in the assignment result storage unit 46 according to the operation details that the administrator terminal 12 receives from the administrator.

In this way, the administrator uses the administrator terminal 12 to check the paired messages of the question message and the answer message that the information processing device 10 has classified and processed the chat logs to be monitored, and has outputted them. After it is corrected, it can be used as knowledge.

<Processing>
FIG. 4 is a flowchart of an example of a processing procedure of the information processing apparatus according to the present embodiment. In step S10, the initial setting unit 34 of the information processing device 10 receives settings for a chat log to be monitored from, for example, an administrator operating the administrator terminal 12.

In step S12, the extraction unit 36 of the information processing device 10 extracts a chat log to be monitored, such as that shown in FIG. 5, from the log information storage unit 32, for example.

FIG. 5 is an explanatory diagram of an example of a chat log. The chat log shown in FIG. 5 includes the date and time, speaker, message, and reply message as items. The item "date and time" represents the date and time when the message was registered. The item "speaker" is an example of identification information of a user who has registered a message. The item "message" is an example of text data representing the message of the statement. Further, the item "reply message" is an example of text data representing a message in response to a comment.

The chat log in FIG. 5 is an example of a chat log in which the speaker of a message, the parent-child relationship of the message, and the order of message exchange can be identified. In the chat log of FIG. 5, a message of a comment and a message of a reply to the comment have a parent-child relationship. For example, the message of the statement is "parent". The message in reply to the comment is "child". The chat log extracted in step S12 may be a chat log in which the parent-child relationship of messages is not specified.

In step S14, the classification unit 38 analyzes and classifies the message shown in FIG. 6, for example, included in the chat log extracted in step S12, as shown in FIGS. 7 and 8. FIG. 6 is an explanatory diagram of an example of messages included in the chat log. FIG. 7 is a flowchart of an example of a processing procedure for message classification. Further, FIG. 8 is an explanatory diagram of an example of message classification processing.

FIG. 6 shows the messages included in the chat log of FIG. 5 in association with the item "Message No.". In step S30 of FIG. 7, the classification unit 38 extracts one unprocessed message from the messages as shown in FIG. 6 included in the chat log extracted in step S12.

In step S32, the classification unit 38 classifies the message extracted in step S30 using a machine-learned classification model as shown in FIG. 8, for example. FIG. 8 shows an example in which the classification unit 38 uses three types of machine-learned classification models 100A to 100C.

The machine-learned classification model 100A has machine-learned the characteristics of question messages, answer messages, and other messages, and when the message extracted in step S30 is input, the input message is classified as a question message. , reply messages, or other messages. In FIG. 8, the classification by the machine-learned classification model 100A is indicated as classification A.

The machine-learned classification model 100B has machine-learned the characteristics of complaint messages, emergency messages, and ordinary messages, and when the message extracted in step S30 is input, the input message is classified as a complaint message. , an urgent message, or a normal message. In FIG. 8, the classification by the machine-learned classification model 100B is indicated as classification B.

In addition, the machine-learned classification model 100C has machine-learned message characteristics such as laws and regulations, specifications, paperwork, and others, and when the message extracted in step S30 is input, the classification model 100C Classify into laws, specifications, administrative procedures, and others. In addition, in FIG. 8, the classification by the machine-learned classification model 100C is indicated as classification C.

In step S34, the classification unit 38 outputs the result of classifying one message extracted in step S30. For example, as shown in FIG. 8, the classification unit 38 classifies one message extracted in step S30 from three types of viewpoints, and outputs the classification results.

By repeating steps S30 to S36 until there are no unprocessed messages in the messages in FIG. 6, the classification unit 38 sorts the messages in FIG. 6 included in the chat log extracted in step S12 from one or more viewpoints. The classification results are output in association with messages as shown in FIG. 9, for example.

FIG. 9 is an explanatory diagram of an example of a message after message classification. In FIG. 9, the results of classification by the classification unit 38 of FIG. 8 are associated with messages. For example, for a message whose message number is "1", the classification unit 38 classifies classification A into "question", classification B into "normal", and classification C into "law".

Returning to step S16 in FIG. 4, the processing unit 40 performs processing for removing personal information, removing unnecessary sentences, and forming sentences for the message shown in FIG. 6, for example, which was included in the chat log extracted in step S12.

FIG. 10 is a flowchart of an example of a message processing procedure. In step S40, the processing unit 40 extracts one unprocessed message from the messages shown in FIG. 6 included in the chat log extracted in step S12.

In step S42, the processing unit 40 removes personal information from the message extracted in step S40. Personal information may be removed using a model that has undergone machine learning of the characteristics of personal information. Further, personal information may be removed using proper name extraction technology. For example, the processing unit 40 lists the named entities (proper names, etc.) included in the message, selects the reference information search target and search method for each entity expression category (place name, person name, etc.), and selects the search target and search method for the named entity. By searching for reference information, personal information may be identified and removed from the listed unique expressions based on the searched reference information. FIG. 11 is an explanatory diagram of an example of a message after personal information has been removed. The message in FIG. 11 has the personal information removed from the message in FIG.

In step S44, the processing unit 40 removes unnecessary sentences from the message of FIG. 11 from which personal information has been removed in step S42. The extra sentences may be removed by identifying the features of the extra sentences using a machine-learned model. Further, the removal of unnecessary sentences may be performed using linguistic analysis such as morphological analysis. For example, the processing unit 40 classifies the text of the message into parts such as main text, question, supplement, greeting, thank you, and others, and removes the parts such as greeting, thank you, and others as redundant text. FIG. 12 is an explanatory diagram of an example of a message after unnecessary text has been removed. Note that morphological analysis is just one example, and kana-kanji conversion, proofreading support, dependency analysis, key phrase extraction, and natural language understanding may be used instead of or in addition to morphological analysis.

In step S46, the processing unit 40 performs sentence shaping on the message of FIG. 12 from which the extra sentences were removed in step S44. The sentence formation may be performed using, for example, deep learning technology. For example, the processing unit 40 performs sentence formation by replacing each word in the message of FIG. 12 from which extra sentences have been removed in step S44 with a word that has a higher occurrence rate.

Returning to step S18 in FIG. 4, the allocation unit 42 performs a process of allocating a message whose classification A is “question” to a message whose classification A is “answer” based on the result of classification by the classification unit 38 in step S14.

FIG. 13 is a flowchart of an example of a message allocation processing procedure. In step S60, the allocation unit 42 extracts one message whose classification A is "question" from the unprocessed messages processed in step S16.

In step S62, the assignment unit 42 specifies a message whose classification A is "answer" to be assigned to the message whose classification A is "question" extracted in step S60. For example, if a message is extracted from a chat log for which a parent-child relationship can be specified, the assignment unit 42 can identify a message whose classification A is "answer" that has a parent-child relationship with the "question" message extracted in step S60.

On the other hand, if the message is extracted from a chat log for which the parent-child relationship cannot be determined, the assignment unit 42 uses natural language processing technology (https://jp.ricoh. com/technology/tech/115_deep_alignment) may be used in the process of step S62.

By using natural language processing technology that utilizes artificial intelligence to automatically match documents containing similar content, the assignment unit 42 is able to automatically match documents containing similar content to the "Question" message extracted in step S60. ” messages can be identified.

Further, if the message is extracted from a chat log in which the order of message exchange can be specified, the assignment unit 42 assigns the message whose classification A is "answer" to the "question" message extracted in step S60, for example, as shown in FIG. It can be identified as shown in FIG.

FIG. 14 is a flowchart of an example of the processing procedure of step S62. Further, FIG. 15 is an explanatory diagram of an example of a semantic space. In step S100, the allocation unit 42 selects messages in a predetermined range (for example, the first 10 messages and the next 20 messages) based on the order of the "question" messages extracted in step S60, which have been processed in step S16. Extract from the message.

In step S102, the allocation unit 42 removes messages whose classification A is "question" from the messages extracted in step S100. In step S104, the allocation unit 42 arranges messages whose classification A is "answer" that was not removed in step S102, for example, in a semantic space as shown in FIG. 15.

In the semantic space of FIG. 15, example messages of the question document and example messages of the answer document are arranged. The message of the question document and the message of the answer document, which are located close to each other in the semantic space of FIG. 15, can be recognized as messages containing similar content.

In step S106, the assignment unit 42 identifies an "answer" message near the position of the "question" message extracted in step S60 on the semantic space of FIG. 15. In the semantic space of FIG. 15, if there is no "answer" message within a predetermined range from the position of the "question" message extracted in step S60, the assignment unit 42 does not need to specify the "answer" message. .

Further, the allocation unit 42 may use classification B when specifying an "answer" message near the location of a "question" message. Classification B represents the urgency of the message. The allocation unit 42 may use classification B as a distance weighting coefficient in the semantic space to make adjustments so that messages with high urgency are more likely to be characterized.

Furthermore, the assignment unit 42 may use classification C when specifying an "answer" message near the location of a "question" message. Classification C represents the field of the message. The assignment unit 42 may use the classification C as a distance weighting coefficient in the semantic space to make adjustments to make it easier for messages in the same field to be characterized.

When the "answer" message is identified in step S62, the assignment unit 42 proceeds to step S64 and assigns the "answer" message identified in step S62 to the "question" message extracted in step S60. The assignment unit 42 can assign "answer" messages to "question" messages by repeating the processing of steps S60 to S66 until there are no unprocessed "question" messages among the messages in FIG.

In FIGS. 13 to 15, the process of identifying, in step S62, an "answer" message containing similar content to the "question" message extracted in step S60 has been described, but as shown in FIG. A "question" message similar to the "question" message (similar question message) may be specified.

FIG. 16 is a flowchart of an example of the processing procedure of step S62. In step S120, the allocation unit 42 selects messages in a predetermined range (for example, the first 10 messages and the next 20 messages) based on the order of the "question" messages extracted in step S60 from the processed messages. Extract.

In step S122, the allocation unit 42 arranges the message extracted in step S120 in a semantic space as shown in FIG. In step S124, the assignment unit 42 identifies an "answer" message that exists in a position close to the "question" message extracted in step S60 on the semantic space of FIG. 15.

In step S126, the allocation unit 42 can identify a "question" message that exists in a position close to the "question" message extracted in step S60 on the semantic space of FIG. 15 as a similar question message.

In this manner, according to the process of step S62 shown in FIG. 16, a pair of messages in which an answer message is assigned to a question message can be presented to the administrator as similar question messages.

Further, returning to step S20 in FIG. 4, the output unit 44 outputs the "question" message and the "answer" message assigned by the assignment unit 42 as a pair of messages.

According to the information processing system 1 according to the present embodiment, a chat log of a chat service in which a questioner can easily register a question message and a responder can respond to the question message with an answer message can be used as knowledge. It becomes easier to use.

For example, in the information processing system 1 according to the present embodiment, the work required by the administrator to use chat logs as knowledge can be reduced. Further, in the information processing system 1 according to the present embodiment, it is possible to reduce the message processing work required to disclose a question and answer message pair exchanged in a chat as knowledge.

[Other embodiments]
Although FIG. 1 describes an example in which the information processing device 10 provides the chat function, the configuration may include a chat service device 16 that provides the chat function, as in the information processing system 1A of FIG. 17.

FIG. 17 is a configuration diagram of an example of an information processing system according to this embodiment. The chat service device 16 shown in FIG. 17 provides a chat function to a plurality of users operating the user terminal 14, and also stores a chat log. Additionally, the chat service device 16 provides chat logs to the information processing device 10. The chat service device 16 may be implemented using a single computer or multiple computers, or may be implemented using a cloud service.

Note that the information processing device 10 uses the chat log provided from the chat service device 16 to perform the same processing as in the first embodiment.

FIG. 18 is a functional configuration diagram of an example of an information processing system according to this embodiment. The functional configuration of FIG. 18 is the same as the functional configuration of FIG. 3 except for some parts, so the description of the same parts will be omitted as appropriate.

For example, the chat service device 16 implements the functional configuration shown in FIG. 18 by executing the OS and programs. The chat service device 16 in FIG. 18 includes a chat processing section 80, a log information storage section 82, and a log information providing section 84. Further, the information processing device 10 realizes the functional configuration shown in FIG. 18 by executing the OS and programs. The information processing device 10 in FIG. 18 includes a log information storage section 32, an initial setting section 34, an extraction section 36, a classification section 38, a processing section 40, an allocation section 42, an output section 44, an allocation result storage section 46, and an operation reception section 48. , an allocation result update section 50, and a log information acquisition section 90.

The chat processing unit 80 provides a chat function to a plurality of users who operate the user terminal 14, and allows the plurality of users to exchange messages written in text or the like. The log information storage unit 82 stores messages exchanged by text or the like between multiple users as a chat log. The log information providing unit 84 provides the information processing device 10 with the chat log stored in the log information storage unit 82 .

The log information acquisition unit 90 of the information processing device 10 acquires a chat log from the chat service device 16 and stores it in the log information storage unit 32. The other configurations are the same as the functional configuration in FIG. 3, so the explanation will be omitted.

As described above, according to this embodiment, chat logs can be used more efficiently as knowledge.

Each function of the embodiments described above can be realized by one or more processing circuits. Here, the term "processing circuit" as used herein refers to a processor programmed to execute each function by software, such as a processor implemented by an electronic circuit, or a processor designed to execute each function explained above. This includes devices such as ASICs (Application Specific Integrated Circuits), DSPs (digital signal processors), FPGAs (field programmable gate arrays), and conventional circuit modules.

The devices described in the Examples are merely illustrative of one of several computing environments for implementing the embodiments disclosed herein. The present invention is not limited to this embodiment, and the components in this embodiment include those that can be easily conceived by those skilled in the art, those that are substantially the same, and those that are within the so-called equivalent range. . Furthermore, various omissions, substitutions, changes, and combinations of components can be made without departing from the gist of the present embodiment.

Aspects of the present invention are, for example, as follows.
<1>
Extracting means for extracting log information of messages written by multiple users for communication;
Classifying means for classifying at least question messages and answer messages from the messages included in the log information using predetermined conditions for classifying the messages;
Assigning means for assigning the answer message to the question message based on the classified result;
Output means for outputting the assigned message of the question and the message of the answer as a pair of messages;
An information processing device having:
<2>
<1> wherein the classification means classifies the question message, the answer message, and other messages from the message using a learned classification model that has learned the characteristics of the question message and the answer message. The information processing device described.
<3>
The assignment means assigns the answer message to the question message based on the classified result and the parent-child relationship information of the message included in the log information, as described in <1> or <2> above. information processing equipment.
<4>
The assignment means according to <1> or <2> above assigns the answer message to the question message based on the classification result and the similarity of the message included in the log information. Information processing device.
<5>
The assignment means assigns the answer message to the question message based on the classification result, the parent-child relationship information of the message included in the log information, and the similarity of the message. 1> or the information processing device described in <2>.
<6>
processing means for removing personal information, removing unnecessary sentences, and forming sentences on the question message and the answer message;
The information processing device according to any one of <1> to <5>, further comprising:
<7>
The information processing device according to any one of <1> to <6>, wherein the classification means classifies the messages included in the log information using a plurality of types of machine-learned classification models. .
<8>
The information processing device according to any one of <1> to <7>, wherein the classification means further classifies messages included in the log information based on urgency and field.
<9>
The information processing device according to any one of <1> to <8>, wherein the output means outputs the pair of messages as learning data for an application program.
<10>
In the information processing device,
Extraction procedure for extracting log information of messages written by multiple users for communication;
a classification procedure for classifying at least question messages and answer messages from the messages included in the log information using predetermined conditions for classifying the messages;
an assignment step of assigning the answer message to the question message based on the classified results;
an output procedure for outputting the assigned message of the question and the message of the answer as a pair of messages;
A program to run.
<11>
An information processing system in which an information processing device and an administrator terminal are communicably connected,
The information processing device includes:
Extracting means for extracting log information of messages written by multiple users for communication;
Classifying means for classifying at least question messages and answer messages from the messages included in the log information using predetermined conditions for classifying the messages;
Assigning means for assigning the answer message to the question message based on the classified result;
Output means for outputting the assigned message of the question and the message of the answer as a pair of messages;
has
The administrator terminal is
An information processing system comprising display means for displaying the pair of messages.
<12>
An information processing method executed by an information processing device, the method comprising:
an extraction step of extracting log information of messages written by multiple users for communication;
a classification step of classifying at least question messages and answer messages from the messages included in the log information using predetermined conditions;
an assigning step of assigning the answer message to the question message based on the classified result;
an output step of outputting the assigned message of the question and the message of the answer as a pair of messages;
An information processing method having
<13>
The classification step is the step of classifying the question message, the answer message, and other messages from the message using a learned classification model that has learned the characteristics of the question message and the answer message. Information processing method described.
<14>
The assignment step includes assigning the answer message to the question message based on the classified result and the parent-child relationship information of the message included in the log information, as described in <12> or <13> above. information processing methods.
<15>
The assigning step includes assigning the answer message to the question message based on the classification result and the similarity of the message included in the log information. Information processing method.
<16>
The assignment step includes assigning the answer message to the question message based on the classification result, the parent-child relationship information of the message included in the log information, and the similarity of the message. 12> or the information processing method described in <13>.
<17>
a processing step of removing personal information, removing unnecessary sentences, and forming sentences for the question message and the answer message;
The information processing method according to any one of <12> to <16>, further comprising:
<18>
The information processing method according to any one of <12> to <17>, wherein the classification step uses multiple types of machine-learned classification models to classify the messages included in the log information. .
<19>
The information processing method according to any one of <12> to <18>, wherein the classifying step further classifies messages included in the log information based on urgency and field.
<20>
The information processing method according to any one of <12> to <19>, wherein the output step outputs the paired messages as learning data for an application program.

1 Information processing system 10 Information processing device 12 Administrator terminal 14 User terminal 18 Network 36 Extraction unit 38 Classification unit 40 Processing unit 42 Allocation unit 44 Output unit 60 Display unit

JP 2021-163167 Publication

Claims (20)

Extracting means for extracting log information of messages written by multiple users for communication;
Classifying means for classifying at least question messages and answer messages from the messages included in the log information using predetermined conditions for classifying the messages;
Assigning means for assigning the answer message to the question message based on the classified result;
Output means for outputting the assigned message of the question and the message of the answer as a pair of messages;
An information processing device having: 2. The classifying means classifies the question message, the answer message, and other messages from the message using a learned classification model that has learned the characteristics of the question message and the answer message. information processing equipment. The information processing according to claim 1 or 2, wherein the assignment means assigns the answer message to the question message based on the classified result and parent-child relationship information of the message included in the log information. Device. The information processing apparatus according to claim 1 or 2, wherein the assignment means assigns the answer message to the question message based on the classification result and the similarity of the message included in the log information. . 2. The assigning means assigns the answer message to the question message based on the classification result, parent-child relationship information of the message included in the log information, and similarity of the message. 2. The information processing device according to 1 or 2. processing means for removing personal information, removing unnecessary sentences, and forming sentences on the question message and the answer message;
The information processing device according to claim 1 or 2, further comprising: 3. The information processing apparatus according to claim 1, wherein the classification means uses a plurality of types of machine-learned classification models to classify the messages included in the log information. 3. The information processing apparatus according to claim 1, wherein said classification means further classifies messages included in said log information based on urgency and field. 3. The information processing apparatus according to claim 1, wherein the output means outputs the paired messages as learning data for an application program. In the information processing device,
Extraction procedure for extracting log information of messages written by multiple users for communication;
a classification procedure for classifying at least question messages and answer messages from the messages included in the log information using predetermined conditions for classifying the messages;
an assignment step of assigning the answer message to the question message based on the classified results;
an output procedure for outputting the assigned message of the question and the message of the answer as a pair of messages;
A program to run. An information processing system in which an information processing device and an administrator terminal are communicably connected,
The information processing device includes:
Extracting means for extracting log information of messages written by multiple users for communication;
Classifying means for classifying at least question messages and answer messages from the messages included in the log information using predetermined conditions for classifying the messages;
Assigning means for assigning the answer message to the question message based on the classified result;
Output means for outputting the assigned message of the question and the message of the answer as a pair of messages;
has
The administrator terminal is
An information processing system comprising display means for displaying the pair of messages. An information processing method executed by an information processing device, the method comprising:
an extraction step of extracting log information of messages written by multiple users for communication;
a classification step of classifying at least question messages and answer messages from the messages included in the log information using predetermined conditions;
an assigning step of assigning the answer message to the question message based on the classified result;
an output step of outputting the assigned message of the question and the message of the answer as a pair of messages;
An information processing method having 13. The classifying step uses a learned classification model that has learned the characteristics of the question message and the answer message to classify the question message, answer message, and other messages from the message. information processing methods. The information processing according to claim 12 or 13, wherein the assigning step assigns the answer message to the question message based on the classified result and parent-child relationship information of the message included in the log information. Method. The information processing method according to claim 12 or 13, wherein the assigning step assigns the answer message to the question message based on the classified result and the similarity of the message included in the log information. . The assigning step assigns the answer message to the question message based on the classification result, parent-child relationship information of the message included in the log information, and similarity of the message. 13. The information processing method according to 12 or 13. a processing step of removing personal information, removing unnecessary sentences, and forming sentences for the message of the question and the message of the answer;
The information processing method according to claim 12 or 13, further comprising: 14. The information processing method according to claim 12, wherein the classifying step classifies the messages included in the log information using a plurality of types of machine-learned classification models. 14. The information processing method according to claim 12, wherein said classifying step further classifies messages included in said log information based on urgency and field. 14. The information processing method according to claim 12, wherein the output step outputs the paired messages as learning data for an application program.

Images