KR102499198B1 - Chatbot service providing system for considering user personaand method thereof - Google Patents

Abstract

A user-customized chatbot service system and method are provided. The method, in a user-customized chatbot service method performed by a chatbot management server, includes receiving conversation data including a query from a user terminal through a chatbot service; generating analysis data by analyzing conversation data including the query; Calculating user understanding using the analysis data; generating search data based on the analysis data; and generating dialogue data including an answer when the search data satisfies the level of understanding of the user. can include

Description

User-customized chatbot service system and its method {CHATBOT SERVICE PROVIDING SYSTEM FOR CONSIDERING USER PERSONAAND METHOD THEREOF}

The present invention relates to a user-customized chatbot service system and method, and in particular, to field workers such as banking, medical, insurance, law, construction, etc., by analyzing the knowledge necessary for field work by situation, field, and individual, It relates to a user-customized chatbot service system and method capable of providing information corresponding to the level of technology.

In general, a chatbot (CHATBOT) is a technology that uses natural language processing to provide a service as if a computer and a person have a conversation with a person. It is a technique to obtain information through

For example, the machine can process services such as telling the weather, movie reservations, and restaurant reservations as if talking to a person.

In addition, chatbots have become a service at the level of providing predetermined answers according to pre-entered algorithms, but with the development of natural language analysis and processing technology along with big data processing technology, they are providing optimal answers in consideration of various variables.

In other words, the chatbot service has made rapid progress with big data analysis, machine learning, and natural language processing technology, inferring the correct answer based on the user's conversation content analysis and predicting the next question to provide a conversation service. Beyond conversation services, it is expanding to handle various operations such as shopping and payment.

The matters described as the background art above are only for improving understanding of the background of the present invention, and should not be taken as an admission that they correspond to prior art already known to those skilled in the art.

Republic of Korea Patent Publication No. 10-2018-0098455

The problem to be solved by the present invention is to provide a user-customized chatbot service system and method.

The problems to be solved by the present invention are not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the description below.

In the user-customized chatbot service method according to an embodiment of the present invention for solving the above problems, in the user-customized chatbot service method performed by the chatbot management server, the conversation data including the query from the user terminal through the chatbot service receiving; generating analysis data by analyzing conversation data including the query; Calculating user understanding using the analysis data; generating search data based on the analysis data; and generating dialogue data including an answer when the search data satisfies the level of understanding of the user. can include

In an embodiment of the present invention, in the generating of the analysis data, the analysis data may be generated by analyzing the conversation data for each situation, each field, and each individual.

In one embodiment of the present invention, the generating of the analysis data may include generating word extraction data by analyzing the dialogue data; generating sentence extraction data by analyzing the dialogue data; generating word analysis data using the word extraction data; generating sentence analysis data using the sentence extraction data; and generating the analysis data by comparing and analyzing the word analysis data and the sentence analysis data. can include

In one embodiment of the present invention, the generating of the word extraction data may include: extracting words by analyzing the conversation data for each situation; extracting words by analyzing the conversation data for each field; and extracting words by analyzing the dialogue data for each individual. can include

In one embodiment of the present invention, the generating of the sentence extraction data may include extracting sentences by analyzing the conversation data for each situation; extracting sentences by analyzing the conversation data for each field; and extracting sentences by analyzing the dialogue data for each individual. can include

In one embodiment of the present invention, before generating the word extraction data, pre-processing the conversation data; may further include.

In one embodiment of the present invention, the step of preprocessing the conversation data is a tokenization operation using a space removal filter and a special character removal filter so that words included in the conversation data are divided into minimum units of meaningful words. performing steps; after the tokenization task is completed, performing a refinement task so that the meanings of the remaining words are highlighted by removing noise data for words that appear in the conversation data with a low frequency or words that are repeated many times; and normalizing the conversation data after the refining operation is completed. can include

In one embodiment of the present invention, the pre-processing of the dialogue data may include: converting the dialogue data into text when the dialogue data is voice; may further include.

In one embodiment of the present invention, the step of calculating the user's understanding may calculate the user's understanding by evaluating discipline, cooperativeness, activeness, practical ability, guidance-assistance, and responsibility using the analysis data. .

In one embodiment of the present invention, in the generating of the search data, the search data corresponding to the analysis data may be generated by searching for data stored in the learning data.

In one embodiment of the present invention, generating the learning data by iterative learning based on big data; may further include.

In one embodiment of the present invention, updating the learning data in real time in response to the conversation data including the answer; may further include.

In one embodiment of the present invention, when the search data is not satisfied with the level of understanding of the user, receiving additional data from the user terminal; can include

In one embodiment of the present invention, automatically checking and converting typos included in the dialogue data; may further include.

In addition, a user-customized chatbot service method according to another embodiment of the present invention for solving the above problems includes receiving, by a chatbot management server, conversation data including user information and queries from a user terminal through a chatbot service; generating analysis data by analyzing conversation data including the query based on the user information by the chatbot management server; Calculating, by the chatbot management server, a level of understanding of the user using the analysis data based on the user information; generating search data by the chatbot management server searching for learning data based on the analysis data; and determining, by the chatbot management server, whether the search data satisfies the level of understanding of the user, and generating dialogue data including an answer. can include

In one embodiment of the present invention, generating the learning data by the manager terminal; Further, the generating of the learning data may include: collecting big data by the manager terminal; generating pre-processed data by the manager terminal by pre-processing the big data and converting the big data into classifiable data; The preprocessed data preprocessed by the manager terminal is Machine Generated Data, Semi Generated Data, Human Generated Data, Enterprise Specialized Data, Enterprise Universal Data ( Enterprise Normalized Data) and enterprise specialized vocabulary data (Enterprise Specialized Vocabulary Data) to classify by task, individual, and company to create classification data; and generating the learning data by iteratively learning the classification data, by the manager terminal. can include

In one embodiment of the present invention, the group manager terminal inspecting the learning data; may further include.

In one embodiment of the present invention, in the step of generating the conversation data including the answer, if the search data satisfies the level of understanding of the user, the chatbot management server converts the search data into conversation data including the answer. And if the search data is not satisfied with the level of understanding of the user, the chatbot management server may request additional data from the user terminal.

In addition, a user-customized chatbot service system according to another embodiment of the present invention for solving the above problems includes a chatbot management server that provides a real-time chatbot service to a user terminal; Including, the chatbot management server analyzes the conversation data including the query received from the user terminal to generate analysis data and user understanding, and then extracts and searches data from the learning data using the generated analysis data Data is generated, and it is determined whether or not the created search data satisfies the level of understanding of the user, and conversation data including an answer may be generated and transmitted to the user terminal.

In addition, a user-customized chatbot service system according to another embodiment of the present invention for solving the above problems includes a chatbot management server and an administrator terminal providing a real-time chatbot service to a user terminal, the chatbot management server Analysis data and user understanding are generated by analyzing conversation data including queries received from user terminals, and then search data is generated and generated using learning data received from the manager terminal using the generated analysis data. It is possible to determine whether the retrieved search data satisfies the level of understanding of the user, generate dialogue data including an answer, and transmit it to the user terminal.

And, the program according to an embodiment of the present invention is combined with a computer that is hardware and stored in a computer-readable recording medium to perform the user-customized chatbot service method.

Other specific details of the invention are included in the detailed description and drawings.

According to the present invention, based on the user's information using the chatbot service, analysis by situation, sector, and individual is provided to field workers such as banking, medical, insurance, law, construction, etc. to satisfy the user with the knowledge required for field work. It can be provided as an answer that can be understood.

According to the present invention, by using a portable device to field workers with a vulnerable external environment, through a chatbot service, the knowledge necessary for field work is searched for by situation, field, and individual and provided with information, thereby increasing the work efficiency of field workers Convenience and reliability can be enhanced while respecting the diversity of users at the same time.

The effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below.

1 is a diagram for explaining a user-customized chatbot service system according to an embodiment of the present invention.
FIG. 2 is a detailed diagram for explaining the user-customized chatbot service system shown in FIG. 1 .
3 is a diagram for explaining a user-customized chatbot service system according to another embodiment of the present invention.
4 is a diagram for explaining a user-customized chatbot service method according to an embodiment of the present invention.
FIG. 5 is a diagram for explaining a step of generating analysis data shown in FIG. 4 .
FIG. 6 is a diagram for explaining a method of calculating user understanding shown in FIG. 4 .
7 is a diagram for explaining a user-customized chatbot service method according to another embodiment of the present invention.

Advantages and features of the present invention, and methods of achieving them, will become clear with reference to the detailed description of the following embodiments taken in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, only these embodiments are intended to complete the disclosure of the present invention, and are common in the art to which the present invention belongs. It is provided to fully inform the person skilled in the art of the scope of the invention, and the invention is only defined by the scope of the claims.

Terminology used herein is for describing the embodiments and is not intended to limit the present invention. In this specification, singular forms also include plural forms unless specifically stated otherwise in a phrase. As used herein, "comprises" and/or "comprising" does not exclude the presence or addition of one or more other elements other than the recited elements. Like reference numerals throughout the specification refer to like elements, and “and/or” includes each and every combination of one or more of the recited elements. Although "first", "second", etc. are used to describe various components, these components are not limited by these terms, of course. These terms are only used to distinguish one component from another. Accordingly, it goes without saying that the first element mentioned below may also be the second element within the technical spirit of the present invention.

Unless otherwise defined, all terms (including technical and scientific terms) used in this specification may be used with meanings commonly understood by those skilled in the art to which the present invention belongs. In addition, terms defined in commonly used dictionaries are not interpreted ideally or excessively unless explicitly specifically defined.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a diagram for explaining a user-customized chatbot service system according to an embodiment of the present invention, and FIG. 2 is a detailed diagram for explaining the user-customized chatbot service system shown in FIG.

As shown in FIGS. 1 and 2 , a user-customized chatbot service system 1 according to an embodiment of the present invention may include a user terminal 10 and a chatbot management server 20 .

Here, the user terminal 10 and the chatbot management server 20 may be synchronized in real time using a wireless communication network to transmit and receive data. The wireless communication network may support various long-distance communication methods, for example, wireless LAN (WLAN), DLNA (Digital Living Network Alliance), WiBro (Wireless Broadband: Wibro), WiMAX (World Interoperability for Microwave Access: Wimax) ), GSM (Global System for Mobile communication), CDMA (Code Division Multi Access), CDMA2000 (Code Division Multi Access 2000), EV-DO (Enhanced Voice-Data Optimized or Enhanced Voice-Data Only), WCDMA (Wideband CDMA) , HSDPA (High Speed Downlink Packet Access), HSUPA (High Speed Uplink Packet Access), IEEE 802.16, Long Term Evolution (LTE), LTEA (Long Term Evolution-Advanced), broadband wireless mobile communication service (Wireless Mobile Broadband Service: Various communication methods such as WMBS), BLE (Bluetooth Low Energy), Zigbee, RF (Radio Frequency), and LoRa (Long Range) may be applied, but are not limited thereto, and various widely known wireless or mobile communications method may be applied.

The user terminal 10 may perform a chatbot service through communication with the chatbot management server 20 .

Specifically, the user terminal 10 may transmit and receive conversation data with the chatbot management server 20 in real time through the chatbot service.

For example, when the user terminal 10 transmits conversation data including inquiry contents to the chatbot management server 20, it receives conversation data including answer contents corresponding to the inquiry contents, or a conversation including inquiry contents. When data is received from the chatbot management server 20, conversation data including answer contents corresponding to the inquiry contents may be transmitted. That is, the user terminal 10 can operate with the chatbot management server 20 and a user belonging to a company or a user capable of asking/answering questions/answers for each job, individual, and company.

In this embodiment, the user terminal 10 is described as a first user terminal 11 and a second user terminal 12, but may be a plurality of user terminals 10.

In addition, the user terminal 10 may receive a query for additional data from the chatbot management server 20 while transmitting and receiving conversation data, and transmit a response to the query.

To this end, the user terminal 10 may generate dialogue data including questions/answers using a separate input tool. For example, the user terminal 10 uses at least one of a microphone, a touch screen, a mouse, a microphone, a keyboard, or a camera to query/request Conversation data including answers may be generated, but is not limited thereto. At this time, the conversation data may be composed of text or voice, but is not limited thereto, and may be composed of video or image.

The user terminal 10 may receive conversation data visually and audibly separately or simultaneously. For example, it may include a display capable of outputting symbols, characters, numbers, etc. on a screen according to an operating state, a lamp outputting color change or flickering, or a speaker outputting audio.

The user terminal 10 may proceed with membership registration by using user information in order to perform chatbot service with the chatbot management server 20, but is not limited thereto. Here, the user information may include name, gender, age, contact information, occupation, etc., but is not limited thereto, and information on the user's current situation may be input.

Depending on the embodiment, the user terminal 10 may generate and transmit a feedback signal for the chatbot service to the chatbot management server 20 after or while using the chatbot service. Unlike this, the user terminal 10 may transmit before using the chatbot service. In this case, the feedback signal may include, but is not limited to, information such as satisfaction, accuracy, and reliability of answer data for each task, individual, and company for the chatbot service.

Depending on the embodiment, the user terminal 10 may receive a feedback control signal corresponding to the feedback signal before, during, or after using the chatbot service. At this time, the feedback control signal may include improved or updated information of the chatbot service corresponding to the feedback signal. Here, the feedback control signal may include event information. Event information may include advertisements, discounts on chatbot services, or information on events, but is not limited thereto.

As such, the user terminal 10 may include various portable electronic communication devices that support communication with the chatbot management server 20 . For example, the user terminal 10 is a separate smart device, such as a smart phone, a personal digital assistant (PDA), a tablet, a wearable device, a watch type terminal, A desktop computer that may include various portable terminals such as Smart Glass, HMD (Head Mounted Display), etc.) and various IoT (Internet of Things) terminals, but is not portable, and It may include an electronic communication device such as a workstation computer.

In addition, the user terminal 10 may operate using an application program (application program or application) in the present disclosure, and such an application program may be downloaded from an external server or chatbot management server 20 through wireless communication. can

The chatbot management server 20 may include a data transmission/reception unit 22, a database unit 24, a monitoring unit 26, and a management control unit 28.

The data transmission/reception unit 22 may transmit/receive dialogue data including questions/answers with the user terminal 10 .

Also, the data transceiver 22 may receive user information from the user terminal 10 .

Depending on the embodiment, the data transceiver 22 may receive a feedback signal from the user terminal 10 and transmit a feedback control signal corresponding to the feedback signal to the user terminal 10 .

The database unit 24 may store data transmitted and received with the user terminal 10 through a wireless communication network.

The database unit 24 may store data supporting various functions of the chatbot management server 20 . The database unit 24 may store a plurality of application programs (application programs or applications) running in the chatbot management server 20, data for the operation of the chatbot management server 20, and commands. At least some of these application programs may be downloaded from an external server through wireless communication.

The monitoring unit 26 screens the operation status of the user terminal 10 by user manipulation, the operation status of the chatbot management server 20, and data transmitted and received between the user terminal 10 and the chatbot management server 20. can be monitored through That is, by checking the use state of the user terminal 10 in real time, it is possible to give the user more confidence by making the user's use more convenient.

The management control unit 28 executes a chatbot service with the user terminal 10, analyzes the conversation data received from the user terminal 10, generates analysis data and user understanding, and then uses the generated analysis data to generate learning data. Search data may be generated by extracting data from the search data, and conversation data may be generated and transmitted to the user terminal 10 by determining whether the generated search data satisfies the level of understanding of the user. At this time, the management control unit 28 has started to generate analysis data, user understanding, and search data for all conversation data received from the user terminal 10, but is not limited thereto.

Specifically, the management and control unit 28 may generate analysis data by analyzing conversation data by situation, field, and individual based on user information.

For example, words and sentences included in conversation data may be extracted and extracted for each situation, field, and individual to generate extracted data, and analysis data may be generated by analyzing the generated extracted data. That is, in response to banking, construction, medical care, insurance, law, etc., words can be extracted by situation, field, and individual to generate word extraction data, and word analysis data can be generated by analyzing the generated word extraction data. Sentence extraction data may be generated by extracting sentences by situation, field, or individual, and sentence analysis data may be generated by analyzing the generated sentence extraction data. Analysis data may be generated by merging the generated word analysis data and sentence analysis data. Accordingly, by analyzing each word and sentence and then comparing/analyzing and merging them, it is possible to clearly understand the intention of the query included in the conversation data and deliver an accurate answer with high understanding to the user.

In addition, the management control unit 28 may calculate the level of understanding of the user by using analysis data based on user information. At this time, the user's understanding may be evaluated for discipline, cooperativeness, activeness, practical ability, guidance-assistance, and responsibility, but is not limited thereto.

For example, the management and control unit 28 evaluates discipline, which evaluates whether the user adheres to set rules and performs work, cooperativeness, which evaluates whether the user gets along well with co-workers and related people, and whether the user voluntarily engages in work without waiting for work instructions. Proactiveness in doing things, practical ability to evaluate whether the assigned task is handled reliably, guidance-support to evaluate whether it is helpful to juniors, drivers or people related to the work, and responsibility to evaluate whether a given task or responsibility is carried out to the end. It can be used to calculate user understanding.

In addition, the management control unit 28 may generate search data corresponding to the generated analysis data by searching for data stored in the learning data. That is, the management control unit 28 may extract data corresponding to the dialogue data from the learning data and generate it as search data.

In addition, the management control unit 28 may generate conversation data in consideration of user understanding based on the generated search data and user information.

For example, if the search data satisfies the level of understanding of the user, the management control unit 28 may store the search data as conversation data and transmit the search data to the user terminal 10 .

In contrast, when the search data does not satisfy the level of understanding of the user, the management and control unit 28 may request additional data from the user terminal 10, and the search data may be repeatedly performed until the level of understanding of the user is satisfied.

At this time, the management control unit 28 may pre-process the dialogue data before generating the analysis data by analyzing the dialogue data.

Specifically, the management control unit 28 may pre-process conversation data by sequentially performing tokenization, cleaning, and normalization so that conversation data is accurately recognized. Here, if the dialogue data is voice, image, or video, it may be converted into text.

Specifically, the management control unit 28 may perform a tokenization operation of filtering conversation data composed of text. That is, the management control unit 28 may perform a tokenization operation by using a space removal filter and a special character removal filter so that words included in the text are divided into minimum units of meaningful words.

In addition, the management control unit 28 may perform a refining operation of removing noise data included in the tokenized conversation data. That is, the management and control unit 28 may remove noise data for words with a low frequency of occurrence or corresponding words that are repeated many times to perform a refinement operation so that the meanings of the remaining words are highlighted.

Depending on the embodiment, the management control unit 28 may digitize the noise data and determine whether or not to delete the noise data corresponding to the user information of the user terminal 10 .

In addition, the management control unit 28 may normalize the text data after the refining operation has been completed.

Depending on the embodiment, the management control unit 28 may automatically inspect and convert typos included in the text.

In addition, the management control unit 28 may generate learning data by repeatedly learning data using a deep learning technique or a machine learning technique based on big data. Here, big data may include various types of data such as text, voice, video, and image.

In addition, the management control unit 28 can generate highly accurate learning data by inspecting the generated learning data.

For example, the management control unit 286 may check whether answer data can be generated by analyzing a query word associated with 'search' and extracting a query sentence associated with the search from learning data.

In addition, the management control unit 28 may generate a feedback control signal capable of improving the chatbot service in response to the feedback signal received from the user terminal 10 .

Depending on the embodiment, the management control unit 28 may increase the use of the chatbot service of the user terminal 10 by generating a feedback control signal including event information.

Such a chatbot management server 20 may be implemented by a hardware circuit (eg, a CMOS-based logic circuit), firmware, software, or a combination thereof. For example, it may be implemented using transistors, logic gates, and electronic circuits in the form of various electrical structures.

3 is a diagram for explaining a user-customized chatbot service system according to another embodiment of the present invention.

Referring to FIG. 3 , a user-customized chatbot service system 2 according to another embodiment of the present invention may include a manager terminal 30 .

Except for the manager terminal 30 shown in FIG. 3 , it may have the same characteristics as the user-customized chatbot service system 1 shown in FIGS. 1 and 2 .

In FIG. 3 below, a detailed description of contents overlapping with those of FIGS. 1 and 2 may be omitted, and a description may focus on different points. Therefore, components performing the same functions as the user-customized chatbot service system 2 shown in FIG. 3 are given the same reference numerals as those in FIGS. 1 and 2, and detailed descriptions thereof are omitted.

First, the administrator terminal 30 may communicate with the user terminal 10 and the chatbot management server 20 using an application program (application program or application), and these applications may communicate with an external server through wireless communication. Alternatively, it may be downloaded from the chatbot management server 20.

The manager terminal 30 may transmit and receive data in real time synchronization with the user terminal 10 and the chatbot management server 20 using a wireless communication network.

The manager terminal 30 may include a signal transmission/reception unit 32, a memory unit 34, a display unit 36, and a terminal control unit 38.

The signal transmitting and receiving unit 32 may transmit and receive conversation data with the user terminal 10 or the chatbot management server 20 .

In addition, the signal transmitting and receiving unit 32 may transmit and receive learning data to and from the chatbot management server 20 .

According to the embodiment, a feedback signal may be received from the user terminal 10 and the chatbot management server 20, and a feedback control signal corresponding to the feedback signal may be transmitted to the user terminal 10 and the chatbot management server 20.

The memory unit 34 may store data transmitted and received between the user terminal 10 and the chatbot management server 20 through a wireless communication network.

The display unit 36 may monitor data transmitted and received between the user terminal 10, the chatbot management server 20, and the manager terminal 30 in real time through a screen.

The terminal control unit 38 may learn the learning data received from the chatbot management server 20 and execute the chatbot service with the user terminal 10 .

In addition, the terminal controller 38 may generate learning data based on big data.

Such manager terminal 30 may include various portable electronic communication devices that support communication with the user terminal 10 and the chatbot management server 20 . For example, as a separate smart device, a smart phone, a personal digital assistant (PDA), a tablet, a wearable device (eg, a watch type terminal (Smartwatch), a glass type terminal (Smart Glass), Head Mounted Display (HMD), etc.) and various Internet of Things (IoT) terminals, but are not limited thereto.

The operation of the user-customized chatbot service system according to an embodiment of the present invention having such a structure is as follows. 4 is a diagram for explaining a user-customized chatbot service method according to an embodiment of the present invention, FIG. 5 is a diagram for explaining a step of generating analysis data shown in FIG. 4, and FIG. It is a diagram for explaining a method of calculating user understanding.

In this embodiment, the first user terminal 11 has user information of a man in his early 50s, male A, who joined the construction company for 10 years, and the second user terminal 12 has user information in his late 20s, male B, who joined the construction company 2 Although it has been disclosed as having yearly user information, it is not limited thereto.

First, the chatbot management server 20 may execute the chatbot service with the user terminal 10 requesting the chatbot service (S100).

Next, the chatbot management server 20 may receive user information (S110). Here, the user information may include name, gender, age, contact information, occupation, etc., but is not limited thereto, and information on the user's current situation may be input.

At this time, in order to execute the chatbot service, the user terminal 10 may advance membership registration through input of user information, but is not limited thereto.

Next, the chatbot management server 20 may receive conversation data including questions/answers from the user terminal 10 (S120).

For example, the chatbot management server 20 may receive conversation data from the user terminal 10, "How do you manage the site when it rains?"

Next, the chatbot management server 20 may generate analysis data by analyzing the conversation data received from the user terminal 10 (S130).

Specifically, referring to FIG. 5 , the chatbot management server 20 may pre-process conversation data received from the user terminal 10 (S200).

For example, the chatbot management server 20 may pre-process conversation data by sequentially performing tokenization, cleaning, and normalization so that conversation data is accurately recognized. Here, if the dialogue data is voice, image, or video, it may be converted into text.

For example, the chatbot management server 20 may perform a tokenization operation by using a space removal filter and a special character removal filter so that words included in text are divided into minimum units of meaningful words.

In addition, the chatbot management server 20 removes noise data for words that rarely appear in conversation data or words that are repeated many times after the tokenization work is finished, and refines the meaning of the remaining words to stand out. can At this time, the chatbot management server 20 may determine whether or not to delete the noise data in response to the user information of the user terminal 10 by digitizing the noise data.

In addition, the chatbot management server 20 may normalize the text data for which the refining work has been completed. At this time, the chatbot management server 20 may automatically inspect and convert typos included in the text.

Next, the chatbot management server 20 may generate word extraction data by extracting words from the preprocessed conversation data (S210).

Specifically, word extraction data may be generated by extracting words included in conversation data by situation, field, and individual.

For example, the chatbot management server 20 receives 'rain', 'construction', 'site', The words 'site management' and 'management' can be extracted. That is, the words 'non' and 'site' are extracted for each situation under the premise that it is a construction site among banking, construction, medical care, insurance, and law, and the words 'construction', 'site', and 'site management' are extracted for each field. and extract the word 'management' for each individual.

Next, the chatbot management server 20 may generate word analysis data by analyzing the extracted word extraction data (S220).

Next, the chatbot management server 20 may generate sentence extraction data by extracting sentences from the preprocessed conversation data (S230).

Specifically, the chatbot management server 20 may generate sentence extraction data by extracting sentences included in conversation data by situation, field, and individual.

For example, the chatbot management server 20 extracts all sentences at once for each sentence among several sentences received from the user terminal 10, and then extracts them for each situation, field, and individual to generate sentence extraction data can do.

Next, the chatbot management server 20 may generate sentence analysis data by analyzing using the extracted sentence extraction data (S240).

Next, the chatbot management server 20 may generate analysis data by comparing/analyzing and merging word analysis data and sentence analysis data (S250).

Next, the chatbot management server 20 may calculate the user's understanding by using the analysis data based on the user information.

Specifically, the chatbot management server 20 evaluates discipline, which evaluates whether the user observes set rules and performs tasks, cooperativeness, which evaluates whether the user gets along well with colleagues and related people, and whether the user voluntarily engages in work without waiting for work instructions. Proactiveness in doing things, practical ability to evaluate whether the assigned task is handled reliably, guidance-support to evaluate whether it is helpful to juniors, drivers or people related to the work, and responsibility to evaluate whether a given task or responsibility is carried out to the end. It can be used to calculate user understanding.

For example, referring to FIG. 6 , the first user terminal 11 scored 2 points for discipline, 5 points for cooperativeness, 5 points for proactiveness, 5 points for practical performance, 4 points for guidance and assistance, and 4 points for responsibility, out of 30 points. With a user understanding of 25 points, the second user terminal 12 scored 22 points out of 30, with 5 points for discipline, 4 points for cooperativeness, 4 points for proactiveness, 1 point for practical performance, 3 points for guidance and assistance, and 5 points for responsibility. of user comprehension. That is, the first user terminal 11 of male A, who is in his early 50s and has worked for a construction company for 10 years, does not follow the rules well, but has a high degree of user understanding due to his cooperativeness, initiative, ability to perform practical tasks, guidance-assistance, and responsibility. Able to know. In addition, it can be seen that the second user terminal 20, male B in his late twenties, in the second year of employment at a construction company, has excellent discipline, cooperativeness, initiative, guidance-assistance, and responsibility, but has low practical performance skills, resulting in low user understanding. .

Next, the chatbot management server 20 may generate search data by extracting data from the learning data using the generated analysis data (S150).

Specifically, the chatbot management server 20 may extract data corresponding to conversation data from learning data and generate search data.

Here, the learning data may be generated by the chatbot management server 20 repeatedly learning data using a deep learning technique or a machine learning technique based on big data.

Next, the chatbot management server 20 determines whether the generated search data satisfies the user's level of understanding (S160), and if the search data satisfies the level of user's level of understanding, the chatbot management server 20 stores the searched data as dialogue data to provide the user terminal 10 ) can be transmitted (S170).

Finally, the chatbot management server 20 may terminate the chatbot service when the user terminal 10 requests termination of the chatbot service (S180).

On the other hand, the chatbot management server 20 determines whether the generated search data satisfies the level of understanding of the user (S160), and if the search data does not satisfy the level of understanding of the user, it may request additional data from the user terminal 10 (S160). S190). That is, the chatbot management server 20 may repeatedly perform the above steps until the search data satisfies the level of understanding of the user.

7 is a diagram for explaining a user-customized chatbot service method according to another embodiment of the present invention.

In FIG. 7, a case in which learning data is generated by the user terminal 30 is limited and described. Accordingly, steps S310 of transmitting dialog data, step S320 of generating dialog data, and step S330 of receiving dialog data may be the same as those shown in FIGS. 4 and 5, and thus detailed descriptions thereof are omitted.

First, as shown in FIG. 7, the user terminal 30 may generate learning data (S310). That is, the user terminal 30 may generate learning data by iterative learning using a deep learning technique or a machine learning technique based on big data.

Specifically, the user terminal 30 may collect various data for each job, individual, and company based on big data. Here, big data may include various types of data such as text, voice, video, and image.

Next, the user terminal 30 may pre-process the collected data and convert it into classifiable data.

For example, the user terminal 30 may convert data into classifiable data through RPA (Robot Process Automation) pre-processing, but is not limited thereto.

Next, the user terminal 30 may classify the collected data. For example, it may be classified according to business, individual, and company, but is not limited thereto.

Specifically, the user terminal 30 converts the collected data into Machine Generated Data, Semi Generated Data, Human Generated Data, Enterprise Specialized Data, Data can be classified by business, individual, and company by type, including Enterprise Normalized Data and Enterprise Specialized Vocabulary Data.

Next, the chatbot management server 20 may receive conversation data from the user terminal 10 through the chatbot service (S310).

Next, the chatbot management server 20 may generate conversation data including answers by classifying and selecting learning data corresponding to conversation data including queries (S320). At this time, learning data may be received from the user terminal 30 .

Next, the user terminal 10 may generate a feedback signal by determining the conversation data received from the chatbot management server 20 (S330 and S340). At this time, the feedback signal may include information such as satisfaction, accuracy, and reliability of the answer through the chatbot service.

Next, the chatbot management server 20 may generate a feedback control signal capable of improving the chatbot service in response to the feedback signal received from the user terminal 10 (S350).

Depending on the embodiment, the user terminal 10 may receive a feedback control signal including improved or updated information of the chatbot service in response to the feedback signal before, during, or after using the chatbot service. At this time, the feedback control signal may include event information. Event information may include advertisements, discounts on chatbot services, or information on events, but is not limited thereto.

Depending on the embodiment, the manager terminal 30 may generate a feedback control signal capable of improving the chatbot service in response to the feedback signal.

Next, the manager terminal 30 may update learning data in real time using conversation data (S360).

Depending on the embodiment, the manager terminal 30 may update learning data in real time in response to a feedback signal received from the user terminal 10 .

Unlike this, the chatbot management server 20 may update learning data in real time in response to conversation data and feedback signals.

Finally, the manager terminal 30 may monitor data transmitted and received between the user terminal 10 and the chatbot management server 20 in real time.

Steps of a method or algorithm described in connection with an embodiment of the present invention may be implemented directly in hardware, implemented in a software module executed by hardware, or implemented by a combination thereof. A software module may include random access memory (RAM), read only memory (ROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, hard disk, removable disk, CD-ROM, or It may reside in any form of computer readable recording medium well known in the art to which the invention pertains.

Although the embodiments of the present invention have been described with reference to the accompanying drawings, those skilled in the art to which the present invention pertains can be implemented in other specific forms without changing the technical spirit or essential features of the present invention. you will be able to understand Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive.

1, 2: User-customized chatbot service system
10: user terminal
20: Chatbot management server
30: manager terminal

Claims (21)

In the user-customized chatbot service method performed by the chatbot management server,
Receiving conversation data including a query from a user terminal through a chatbot service;
generating analysis data by analyzing conversation data including the query;
Calculating user understanding using the analysis data;
generating search data based on the analysis data; and
generating dialogue data including an answer when the search data satisfies the level of understanding of the user; including,
The step of calculating the user understanding is,
A user-customized chatbot service method that calculates the user's understanding by evaluating discipline, cooperation, initiative, practical performance, guidance-assistance, and responsibility using the analysis data. According to claim 1,
The step of generating the analysis data,
A user-customized chatbot service method for generating the analysis data by analyzing the conversation data by situation, field, and individual. According to claim 2,
The step of generating the analysis data,
generating word extraction data by analyzing the dialogue data;
generating sentence extraction data by analyzing the dialogue data;
generating word analysis data using the word extraction data;
generating sentence analysis data using the sentence extraction data; and
generating the analysis data by comparing and analyzing the word analysis data and the sentence analysis data; Including, user-customized chatbot service method. According to claim 3,
The step of generating the word extraction data,
extracting words by analyzing the conversation data for each situation;
extracting words by analyzing the conversation data for each field; and
extracting words by analyzing the dialogue data for each individual; Including, user-customized chatbot service method. According to claim 3,
The step of generating the sentence extraction data,
extracting sentences by analyzing the conversation data for each situation;
extracting sentences by analyzing the conversation data for each field; and
extracting sentences by analyzing the dialogue data for each individual; Including, user-customized chatbot service method. According to claim 3,
Prior to the step of generating the word extraction data,
pre-processing the dialogue data; Further comprising a, user-customized chatbot service method According to claim 6,
The step of pre-processing the conversation data,
performing a tokenization operation using a space removal filter and a special character removal filter so that words included in the conversation data are divided into minimum units of meaningful words;
after the tokenization task is completed, performing a refinement task so that the meanings of the remaining words are highlighted by removing noise data for words that appear in the conversation data with a low frequency or words that are repeated many times; and
normalizing the conversation data after the refining operation is completed; Including, user-customized chatbot service method. According to claim 6,
The step of pre-processing the conversation data,
converting the dialogue data into text when the dialogue data is voice; Further comprising a, user-customized chatbot service method. delete According to claim 1,
Generating the search data,
A user-customized chatbot service method for generating the search data corresponding to the analysis data by searching for data stored in the learning data. According to claim 10,
generating the learning data by iterative learning based on big data; Further comprising a, user-customized chatbot service method. According to claim 11,
updating the learning data in real time in response to the dialogue data including the answers; Further comprising a, user-customized chatbot service method. According to claim 1,
receiving additional data from the user terminal when the search data does not satisfy the level of understanding of the user; Including, user-customized chatbot service method. According to claim 7,
The step of pre-processing the conversation data,
automatically checking and converting typos included in the conversation data; Further comprising a, user-customized chatbot service method. receiving, by a chatbot management server, conversation data including user information and queries from a user terminal through a chatbot service;
generating analysis data by analyzing conversation data including the query based on the user information by the chatbot management server;
Calculating, by the chatbot management server, a level of understanding of the user using the analysis data based on the user information;
generating search data by the chatbot management server searching for learning data based on the analysis data; and
the chatbot management server determining whether the search data satisfies the level of understanding of the user and generating conversation data including an answer; including,
The step of calculating the user understanding is,
A user-customized chatbot service method that calculates the user's understanding by evaluating discipline, cooperation, initiative, practical performance, guidance-assistance, and responsibility using the analysis data. According to claim 15,
Generating, by a manager terminal, the learning data; Including more,
The step of generating the learning data,
Collecting big data by the manager terminal;
generating pre-processed data by the manager terminal by pre-processing the big data and converting the big data into classifiable data;
The manager terminal converts the preprocessed data into machine generated data, semi-automated data, human generated data, enterprise specialized data, and enterprise universal data. Normalized Data) and enterprise specialized vocabulary data (Enterprise Specialized Vocabulary Data) classifying by task, individual, and company to generate classification data; and
generating the learning data by iteratively learning the classification data, by the manager terminal; Including, user-customized chatbot service method. According to claim 16,
inspecting the learning data by the manager terminal; Further comprising a, user-customized chatbot service method. According to claim 15,
In the step of generating conversation data including the answer,
When the search data satisfies the level of understanding of the user, the chatbot management server generates the search data as conversation data including the answer,
If the search data is not satisfied with the level of understanding of the user, the chatbot management server requests additional data from the user terminal. A chatbot management server providing a real-time chatbot service to user terminals; including,
The chatbot management server generates analysis data by analyzing conversation data including a query received from the user terminal, generates user understanding using the analysis data, and then extracts data from learning data using the analysis data. Extracting and generating search data, determining whether the search data satisfies the user's level of understanding, generating conversation data including answers, and transmitting the data to the user terminal;
The user's understanding is generated by evaluating discipline, cooperation, activeness, practical performance, guidance-assistance, and responsibility using the analysis data. Including a chatbot management server and an administrator terminal that provide real-time chatbot services to user terminals,
The chatbot management server analyzes the conversation data including the query received from the user terminal to generate analysis data, uses the analysis data to generate user understanding, and then uses the analysis data to receive from the manager terminal Search data is generated using the learned learning data, it is determined whether the search data satisfies the user's level of understanding, and conversation data including answers are generated and transmitted to the user terminal;
The user's understanding is generated by evaluating discipline, cooperation, activeness, practical performance, guidance-assistance, and responsibility using the analysis data. A computer program stored in a computer-readable recording medium to perform the method of any one of claims 1 to 8 and 10 to 18 by being combined with a computer that is hardware.

Images