KR20200141863A - System for providing personal recording data based ai chatting service - Google Patents

Abstract

Provided is a personal recording data-based artificial intelligence dialogue service providing system, which comprises: a user terminal which provides at least one type of personal record data to generate a character of an artificial intelligence dialogue agent, and conducts a dialogue with the artificial intelligence dialogue agent generated based on the personal record data; a collection unit which collects at least one type of personal record data from the user terminal; a decision unit which trains the collected personal record data to determine the character of the artificial intelligence dialogue agent; a retraining unit which retrains the artificial intelligence dialogue agent with a determined character; and an artificial intelligence dialogue service providing server which sets the artificial intelligence dialogue agent corresponding to the selected character to run on the user terminal when the character is selected from the user terminal.

Description

System for providing artificial intelligence dialogue service based on personal records data {SYSTEM FOR PROVIDING PERSONAL RECORDING DATA BASED AI CHATTING SERVICE}

The present invention relates to a system for providing an artificial intelligence conversation service based on personal records data, and provides a platform for learning from personal records data to communicate with a chat agent having a character and an identity.

Artificial Intelligence is revolutionizing the way business, organizational operations, lifestyles, and communication are done. Various informatization projects are underway to provide optimal services to the lifestyles of modern cultures that change rapidly every day and to the needs of customers that are constantly changing in a variety of ways. Among them, big data and deep learning-related technologies are in progress. Artificial intelligence technology, which has been developed and applied to real life in specific fields, has been implemented, and is also applied to intelligent personal services that provide and utilize information in various fields specialized to individual and analysis of specific data. Currently, the interaction between artificial intelligence and humans is limited, but in everyday natural language, that is, in the form of conversation. Although it is still in a rudimentary stage, various household appliances connected through a network are being controlled through a conversation through voice, and a search, query, and response for specific information is possible through a knowledge base to which deep learning is applied.

At this time, research and development on a method of forming a character to have an identity in an artificial intelligence chatbot or conversation agent was conducted. In this regard, Korean Patent Publication No. 2004-0107858 (December 23, 2004) In one public), when the content is provided through a chat with the chatbot, the method of providing the content is set differently according to the character of the character selected on the chatbot, and the content necessary for the conversation with the chatbot is collected and collected. By processing the processed content according to the character and transmitting the processed content, the user gets information through conversation and encounters the information in various ways, and each character has a unique identity, so it is not a simple program but a single character. A configuration is disclosed that can be recognized and approached more user-friendly.

However, the above-described characters are not imitating the person's personality, but are merely expressing one of the various aspects that a person has, so the person who is still talking to the chatbot feels that he is talking with the chatbot. . In other words, a person does not simply have one aspect, but has the complexity of possessing various aspects at the same time, and shows different personalities according to social status or family relations, so even if one character is formed, it is applied to all. , After all, you have no choice but to think that the other person is talking with the chatbot, not with people again.

An embodiment of the present invention is not limited to analyzing the object through personal record data, but also by analyzing the human relationship between the user and the object with whom the conversation is made, and constructing a database of each word, and determining the characteristics of each relationship. By creating a complex on the overall character, the conversation partner can give a sense of reality like talking with a real object, rather than a conversation with a chatbot, and further analyze the character of the deceased object and the character of each relationship to communicate. When creating an agent, it is possible to provide a method of providing an artificial intelligence conversation service based on personal records data that can provide an experience like having a conversation with a deceased person. However, the technical problem to be achieved by the present embodiment is not limited to the technical problem as described above, and other technical problems may exist.

As a technical means for achieving the above-described technical problem, an embodiment of the present invention provides at least one type of personal record data to generate a character of an artificial intelligence dialogue agent, and is created based on the personal record data. A user terminal that conducts a conversation with the intelligent conversation agent, and a collection unit that collects at least one type of personal record data from the user terminal, a determination unit that determines the character of the artificial intelligence conversation agent by learning the collected personal record data, An artificial intelligence dialogue service providing server including a relearning unit for retraining an artificial intelligence dialogue agent as a character, and a setting unit for setting an artificial intelligence dialogue agent corresponding to the selected character to be run on the user terminal when a character is selected at the user terminal Includes.

According to any one of the above-described problem solving means of the present invention, not only analyzing the object through personal record data, but also analyzing the human relationship between the user who is a conversation partner and the object, constructing each word database, and By creating the personality of each relationship on the overall character of the object, the person who communicates can give a sense of reality like talking with an actual object rather than the feeling of having a conversation with the chatbot, and furthermore, the character and relationship of the deceased object. When a conversation agent is created by analyzing individual characteristics, it is possible to provide an experience similar to conversation with the deceased counterpart.

1 is a diagram illustrating a system for providing an artificial intelligence conversation service based on personal record data according to an embodiment of the present invention.
FIG. 2 is a block diagram illustrating an artificial intelligence conversation service providing server included in the system of FIG. 1.
3 is a diagram for explaining an embodiment in which an artificial intelligence conversation service based on personal record data according to an embodiment of the present invention is implemented.
4 is a diagram illustrating a process of transmitting and receiving data between components included in the system for providing an artificial intelligence conversation service based on personal record data of FIG. 1 according to an embodiment of the present invention.
5 is a flowchart illustrating a method of providing an artificial intelligence conversation service based on personal record data according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those of ordinary skill in the art can easily implement the present invention. However, the present invention may be implemented in various different forms and is not limited to the embodiments described herein. In the drawings, parts irrelevant to the description are omitted in order to clearly describe the present invention, and similar reference numerals are assigned to similar parts throughout the specification.

Throughout the specification, when a part is said to be "connected" to another part, this includes not only "directly connected" but also "electrically connected" with another element interposed therebetween. . In addition, when a part "includes" a certain component, it means that other components may be further included, and one or more other features, not excluding other components, unless specifically stated to the contrary. It is to be understood that it does not preclude the presence or addition of any number, step, action, component, part, or combination thereof.

The terms "about", "substantially" and the like, as used throughout the specification, are used in or close to the numerical value when manufacturing and material tolerances specific to the stated meaning are presented, and are used in the sense of the present invention. To assist, accurate or absolute figures are used to prevent unfair use of the stated disclosure by unscrupulous infringers. As used throughout the specification of the present invention, the term "step (to)" or "step of" does not mean "step for".

In the present specification, the term "unit" includes a unit realized by hardware, a unit realized by software, and a unit realized using both. Further, one unit may be realized using two or more hardware, or two or more units may be realized using one hardware.

In this specification, some of the operations or functions described as being performed by the terminal, device, or device may be performed instead in a server connected to the terminal, device, or device. Likewise, some of the operations or functions described as being performed by the server may also be performed by a terminal, device, or device connected to the server.

In this specification, some of the operations or functions described as mapping or matching with the terminal means mapping or matching the unique number of the terminal or the identification information of the individual, which is the identification information of the terminal. Can be interpreted as.

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

1 is a diagram illustrating a system for providing an artificial intelligence conversation service based on personal records data according to an embodiment of the present invention. Referring to FIG. 1, the system 1 for providing an artificial intelligence dialogue service based on personal record data includes at least one user terminal 100, an artificial intelligence dialogue service providing server 300, and at least one object terminal 400. Can include. However, since the system 1 for providing an artificial intelligence conversation service based on personal record data of FIG. 1 is only an embodiment of the present invention, the present invention is not limitedly interpreted through FIG. 1.

In this case, each component of FIG. 1 is generally connected through a network 200. For example, as shown in FIG. 1, at least one user terminal 100 may be connected to an artificial intelligence conversation service providing server 300 through a network 200. In addition, the artificial intelligence conversation service providing server 300 may be connected to at least one user terminal 100 and at least one object terminal 400 through the network 200. In addition, at least one object terminal 400 may be connected to the artificial intelligence conversation service providing server 300 through the network 200.

Here, the network refers to a connection structure in which information exchange is possible between respective nodes such as a plurality of terminals and servers, and examples of such networks include RF, 3rd Generation Partnership Project (3GPP) network, and Long Term (LTE). Evolution) network, 5GPP (5th Generation Partnership Project) network, WIMAX (World Interoperability for Microwave Access) network, Internet, LAN (Local Area Network), Wireless LAN (Wireless Local Area Network), WAN (Wide Area Network) , Personal Area Network (PAN), Bluetooth (Bluetooth) network, NFC network, satellite broadcasting network, analog broadcasting network, Digital Multimedia Broadcasting (DMB) network, and the like, but are not limited thereto.

In the following, the term “at least one” is defined as a term including the singular number and the plural number, and even if the term “at least one” does not exist, each component may exist in the singular or plural, and may mean the singular or plural. It will be self-evident. In addition, it will be possible to change according to the embodiment that each component is provided in a singular or plural.

At least one user terminal 100 may be a terminal that provides personal records data to the artificial intelligence chat service providing server 300 using a web page, app page, program, or application related to an artificial intelligence chat service based on personal records data. have. In addition, the at least one user terminal 100 may be a terminal that transmits and receives a chat agent and a chat agent provided by the artificial intelligence chat service providing server 300.

Here, the at least one user terminal 100 may be implemented as a computer capable of accessing a remote server or terminal through a network. Here, the computer may include, for example, a navigation system, a notebook equipped with a web browser, a desktop, a laptop, and the like. In this case, the at least one user terminal 100 may be implemented as a terminal capable of accessing a remote server or terminal through a network. At least one user terminal 100, for example, as a wireless communication device that is guaranteed portability and mobility, navigation, PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular), PHS(Personal Handyphone System), PDA(Personal Digital Assistant), IMT(International Mobile Telecommunication)-2000, CDMA(Code Division Multiple Access)-2000, W-CDMA(W-Code Division Multiple Access), Wibro(Wireless Broadband Internet) ) All types of handheld-based wireless communication devices such as terminals, smartphones, smartpads, and tablet PCs may be included.

The artificial intelligence chat service providing server 300 may be a server that provides an artificial intelligence chat service web page, an app page, a program, or an application based on personal record data. In addition, the artificial intelligence conversation service providing server 300 may be a server that collects personal record data from the user terminal 100. In addition, the artificial intelligence conversation service providing server 300 may be a server that defines a relationship between a user of the user terminal 100 and an object that has generated personal record data, and analyzes a personality based on the definition. In addition, the artificial intelligence dialogue service providing server 300 may be a server that analyzes and stores a character of an object that has generated personal record data. In addition, the artificial intelligence conversation service providing server 300 may be a server that builds a word database based on characters and personalities, and learns and trains conversation agents. In addition, when the user terminal 100 selects a character of the conversation agent, the artificial intelligence conversation service providing server 300 may be a server that provides a conversation interface to transmit and receive a conversation with the selected character. In addition, the artificial intelligence dialog service providing server 300, when feedback of error data that denies a personality or character from the user terminal 100, retrains and retrains the personality or character of the dialog agent based on the feedback data. It could be a server.

Here, the artificial intelligence conversation service providing server 300 may be implemented as a computer that can access a remote server or terminal through a network. Here, the computer may include, for example, a navigation system, a notebook equipped with a web browser, a desktop, a laptop, and the like.

The at least one object terminal 400 may be a terminal that transmits personal records data to the artificial intelligence conversation service providing server 300 using a web page, app page, program, or application related to an artificial intelligence conversation service based on personal records data. have. In addition, at least one object terminal 400 is identified by an identifier input to the artificial intelligence dialog service providing server 300 from the user terminal 100, and responds to the request of the artificial intelligence dialog service providing server 300 or provides data It may be a terminal that feeds back.

Here, the at least one object terminal 400 may be implemented as a computer capable of accessing a remote server or terminal through a network. Here, the computer may include, for example, a navigation system, a notebook equipped with a web browser, a desktop, a laptop, and the like. At this time, the at least one object terminal 400 may be implemented as a terminal capable of accessing a remote server or terminal through a network. At least one object terminal 400, for example, as a wireless communication device that is guaranteed portability and mobility, navigation, PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular), PHS(Personal Handyphone System), PDA(Personal Digital Assistant), IMT(International Mobile Telecommunication)-2000, CDMA(Code Division Multiple Access)-2000, W-CDMA(W-Code Division Multiple Access), Wibro(Wireless Broadband Internet) ) All types of handheld-based wireless communication devices such as terminals, smartphones, smartpads, and tablet PCs may be included.

FIG. 2 is a block diagram illustrating an artificial intelligence dialog service providing server included in the system of FIG. 1, and FIG. 3 is an embodiment of implementing an artificial intelligence dialog service based on personal records data according to an embodiment of the present invention. It is a figure for explaining.

Referring to FIG. 2, the artificial intelligence conversation service providing server 300 includes a collection unit 310, a determination unit 320, a relearning unit 330, a setting unit 340, a relationship definition unit 350, and Study 360, may include a connection unit 370.

The artificial intelligence conversation service providing server 300 according to an embodiment of the present invention or another server (not shown) operating in conjunction with at least one user terminal 100 and at least one object terminal 400 When transmitting a data-based artificial intelligence conversation service application, program, app page, web page, etc., at least one user terminal 100 and at least one object terminal 400 may be an artificial intelligence conversation service application based on personal record data , Programs, app pages, web pages, etc. can be installed or opened. In addition, a service program may be driven in at least one user terminal 100 and at least one object terminal 400 by using a script executed in a web browser. Here, the web browser is a program that enables you to use the web (WWW: world wide web) service, and refers to a program that receives and displays hypertext described in HTML (hyper text mark-up language). For example, Netscape , Explorer, chrome, etc. In addition, the application means an application on the terminal, and includes, for example, an app that is executed on a mobile terminal (smartphone).

Referring to FIG. 2, the collection unit 310 may collect at least one type of personal record data from the user terminal 100. In this case, the user terminal 100 may provide at least one type of personal record data to generate a character of an artificial intelligence conversation agent. Here, the personal record data may include moving pictures, conversation history logs, images including photos, audio data, and form record data, but the types are not limited to those described above.

At this time, the conversation agent (Chatting Agent, or chatbot) is a platform through which humans and machines exchange information through voice or text, and was created to show the superficiality of communication between humans and machines, and uses pattern matching and alternative methodologies. Conversation can be configured, and it will provide a platform for automatic speech recognition technology that converts human speech content into text based on human dialogue, dialogue controller to manipulate interactions with humans, and information delivery methods to humans. I can. In this case, the conversation agent may physically analyze a user's query using an artificial intelligence algorithm in the artificial intelligence conversation service providing server 300, and the user may be provided with information through an app or web. In the artificial intelligence conversation service providing server 300, technologies such as natural language processing (NLP), context recognition, and big data analysis may be used to provide appropriate answers to user queries. In addition, a distributed computing technology such as Hadoop processing large-scale message traffic may be further used between the artificial intelligence conversation service providing server 300 and the user. In addition, the conversation agent can use a platform to which an ontology-based knowledge base is applied rather than a simple rule (pattern match, rule-based) basis, and natural language processing (NLP) and ontology-based search You can use technology to process commands.

The determination unit 320 may determine the character of the artificial intelligence conversation agent by learning the collected personal record data. First of all, for example, there is a unique character that characterizes a person's identity. In addition, the character is inherent and revealed throughout human relationships, but personality may appear differently for each human relationship. For example, assuming that the character is extroverted, this is revealed throughout the relationship, but the status of each relationship, i.e. at work, as a subordinate to a supervisor, as a supervisor to a subordinate, as a parent to a child, and as a child to parents. They may appear differently, and the language and tone used will be different. And, even with the same child, the relationship with the first child and the second child are different, so this can also appear differently. If the character A is set and the personality as a mother is defined, but the relationship with the second child is not defined, when the second child communicates with the character A, it may feel that something is not their mother. Accordingly, in the present invention, a character is defined in an upper node, a personality by human relationship is defined in the lower node, and even if the same human relationship is used in the lower node of the personality by human relationship, it is used according to each other. By separately defining the vocabulary and word database and the persona (persona), it is possible to avoid feeling that the person talking to the conversation agent is talking with the machine or with someone other than the character A (character-human relationship characteristics) -Define the personality of each other).

At this time, various methods exist as a tool for analyzing personality types, and various methods can be used. For example, a DISC model may be used. At this time, the personality types in the DISC model are divided into four types: Dominance (D), Influences (I), Conscientiousness (C), and Steadiness (S). The leading type D-type is work-oriented and places importance on quick determination and momentum, while the sociable type I-type is good at influencing others and is active and fun. Type C is a cautious type, has few words, is analytical and accurate. S-type is a stable, relaxed and rational personality. Alternatively, the Big Five Personality model of personality psychology can be used. Personality and persona are measured in the five axes of experiential openness, conscientiousness, extraversion, affinity, and neuroticism. You can also define it. The affect model can be composed of PAD (Pleasure Arousal Dominance) and OCC (Ortony, Clore & Collins), where P (Pleasure) of PAD represents positive or negative emotion, and A (Arousal) is the degree of arousal, And D (Dominance) represents a sense of control over the other person. Sentiment is mapped between -1 and 1 on each axis of P, A, and D. If PAD expresses emotion in one dimension, the OCC model complements it. The OCC computational model categorizes emotions into 22, and defines necessary components for each emotion among objects, events, and actions. For example, feeling happiness becomes an entity that needs an event, an emotion of love requires an object, and an emotion of pride is an action to be proud of. Is the most important. Of course, in addition to the above-described models, various character, personality, and persona definition models may be used, and it will be apparent that the emotional model is not limited to those described above.

In addition, AI characters act according to user choices and a vast amount of predefined decision trees. Here, context is all information used to specify a situation, and context recognition is adopted according to changes in objects over time, as well as nearby people and objects, and places used. Also, according to the behavioral model of cognitive psychology, understanding and perceiving the world is possible only through subjective perception. In other words, the context recognition according to the behavioral model is that not only people, objects, places, but also changes of objects over time are progressed through selection based on a subjective perspective. Accordingly, the character-based interactive artificial intelligence according to an embodiment of the present invention is defined as an interactive artificial intelligence capable of understanding the situation of dialogue through continuous dialogue execution and expressing subjective opinions and emotions based on this.

At this time, it is possible to enable continuous conversation by learning the answer sentence for the input sentence, such as interactive artificial intelligence using machine learning, or by adding the sentence just before the conversation as learning data to the above method through Generative Hierachical Neural Network Models. May be. In addition, multiple agents can be introduced for character-based interactive artificial intelligence. For example, it is divided into perception, decision making, planning/action, etc., and each agent is subdivided into knowledge base and reasoning engine. Therefore, in a platform composed of a plurality of agents, it can proceed in the order of reading and grasping the conversation of the other party, recognizing the overall conversation situation in consideration of one's emotional state, devising a response strategy, and establishing an action plan. .

In this case, the target language analysis agent is an agent that analyzes the meaning of the target and the emotions of the target through the sentences received from the other party. In this agent, through short analysis and in-depth analysis from the feedback of the decision-making agent and the speech of the other party, the meaning and emotion that the other party wants to convey is grasped and transmitted to the conversation situation determination agent. Short analysis refers to a semantic and emotional analysis through the sentences received from the other party. In-depth analysis is an analysis module for inferring the other's actual intentions and feelings by accumulating the opponent's intentions and emotions updated through feedback received from the context-aware agent. Since in-depth analysis is composed of only the values of short analysis at the beginning of the dialogue, prejudices about the subject are reflected, and this is updated according to the intention and emotion of the actual subject as the dialogue proceeds.

The self-emotion grasping agent is an agent for modeling changes in the artificial intelligence's own emotions through sentences received from the other party. Like the target language analysis agent, this agent also grasps the AI's own emotions from the counterpart's remarks through short analysis and in-depth analysis and delivers them to the conversational situation determination agent. Modeling through fragmentary reactions is an instantaneous emotional change that occurs through the other's strong reaction over a certain level. This affects the sudden change of emotions or momentary expression of emotions caused by strong stimulation. Modeling through in-depth reaction models the cumulative emotional changes arising from the situation in which the conversation with the other party proceeds.

The conversational situation determination agent grasps the context of the conversation through the meaning and emotional state received from the target language analysis agent and the self-emotion grasping agent, and the feedback of the decision-making agent, and delivers it to the decision-making agent. In addition, feedback is provided to the target language analysis agent and the self-emotion grasping agent on the identified situation, and provides for updating each in-depth analysis module.

The decision-making agent is an agent that determines the intention of the next speech, including the purpose of speech of artificial intelligence, in the situation identified through the conversational situation determination agent. The decision-making agent provides feedback to the inference engine of the target language analysis agent and the conversational situation determination agent to achieve a subjective interpretation of a given situation. In order to achieve its own purpose, artificial intelligence makes decisions through simulation through an emotional model created through a target language analysis agent and a self-emotion grasp agent.

The response generation agent is an agent that generates the intention of the next speech determined by the decision-making agent at the level of the sentence. The sentence expression agent is an agent that transforms the result sentence of the response generating agent to fit the grammar and specific tone (dialect, etc.). Therefore, in the case of defining all characters, personalities, and personalities by relationship so that interaction is possible and an agent is defined, a dialog agent that simulates the personality of the object that has been in the relationship with the user may be created, and accordingly, the user terminal 100 ) Can chat with the created chat agent.

The relearning unit 330 may relearn the artificial intelligence conversation agent with the determined character. In this case, various methods such as neural network learning, machine learning, and deep learning may be used, and active learning may be used. In this case, active learning means that the agent learns knowledge by itself by interacting with the environment beyond learning the conversational agent with a given dataset. The conversation agent actively acts on the environment, collects information from the environment, and reuses it for learning to improve performance. Looking at human learning, it can be said that it is active learning in that it not only acquires information that is refined, but also works by itself and obtains information. In order for active learning to take place in the question-and-answer field, questions must be generated from conversation agents. The conversation agent asks the generated question to the environment, and the user terminal 100 or big data gives the answer to the conversation agent. In terms of cost, the interaction between the conversation agent and the environment is less expensive than the cost of collecting the dataset. In Q&A, the environment is the user. The main purpose of the user is to use the agent, but it can easily answer the questions the agent asks. If the agent's performance can gradually increase through this process, the need to collect vast amounts of data during neural network training is reduced. The questions generated by the neural network are classified based on questionable words such as WHAT, HOW, HOW, WHEN, WHICH, WHERE, and WHY, and can show good accuracy in each case. For example, a neural network takes two words and their relationship as input. It then creates a question in the form of asking what is related to a word. If the generated question is based on an existing Q&A dataset, we can expand the dataset by asking the user back questions and receiving answers.

When selecting a character in the user terminal 100, the setting unit 340 may set an artificial intelligence conversation agent corresponding to the selected character to be driven in the user terminal 100. In this case, the user terminal 100 may conduct a conversation with an artificial intelligence conversation agent generated based on the personal record data. And, even if the character, that is, the character "Mother" is selected, the setting unit 340 must check who the other party is, that is, the number of children of the "Mother". The setting unit 340 may use the relationship defined by the user terminal 100, or may automatically define and grasp the other party through the conversation history log. When the relationship between the other party and the character is defined, a conversation agent learned by the relationship may be provided, and the conversation is controlled to proceed using a word database and personality generated within the relationship.

The relationship definition unit 350 defines a relationship that forms a human network between a user of the user terminal 100 and an object to create a character of the artificial intelligence conversation agent from the user terminal 100, and has artificial intelligence to correspond to the defined relationship. You can train conversation agents. In addition, the relationship definition unit 350 may establish a word database to correspond to the defined relationship, and set a plurality of word databases corresponding to a plurality of relationships to one character, respectively. At this time, since the relationship definition is the same as described above, a duplicate description will be omitted.

The providing unit 360 may provide a conversation interface for transmitting and receiving a conversation between the artificial intelligence conversation agent and the user terminal 100. At this time, the same is true for the text-based conversation interface, but the problem is the voice interface. A conversation agent that imitates not only the voice of the object of the object terminal 400, but also tone, tone, tone, etc. should be provided using an algorithm related to human robot interaction (HRI) personification. For example, stress is a change in rhyme emphasis. To measure this, average energy and frequency spectrum entropy of the basic form are extracted from each voice section, and then an estimate of the average scale standard deviation of energy, formant frequency and spectral entropy is obtained. This standard deviation measures speaking stress by focusing on the sum of z-scoreds, and applies it to conversational agents to train and learn. Similarly, it is possible to learn the voice, tone, dialect, and word database to be used.

When the identifier of the object terminal of the object for which the character of the conversation agent is to be created is input from the user terminal 100, the connection unit 370 searches for the object terminal based on the identifier, and at least one content included in the searched object terminal Can be collected. For example, in order to find the data of the deceased object, it may be collected from the user terminal 100, but it is more likely to collect data by directly accessing the object terminal 400 to collect data. Accordingly, assuming that the access unit 370 provides, for example, a MAC address, the object terminal 400 is searched for the MAC address to request and receive data from the object terminal 400, while the personal record data is stored. Can be collected.

When collecting personal record data, you can also use a method of constructing big data. This is because even one object needs various data to be classified by each character and personality. At this time, the artificial intelligence conversation service providing server 300 may store collected raw data in parallel and distributed, and unstructured data, structured data, and semi-structured data included in the stored raw data Data (semi-structured) can be refined, pre-processing including classification as meta data can be performed, and analysis including data mining of pre-processed data can be performed. In addition, the artificial intelligence conversation service providing server 300 may visualize and output the analyzed data. At this time, in data mining, classification (Classification), which predicts a class of new data by learning a training data set with a known class by searching for an intrinsic relationship between preprocessed data, or clustering that group data based on similarity without class information ( Clustering) can be performed. Of course, there may be various other mining methods, and may be mined differently according to the type of big data to be collected and stored or the type of query to be requested later. The big data constructed in this way may be verified through deep learning or machine learning of artificial neural networks.

At this time, the artificial neural network can use a CNN (Convolutional neural network) structure, which is a network structure using a convolutional layer, which is suitable for image processing, and can classify images based on features in the image by inputting image data. Because it can. In addition, text mining is a technology that aims to extract and process useful information from non/semi-structured text data based on natural language processing technology. Through text mining technology, meaningful information can be extracted from a vast bundle of texts, linkages with other information can be grasped, and the results can be obtained beyond finding a category of text or simply searching for information. Using this, in the artificial intelligence conversation service according to an embodiment of the present invention, large-capacity language resources and statistical and regular algorithms can be used to analyze an identifier or natural language input as a query and discover hidden information therein. . In addition, cluster analysis can be used to finally discover groups of similar characteristics while combining objects with similar characteristics.

Hereinafter, the operation process according to the configuration of the artificial intelligence conversation service providing server of FIG. 2 will be described in detail with reference to FIG. 3 as an example. However, it will be apparent that the embodiment is only any one of various embodiments of the present invention, and is not limited thereto.

Referring to FIG. 3, (a) the artificial intelligence dialogue service providing server 300 collects personal record data from the user terminal 100 and accesses the object terminal 400 to collect data. Then, (b) the artificial intelligence dialogue service providing server 300 establishes a character for an object of the object terminal 400 and defines a personality for each human relationship to analyze, learn, and train the character and personality for each relationship. In addition, (c) the artificial intelligence dialogue service providing server 300 learns the dialogue agent based on the analyzed character and personality, and (d) identifies the user of the user terminal 100 to provide a dialogue interface.

Matters that are not described for the method of providing artificial intelligence conversation service based on personal records data of FIGS. 2 and 3 are the same as or described above for the method of providing artificial intelligence conversation service based on personal records data through FIG. Since it can be easily inferred from the contents, the description will be omitted below.

4 is a diagram illustrating a process of transmitting and receiving data between components included in the system for providing an artificial intelligence conversation service based on personal record data of FIG. 1 according to an embodiment of the present invention. Hereinafter, an example of a process in which data is transmitted/received between each component will be described with reference to FIG. 4, but the present application is not limitedly interpreted as such an embodiment, as illustrated in FIG. 4 according to various embodiments described above. It is apparent to those skilled in the art that the process of transmitting and receiving data may be changed.

Referring to FIG. 4, the artificial intelligence conversation service providing server 300 receives personal record data from the user terminal 100 (S4100) and stores it (S4200). In addition, the artificial intelligence dialogue service providing server 300 receives definition data defining the relationship between the user and the object from the user terminal 100 (S4300), defines and stores the relationship between the user and the object (S4400), and After accessing the terminal 400 to collect more data (S4500), analysis of the collected data is started (S4600).

Based on the analysis result, the artificial intelligence dialogue service providing server 300 generates a character of the dialogue agent (S4610), extracts personality for each relationship based on the character (S4700), and builds a word database for each relationship (S4800). ), the learning is conducted based on this (S4810). The artificial intelligence conversation service providing server 300 provides a conversation interface when a conversation is requested from the user terminal 100 after the learning is completed (S4900), and a word or sentence that raises questions about personality is input. (S4910), the conversation immediately before raising a question about the personality is collected, stored as an error feedback, and relearning is started (S4920).

The order between the above-described steps S4100 to S4920 is only an example and is not limited thereto. That is, the order of the above-described steps (S4100 to S4920) may be mutually changed, and some of the steps may be executed or deleted at the same time.

The matters not described with respect to the method of providing an artificial intelligence dialogue service based on personal records data of FIG. 4 are the same as or described above with respect to the method of providing an artificial intelligence dialogue service based on personal records data through FIGS. 1 to 3. Since it can be easily inferred from the contents, the description will be omitted below.

5 is a flowchart illustrating a method of providing an artificial intelligence conversation service based on personal record data according to an embodiment of the present invention. 5, the artificial intelligence conversation service providing server collects at least one type of personal record data from a user terminal (S5100), learns the collected personal record data, and determines a character of the artificial intelligence conversation agent ( S5200).

And, the artificial intelligence dialogue service providing server re-learns the artificial intelligence dialogue agent with the determined character (S5300), and when a character is selected in the user terminal, the artificial intelligence dialogue agent corresponding to the selected character is set to run on the user terminal. Do (S5400).

The matters not described with respect to the method of providing an artificial intelligence dialogue service based on personal records data of FIG. 5 are the same as or described above with respect to the method of providing an artificial intelligence dialogue service based on personal records data through FIGS. 1 to 4 above. Since it can be easily inferred from the contents, the description will be omitted below.

The method of providing an artificial intelligence conversation service based on personal records data according to an embodiment described with reference to FIG. 5 is also implemented in the form of a recording medium including instructions executable by a computer such as an application executed by a computer or a program module. Can be. Computer-readable media can be any available media that can be accessed by a computer, and includes both volatile and nonvolatile media, removable and non-removable media. Further, the computer-readable medium may include all computer storage media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data.

In the above-described method for providing an artificial intelligence conversation service based on personal records data according to an embodiment of the present invention, an application basically installed on a terminal (this may include a program included in a platform or operating system basically mounted on the terminal). It may be executed by, and may be executed by an application (ie, a program) directly installed on the master terminal by a user through an application providing server such as an application store server, an application, or a web server related to the service. In this sense, the method for providing an artificial intelligence conversation service based on personal records data according to an embodiment of the present invention described above is implemented as an application (i.e., a program) installed in a terminal or directly installed by a user, and It can be recorded on a recording medium that can be read by.

The above description of the present invention is for illustrative purposes only, and those of ordinary skill in the art to which the present invention pertains will be able to understand that other specific forms can be easily modified without changing the technical spirit or essential features of the present invention will be. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not limiting. For example, each component described as a single type may be implemented in a distributed manner, and similarly, components described as being distributed may also be implemented in a combined form.

The scope of the present invention is indicated by the claims to be described later rather than the detailed description, and all changes or modified forms derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present invention. do.

Claims (6)

A user terminal that provides at least one type of personal record data for generating a character of the artificial intelligence chat agent, and conducts a chat with the artificial intelligence chat agent generated based on the personal record data; And
A collection unit that collects at least one type of personal record data from the user terminal, a determination unit that learns the collected personal record data to determine a character of an artificial intelligence conversation agent, and replays the artificial intelligence conversation agent with the determined character. An artificial intelligence dialogue service providing server including a relearning unit for learning, and a setting unit for setting an artificial intelligence dialogue agent corresponding to the selected character to be driven on the user terminal when the character is selected in the user terminal;
Personal record data-based artificial intelligence dialogue service providing system comprising a.
The method of claim 1,
The personal record data includes a video, a conversation history log, an image including a photo, audio data, and form record data.
The method of claim 1,
The artificial intelligence conversation service providing server,
A relationship defining unit that defines a relationship forming a human network between a user of the user terminal and an object to generate a character of the artificial intelligence conversation agent from the user terminal, and learns the artificial intelligence conversation agent to correspond to the defined relationship;
Personal record data-based artificial intelligence dialogue service providing system, characterized in that it further comprises.
The method of claim 3,
The relationship definition unit,
A system for providing an artificial intelligence conversation service based on personal records data, characterized in that a word database is constructed to correspond to the defined relationship, and a plurality of word databases corresponding to a plurality of relationships are set to be linked to one character.
The method of claim 1,
The artificial intelligence conversation service providing server,
A providing unit that provides a conversation interface for transmitting and receiving a conversation between the artificial intelligence conversation agent and the user terminal;
Personal record data-based artificial intelligence dialogue service providing system, characterized in that it further comprises.
The method of claim 1,
The artificial intelligence conversation service providing server,
When the user terminal inputs the identifier of the object terminal of the object for which the character of the conversation agent is to be created, the connection unit searches for the object terminal based on the identifier and collects at least one content included in the searched object terminal ;
Personal record data-based artificial intelligence dialogue service providing system, characterized in that it further comprises.

Images