KR20190058257A - Chatbot trainer platform and method of operating the same - Google Patents

Abstract

Provided are a chatbot trainer platform and an operating method thereof. The chatbot trainer platform comprises: a chatbot editor for receiving training data including data or information from a chatbot trainer who is a person allowing a chatbot to learn and providing the training data to a chatbot to learn and build the chatbot; a trainer database for storing and managing information of the chatbot trainer; and a chatbot database for storing and managing information of the chatbot wherein the chatbot editor provides a chatbot as a basic chatbot, which has learned basic data or information in advance to the chatbot trainer and the basic chatbot is allowed to learn the training data received from the chatbot trainer, thereby generating a professional chatbot.

Description

[0001] CHATBOT TRAINER PLATFORM AND METHOD OF OPERATING THE SAME [0002]

The present invention relates to a chatbot, and more particularly, to a chatbot trainer platform and a method of operating the same.

A chatbot is a computer program that can perform conversations using voice or text. Chatbots are also called talkbots, chatterbots, or simply bots. Chatbots can be used for various purposes in an interactive interface system such as a messenger service, including customer service or information acquisition. Some chatbots scan the keywords in the message, search the database for matches matched to the scanned keywords, and provide the retrieved responses. Some other chatbots are implemented using more sophisticated natural language processing systems. Natural language processing is a computer system that processes human language through morphological analysis, semantic analysis, dialogue analysis, etc., and enables more natural conversation.

Japanese Patent Application Laid-Open No. 10-1731867, Jul.

SUMMARY OF THE INVENTION The present invention provides a chattbot trainer platform.

Another object of the present invention is to provide a method of operating a chattbot trainer platform.

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

According to an aspect of the present invention, there is provided a chattbot trainer platform for receiving training data including data or information from a person who learns a chattb trainer-chatbot, providing the training data to a chatbot, And a chatbot database for storing and managing the information of the chatbot. The chatbot editor is a basic chatbot, and the basic data or information is stored in a dictionary Provides the learned chatbot to the chatbot trainer, and causes the basic type chatbot to learn the training data received from the chatbot trainer to generate the professional chatbot.

According to another aspect of the present invention, there is provided a method of operating a chattbot trainer platform, the method comprising: receiving training data including data or information from a person who learns a chattb trainer-chatbot; Learning and building the chatbot, storing and managing the information of the chatbot trainer, and storing and managing the information of the chatbot, wherein the step of learning and building the chatbot comprises the steps of: A chatbot that has learned basic data or information is provided to the chatbot trainer and a professional chatbot is generated by causing the basic chatbot to learn the training data received from the chatbot trainer.

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

According to the chattbot trainer platform of the present invention and its operating method of the present invention, various information or data provided from experts in various fields can be learned and built (i.e., generated) by the chatbot, and the individualized, By providing a chatbot having unique characteristics, the chatbot can be utilized in various fields and purposes.

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

1 is a schematic view illustrating an environment in which a chattbot trainer platform and a chattbock store platform according to an embodiment of the present invention are provided.
FIG. 2 is a view conceptually showing a chatbot trainer of the chatbot trainer platform of FIG. 1 being trained.
FIG. 3 is a view schematically showing a detailed configuration of the chattbot trainer platform of FIG. 1. FIG.
FIG. 4 is a view schematically showing a drive screen of an example of the chatbot editor of FIG.
5 is a diagram conceptually showing that a basic type chatbot learns learning data and is built into a professional type chatbot.
FIG. 6 conceptually illustrates registering a chatbot trained in the chatbot trainer platform of FIG. 1 in a chatbot store platform.
FIG. 7 is a view schematically showing a detailed configuration of the chat shop store platform of FIG. 1. FIG.
8 is a view schematically showing an example of chat or API use of a registered chatbot.
9 is a view conceptually showing that a chatbot is trained by another chatbot.
FIG. 10 is a view schematically showing a method of operating a chattbore store platform according to an embodiment of the present invention.
11 is a view schematically showing a method of operating a chattbore store platform according to another embodiment of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS The advantages and features of the present invention and the manner of achieving them will become apparent with reference to the embodiments described in detail below with reference to the accompanying drawings. It should be understood, however, that the invention is not limited to the disclosed embodiments, but may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein, Is provided to fully convey the scope of the present invention to a technician, and the present invention is only defined by the scope of the claims.

The terminology used herein is for the purpose of illustrating embodiments and is not intended to be limiting of the present invention. In the present specification, the singular form includes plural forms unless otherwise specified in the specification. The terms " comprises " and / or " comprising " used in the specification do not exclude the presence or addition of one or more other elements in addition to the stated element. Like reference numerals refer to like elements throughout the specification and " and / or " include each and every combination of one or more of the elements mentioned. Although " first ", " second " and the like are used to describe various components, it is needless to say that these components are not limited by these terms. These terms are used only to distinguish one component from another. Therefore, it goes without saying that the first component mentioned below may be the second component within the technical scope of the present invention.

Unless defined otherwise, all terms (including technical and scientific terms) used herein may be used in a sense that is commonly understood by one of ordinary skill in the art to which this invention belongs. In addition, commonly used predefined terms are not ideally or excessively interpreted unless explicitly defined otherwise.

&Quot; Chatbot " means a computer program that can perform a conversation using voice or text. "Chatbot" is a computer program that can also mimic human behavior and perform certain automated tasks.

&Quot; Training " means that the chatbot provides the chatbots with data or information to perform a conversation or perform a predetermined task in advance. In the following, " training " may simply be referred to as " learning ".

&Quot; Chat " means a conversation taking place on a communication network using a computer system.

Conventional chatbots have only learned data or information provided by a manufacturer or service provider (hereinafter referred to as " manufacturer, etc. "). Therefore, the conventional chatbot functioned only by a manufacturer or the like for a predetermined purpose. Further, even if not updated or updated, a plurality of conventional chatbots provided by one manufacturer or the like all have the same fixed feature or attribute. Therefore, conventional chatbots are not individuated and exhibit fixed uniform characteristics.

The chattbot trainer platform and its operating method according to the embodiment of the present invention learns and builds (i.e., generates) various data or information provided from experts in various fields to a chatbot, By providing a chatbot having unique characteristics, the chatbot can be utilized in various fields and purposes.

In addition, the chatbot store platform and the operation method thereof according to the embodiment of the present invention can open and share a chatbot having an individualized, fluid and unique characteristic, so that the chatbot can be utilized in various fields and purposes, The chattbot trainer will be able to distribute the profit from the use of the chattbot. In addition, by allowing the chatbot to be trained by other chatbots, it is possible for one chatbot to learn various data or information more easily without being restricted to specific data or information.

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

1 is a schematic view illustrating an environment in which a chattbot trainer platform and a chattbock store platform according to an embodiment of the present invention are provided.

Referring to FIG. 1, a chattbot trainer platform 100, a chattbot store platform 200, and a plurality of user devices 300 are interconnected through one or more networks. The network includes a wired communication network, a wireless communication network, or a combination thereof. The topology, protocol, communication method, etc. of the network can be variously modified according to the embodiment.

The chatbot trainer platform 100 trains and builds the chatbots. The chattbot trainer platform 100 learns not only its own internal data or information but also external data or information provided by the user to the chatbot. The chatbot trainer platform 100 enables a chatbot that a user is training or has completed training to use in connection with a predetermined app / web service.

The chatbot store platform 200 sells chatbots. The chatbots built in the chatbot trainer platform 100 can be registered in the chatbot store platform 200. The registered chatbots are disclosed by the chatbot store platform 200 and can be shared with the users. The user can pay a predetermined fee and chat with the registered chatbot or use it in connection with a predetermined web / app service through the API.

User device 300 represents a computer system used by a user of platform 100, 200. For example, the user device 300 may be, but is not limited to, a computer system such as a smartphone, a tablet, a personal digital assistant (PDA), a desktop, a laptop, . User device 300 may be any computer system including a communication module, a processor, and a memory.

Here, the user includes a user of the chatbot trainer platform 100, a user of the chatbot store platform 200, or a combination thereof. Hereinafter, the user of the chatbot trainer platform 100 may be referred to as a chatbot trainer. Chatbot Trainer is a person who learns chatbots by providing data or information to chatbots. A user of the chatbot store platform 200 registers a chatbot in the chatbot store platform 200 to allow a user to open and monetize the chatbot, a user to chat with the chatbot registered in the chatbot store platform 200, Means a user who uses the chatbot registered in the web browser in association with a predetermined web / app service. In some cases, the above users may be the same person or different persons.

1, the chattbot trainer platform 100 and the chattboste platform 200 are shown as separate components. However, the present invention is not limited thereto. For example, Or may be integrally formed.

FIG. 2 is a view conceptually showing a chatbot trainer of the chatbot trainer platform of FIG. 1 being trained.

Referring to FIG. 2, the chattbot trainer may train the chatbot 30 in the chattbot trainer platform 100. The chattbot trainer can provide the chattbot 30 in the chattbot trainer platform 100 with data or information to be learned using the chattbot trainer device 300A.

One chatbot 30 can be learned by one chatbot trainer. Also, one chatbot 30 can be learned by a plurality of chatbot trainer. Thus, the chatbot 30 can learn various data or information from experts in various fields.

FIG. 3 is a view schematically showing a detailed configuration of the chattbot trainer platform of FIG. 1. FIG.

3, the chattbot trainer platform 100 includes a chatbot editor 110, a chatbot 120, an API interface unit 130, a trainer database 140, and a chatbot database 150.

The chatbot editor 110 learns and builds the chatbot. The chatbot editor 110 receives data or information from the chatbot trainer and provides the received data or information to the chatbot for learning. The learning of chatbots is based on supervised learning, unsupervised learning, reinforcement learning, deep learning, ANN (Artificial Neural Networks) algorithm, RNN (Recurrent Neural Network) Algorithms, and the like, but are not limited thereto.

The chatbot 120 includes a basic type chatbot described later or a professional type chatbot that learns additional data or information. The chatbot 120 may be in a training state or a completed state.

The API interface unit 130 provides an API interface. The chatbot trainer can use the chatbots that he / she has learned in connection with a predetermined web / app service. At this time, the chatbot 120 in the chattob trainer platform 100 can communicate with the web / app service through an API. For example, the web / app service may be any one of commercial messenger services such as KAKAO TALK, LINE, Facebook Messenger, Wechat, etc., but is not limited thereto. The API interface unit 130 mediates communication between the chatbot 120 in the chattobo trainer platform 100 and the web / app service.

The trainer database 140 stores and manages information of the trainer. For example, the information of the trainer may include, but is not limited to, a name, an ID, an address, an email address, a contact, a sex, a job, a photograph, a self introduction, a profile, Here, a related chatbot means a chatbot trained by the corresponding chatbot trainer. As described above, since one chatbot can be trained by a plurality of chatbot trainer, related chatbind IDs can be overlapped among information of each chatbot trainer.

The chatbot database 150 stores and manages information of the chatbots. For example, the information of the chatbot may include, but is not limited to, the name of the chatbot, the name, the description, the language of use, the field, and the associated chatbot trainer ID. Here, the related chatbot trainer means a chatbot trainer that has trained the corresponding chatbot. As described above, since one chatbot can be trained by a plurality of chatbot trainer, the associated chatbot trainer ID can include a plurality of IDs.

Meanwhile, the chatbox database 150 may further include a related chatbind ID. Here, the related chatbot means a chatbot that has trained the corresponding chatbot. This is because one chatbot can be learned by one or more chatbots without human involvement, as will be described later.

The chattbot trainer platform 100 may further include a communication unit (not shown) for communication with the chattbot trainer device 300A.

Although not explicitly shown in FIG. 3, the chattbot trainer platform 100 may be modified to include some of the components shown, or some components not shown, depending on the embodiment.

FIG. 4 is a view schematically showing a drive screen of an example of the chatbot editor of FIG.

Referring to FIG. 4, the chatbot editor 110 may be provided in a visual-based coding scheme rather than a text-based coding scheme. The chatbot editor 110 provides a plurality of types of blocks 52. The chattbot trainer may input predetermined data or information (e.g., message to be displayed, context, intention, external link, etc.) to the block 52 using the edit window 51. [ The chatbot trainer can train chatbots by adding / deleting / editing / linking one or more blocks 52 as they draw the mind map. The connection relationship of each block 52 indicates a logical upper / lower or pre / post-relationship. Each block 52 may be hierarchically linked (e. G., A parent relationship or a sibling relationship). The chattbot trainer can execute the chatbot under training using the menu 53, acquire the guide information of the chatbot training, or manage the chatbot that is being trained or completed. This allows the chattbot trainer to easily train the chatbot without having to do separate coding (writing text-based code).

4, an editor having a user interface based on a mind map (i.e., learning a chatbot in a manner of drawing a mind map) is shown as an example of the chatbot editor 110. However, the present invention is not limited thereto, The user interface of the chatbot editor 110 may be modified in various ways.

5 is a diagram conceptually showing that a basic type chatbot learns learning data and is built into a professional type chatbot.

Referring to FIG. 5, a basic type chatbot is provided by a chatbot trainer platform 100.

The basic type chatbot 10 refers to a chatbot in which basic data or information is learned in advance. For example, basic data or information may be provided to guide the chattbot trainer to use the chattbot trainer platform 100, how to train the chattbot, how to connect the external API, and the like. In another example, the basic data or information may be for sending a chat start message, greeting the other party, or responding to simple messages such as weather / stocks. The basic data or information learned in the basic type chatbot 10 is not limited to the above examples.

The chattbot trainer learns the training data 20 including predetermined data or information in the basic type chatbot 10 to generate the professional type chatbot 30. Although not explicitly shown, the chattbot trainer platform 100 provides an algorithm for combining the training data 20 added by the chattb trainer into the basic data or information of the basic chattbot 10. &Quot; Professional type " of the professional chatbot 30 means that new data or information has been learned in addition to the basic data or information previously learned as a concept against the " basic type ".

FIG. 6 conceptually illustrates registering a chatbot trained in the chatbot trainer platform of FIG. 1 in a chatbot store platform.

Referring to FIG. 6, the chattbot trainer may train the chatbot 30 in the chatbot trainer platform 100 and may selectively register the chatbot in the chatbot store platform 200 when the training is completed. The chatbots registered in the chatbot store platform 200 can be publicly disclosed and shared. The information stored and managed in the trainer database 140 and the chatbot database 150 of the chatbot trainer platform 100 are transferred to the chatbot store platform 200. [

FIG. 7 is a view schematically showing a detailed configuration of the chat shop store platform of FIG. 1. FIG.

7, the chatbot store platform 200 includes an API interface unit 210, a chatbot 220, a profit settlement unit 230, a user database 240, a trainer database 250, and a chatbox database 260 .

The API interface unit 210 provides an API interface. A user can use a desired chatbot among a public chatbot in connection with a predetermined web / app service. At this time, the chatbot 220 in the chatbot store platform 200 can communicate with the web / app service through an API. For example, the web / app service may be any one of commercial messenger services such as KAKAO TALK, LINE, Facebook Messenger, Wechat, etc., but is not limited thereto. The API interface unit 210 mediates communication between the web server 220 and the web / app service in the chattop store platform 200.

The chatbot 220 means a chatbot registered by a chatbot trainer.

The profit settlement unit 230 calculates the cost and profit of the user using the chatbot. The profit settlement unit 230 may distribute the profit to the chatbot trainer associated with the chatbot. The profit settlement unit 230 may calculate the distribution ratio of the profit or the amount to be distributed as the profit.

The user database 240 stores and manages user information. For example, the user's information may include, but is not limited to, a user's name, ID, address, email address, contact, gender, job,

The trainer database 250 stores and manages information of the trainer. For example, the information of the trainer may include, but is not limited to, a name, an ID, an address, an email address, a contact, a sex, a job, a photograph, a self introduction, a profile,

The chatbot database 260 stores and manages information of the chatbots. For example, the information of the chatbot may include, but is not limited to, the name of the chatbot, the name, the description, the language used, the field, the amount of use, the image of the chatbot, and the associated chatbot trainer ID.

Meanwhile, the chatbox database 260 may further include a related chatbind ID.

The chatbot store platform 200 may further include a communication unit (not shown) for communication with the user device 300 or the chattbot trainer device 300A.

Although not clearly shown in Fig. 7, according to the embodiment, the cha-bob store platform 200 may be modified so as not to include some of the components shown in the figure, or to further include some components not shown. For example, the chatbot store platform 200 may be modified to include a chatbot editor for training the chatbot 30.

8 is a view schematically showing an example of chat or API use of a registered chatbot.

8, the chatbot 30 registered in the chatbot store platform 200 communicates with a predetermined web / app service server 400 through a chat with the user device 300 or via an API, And can chat with the user of the Web / App service in association with the service.

The chatbot 30 may respond to a query from the user device 300. Alternatively, the chatbot 30 may perform a function such as a customer service (CS) of the Web / App service to respond to a query. However, the present invention is not limited thereto, and the chatbot 30 can perform an arbitrary operation.

When the chatbot 30 is unable to answer while performing a predetermined task (for example, in the case of an intention or context that has not been learned), the chatbot trainer associated with the chatbot 30 is inquired to the chatbot trainer device 300A And delivers the message received from the user or notifies the user of the fact. The inquiry of the associated chatbot trainer or the delivery of the user message may be performed by the chatbot store platform 200. To this end, the chatbot store platform 200 may further include a chatbot manager (not shown).

9 is a view conceptually showing that a chatbot is trained by another chatbot.

Referring to FIG. 9, the chatbot 30 can be trained by another chatbot 30. The chatbot 30 in the chatbot trainer platform 100 can be trained by another chatbot 30 registered in the chatbot store platform 200. [ The chatbot 30 can learn a method of training another chatbot 30 in advance. This can be learned basically or professionally by a chattb trainer.

One chatbot 30 can be learned by one chatbot 30 or a plurality of chatbots 30. In addition, one chatbot 30 can be trained by a chatbot 30 and another chatbot 30. Thus, since the training of the chatbot 30 is not required to be directly performed by a person, the training of the chatbot 30 can be performed quickly and conveniently.

The chattbot trainer of the chatbot 30 to be trained may pay a predetermined training cost to the chadbot trainer of the chambot 30 to train. This payment process can be performed by the chatbot training platform 100 or the chatbot store platform 200. [ To this end, the chatbot store platform 200 may further include a cost processor (not shown). Or, the profit can be distributed according to the contribution of the chatbot trainer of the chatbot 30 that is trained in the distribution process of the profit.

FIG. 10 is a view schematically showing a method of operating a chattbore store platform according to an embodiment of the present invention.

Referring to FIG. 10, in step S510, the chatbot receives a user message. Then, in step S520, the chatbot determines whether a response to the user message is possible. If an answer to the user message is possible, the chatbot sends a response message to the user in response to the user message in step S530. On the other hand, if the user message is not answered, the chatbot trainer associated with the chatbot is inquired in step S540. The chatbot can inquire a chattbot trainer related to itself based on the trainer database 250, the chatbot database 260, and the like. Alternatively, the chatbot notifies the chatbot management unit that the response is not possible, and the chatbot management unit can inquire the chatbot trainer related to the chatbot based on the trainer database 250, the chatbot database 260, and the like. Also, when the chatbot is trained by a plurality of chatbot trainer, a chatbot trainer in a field related to a user message among a plurality of chatbot trainers can be inquired. To this end, for example, the intention or context of a keyword or user message contained in a user message may be analyzed. Depending on the results of the analysis, the chattbot trainer or related chattbot trainer that has been trained more than once may be queried. Then, in step S550, the chatbot transmits a user message to the chatbot trainer inquired. The chatbot trainer can send a response message directly to the user. Alternatively, the chatbot trainer may re-train the chatbot to allow the chatbot to send an appropriate response message.

11 is a view schematically showing a method of operating a chattbore store platform according to another embodiment of the present invention.

Referring to FIG. 11, in step S610, the chatbot receives a user message. Then, in step S620, the chatbot transmits a response message to the user to the user message. Then, in step S630, the chatbot trainer related to the transmitted response message is inquired. The chatbot management unit may inquire, or the profit accounting unit 230 may inquire. Then, in step S640, the profit associated with the response message is distributed to the inquired chatbot trainer. Depending on the embodiment, the number of times, the time of the response message, or the revenue associated with a series of responses may be distributed to the chatbot trainer, but this is not so limited. The revenue associated with the response message includes any revenue generated by the transmission of the response message to the user. Also, when the chatbot is trained by a plurality of chatbot trainer, the chatbot trainer in the field related to the response message among the plurality of chatbot trainers can be inquired. That is, the chatbot trainer that trains the chatbot can be inquired so that the response message can be transmitted. And, profit can be distributed to inquired chattb trainer. When a plurality of chatbot trainer are associated with one response message, the revenue can be distributed to all of the plurality of chatbot trainer. At this time, the ratio of each of the chattbot trainer can be determined differently depending on the contribution degree. Also, when the chatbot is trained by another chatbot, the chatbot trainer associated with the other chatbot is inquired and the profit can be distributed to the inquired chatbot trainer. Even if the chatbot trainer and other chatbots are involved in one response message, the revenue can be distributed as described above.

The steps of a method or algorithm described in connection with the embodiments of the present invention may be embodied directly in hardware, in software modules executed in hardware, or in a combination of both. The software module may be a random access memory (RAM), a read only memory (ROM), an erasable programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a flash memory, a hard disk, a removable disk, a CD- May reside in any form of computer readable recording medium known in the art to which the invention pertains.

While the present invention has been described in connection with what is presently considered to be practical exemplary embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, You will understand. Therefore, it should be understood that the above-described embodiments are illustrative in all aspects and not restrictive.

100: Chattbot Trainer Platform
200: Chambers Store Platform
300: User device

Claims (10)

A chatbot editor that receives training data including data or information from a chatbot trainer and a person who learns chatbots, and provides the training data to chatbots to learn and build the chatbots;
A trainer database for storing and managing information of the chattbot trainer; And
And a chatbot database for storing and managing information of the chatbot,
The chatbot editor provides a chatbot that has been learned basic data or information as a basic chatbot to the chatbot trainer and allows the basic chatbot to learn the training data received from the chatbot trainer to generate a professional chatbot doing,
Chattbot Trainer Platform. The method according to claim 1,
Wherein the basic data or information includes at least one of a usage method of the chattobutor trainer platform, a training method of the chatbot, and an external API connection method,
Chattbot Trainer Platform. The method according to claim 1,
Wherein the basic data or information includes at least one of a start message of the chatbot, a message for greeting, a message for weather or stock response,
Chattbot Trainer Platform. The method according to claim 1,
Wherein the chatbot editor provides an algorithm for combining the training data with the basic data or information,
Chattbot Trainer Platform. The method according to claim 1,
The visual coding scheme of the chatbot editor provides a plurality of types of blocks, the chatbot trainer inputs predetermined data or information into one or more blocks, and connects the one or more blocks in such a manner as to draw a mind map, ≪ / RTI >
Chattbot Trainer Platform. 6. The method of claim 5,
Wherein the predetermined data or information input to the block includes one or more of a message, context, intent or external link to be displayed,
Chattbot Trainer Platform. The method according to claim 1,
The chatbot editor provides guide information of chatbot training,
Chattbot Trainer Platform. The method according to claim 1,
Wherein the information of the chatbot trainer includes information of a chatbot trained by the chatbot trainer,
Chattbot Trainer Platform. The method according to claim 1,
Wherein the information of the chatbot includes information of a chatbot trainer that has trained the chatbot,
Chattbot Trainer Platform. As a method realized by a computer,
Receiving training data including data or information from a chattb trainer and a person learning a chatbot, and providing the training data to a chatbot to learn and build the chatbot;
Storing and managing information of the chatboat trainer; And
Storing and managing information of the chatbot,
The step of learning and building the chatbot may include providing a chatbot that has been preliminarily learned basic data or information as a basic type chatbot to allow the basic type chatbot to learn the training data received from the chatbot trainer, ≪ / RTI >
How to operate the Chattbot Trainer platform.

Images