KR20190123708A - Server, device and method for providing instant messeging service by using relay chatbot - Google Patents

Abstract

According to one embodiment of the present invention, provided is a chatbot server for providing an instant messaging service using a relay chatbot relaying a plurality of chatbots. The chatbot server comprises: a reception unit receiving a query message received for a service selected by a user terminal of a plurality of services and service account information corresponding to the selected service from a messenger server; a deriving unit deriving a response to the query message using a relay chatbot corresponding to the service account information; and a transmission unit transmitting the response to the messenger server. The chatbot derives the response by at least one chatbot of the plurality of chatbots by relaying the plurality of chatbots connected with the relay chatbot and the user terminal.

Description

SERVER, DEVICE AND METHOD FOR PROVIDING INSTANT MESSEGING SERVICE BY USING RELAY CHATBOT}

The present invention relates to a server, a terminal, and a method for providing an instant messaging service, and more particularly, to a server, a terminal, and a method for providing an instant messaging service using a relay chatbot for relaying a plurality of chatbots.

Chatbot is a chat robot, or AI communication program that provides answers to questions and related information through text conversations with people based on artificial intelligence. These chatbots are provided via rules or artificial intelligence technology as services that interact via a messenger application.

Looking at the structure of the conventional chatbot service, the chatbot service structure has a structure in which a chatbot is mapped in a one-to-one relationship for each chatbot service to provide a chat service through the mapped chatbot. For example, when the chatbot server receives a chatbot ID and a query message for a first chatbot service from a user terminal through a messenger server, the chatbot server sends a response to the query message through the first chatbot corresponding to the received chatbot ID. When providing a chatbot ID and a query message for the second chatbot service through the messenger server through the messenger server, the user terminal sends a response to the query message through the second chatbot corresponding to the received chatbot ID. Provided to user terminal via server.

Since the conventional chatbot service provides a service through a chatbot corresponding to each service, the response to the query message is inevitably constant, and it is difficult to derive the best response to the query message.

Japanese Laid-Open Patent Publication No. 2001-337827 (published Dec. 7, 2001)

The present invention is to solve the above-mentioned problems of the prior art, any one of a plurality of chatbots connected to the relay chatbot via a chat chatbot corresponding to the service to the query message associated with the service selected by the user terminal of the plurality of services A server, a terminal, and a method for deriving a response to a query message from the same and providing the derived response to a user terminal are provided.

The present invention also selects a chatbot that processes a query message among a plurality of chatbots using a machine learning model constituting the relay chatbot, derives a response to the query message through the selected chatbot, and sends the derived response to the user terminal. To provide a server, a terminal and a method for providing.

The present invention also provides a server, a terminal, and a method for selecting a response based on a ranking obtained by deriving a ranking of responses to a query message received from a plurality of chatbots through a relay chatbot, and providing the selected response to a user terminal. To provide.

The present invention also provides a server that selects a chatbot that processes a query message using a machine learning model constituting a relay chatbot, derives a response to the query message through the selected chatbot, and provides the derived response to the user terminal. It is intended to provide a terminal and a method.

The present invention also provides a server that derives the response most suitable for the user's query intention among the responses derived by the plurality of chatbots by using a relay chatbot that relays a plurality of chatbots and user terminals without transmitting a direct query message to each chatbot. To provide a terminal and a method.

As a technical means for achieving the above technical problem, a chatbot server that provides an instant messaging service using a relay chatbot relaying a plurality of chatbots and user terminals according to an embodiment of the present invention is a user terminal of a plurality of services A receiving unit receiving a query message received for the selected service and service account information corresponding to the selected service from the messenger server; A derivation unit for deriving a response to the query message using a relay chatbot corresponding to the service account information; And a transmission unit for transmitting the response to the messenger server, wherein the relay chatbot relays the plurality of chatbots and user terminals connected to the relay chatbot to derive the response by at least one chatbot among the plurality of chatbots. .

In a chatbot server according to another embodiment of the present invention, a method for providing an instant messaging service using a relay chatbot connected to a plurality of chatbots includes a query message received for a service selected from a user terminal among a plurality of services and the selected service. Receiving service account information corresponding to the messenger server; Deriving a response to the query message by at least one chatbot of the plurality of chatbots by relaying a plurality of chatbots and a user terminal connected to the relay chatbot using a relay chatbot corresponding to the service account information; And transmitting the derived response to the messenger server.

The present invention is to solve the above-mentioned problems of the prior art, any one of a plurality of chatbots connected to the chatbot connected to the relay chatbot via the middle chatbot corresponding to the service selected by the user terminal of the plurality of services corresponding to the service A server, a terminal, and a method for deriving a response to a query message from the terminal and providing the derived response to the user terminal can be provided.

The present invention also selects a chatbot that processes a query message among a plurality of chatbots using a machine learning model constituting the relay chatbot, derives a response to the query message through the selected chatbot, and sends the derived response to the user terminal. It can provide a server, a terminal and a method for providing.

The present invention also provides a server, a terminal, and a method for selecting a response based on a ranking obtained by deriving a ranking of responses to a query message received from a plurality of chatbots through a relay chatbot, and providing the selected response to a user terminal. Can provide.

The present invention also provides a server that selects a chatbot that processes a query message using a machine learning model constituting a relay chatbot, derives a response to the query message through the selected chatbot, and provides the derived response to the user terminal. A terminal and a method can be provided.

The present invention also provides a server that derives a response most suitable for a user's query intention among the responses derived by the plurality of chatbots by using a relay chatbot relaying a plurality of chatbots and user terminals without transmitting a direct query message to each chatbot. A terminal and a method can be provided.

1 is a diagram schematically illustrating a system for providing an instant messaging service according to an embodiment of the present invention.
2 is a block diagram of a chatbot server according to an embodiment of the present invention.
3A to 3C are diagrams for describing a method for providing an instant messaging service using a relay chatbot according to an embodiment of the present invention.
4A to 4C illustrate a method of deriving a response to a query message according to an embodiment of the present invention.
5 is a flowchart of a method for providing an instant messaging service through a chatbot server according to an embodiment of the present invention.
6 is an exemplary diagram of a response to a query message according to an embodiment of the present invention.

DETAILED DESCRIPTION Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art may easily implement the present invention. As those skilled in the art would realize, the described embodiments may be modified in various different ways, all without departing from the spirit or scope of the present invention. In the drawings, parts irrelevant to the description are omitted in order to clearly describe the present invention, and like reference numerals designate like parts throughout the specification.

Throughout the specification, when a part is "connected" to another part, this includes not only "directly connected" but also "electrically connected" with another element in between. . In addition, when a part is said to "include" a certain component, which means that it may further include other components, except to exclude other components unless otherwise stated.

In the present specification, the term 'unit' includes a unit realized by hardware, a unit realized by software, and a unit realized by both. In addition, one unit may be realized using two or more pieces of hardware, and two or more units may be realized by one piece of hardware. Meanwhile, '~' is not limited to software or hardware, and '~' may be configured to be in an addressable storage medium or may be configured to reproduce one or more processors. Thus, as an example, '~' means components such as software components, object-oriented software components, class components, and task components, and processes, functions, properties, procedures, and the like. Subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, and variables. The functionality provided within the components and the 'parts' may be combined into a smaller number of components and the 'parts' or further separated into additional components and the 'parts'. In addition, the components and '~' may be implemented to play one or more CPUs in the device or secure multimedia card.

The "user terminal" referred to below may be implemented as a computer or a portable terminal that can access a server or another terminal through a network.

1 is a diagram schematically illustrating a system for providing an instant messaging service according to an embodiment of the present invention.

Referring to FIG. 1, an instant messaging service providing system may include a user terminal 100, a messenger server 110, and a chatbot server 120. Since the instant messaging service providing system of FIG. 1 is only an embodiment of the present invention, the present invention is not limitedly interpreted through FIG. 1. That is, according to various embodiments of the present invention, the instant messaging service providing system may be configured differently from FIG. 1.

Friends that can be registered through a messenger app installed in the user terminal 100 may include real friends and virtual friends.

A user of the user terminal 100 may register another user (eg, family, acquaintances, work colleagues, etc.) having no commercial purpose as a real friend through a messenger app.

The user terminal 100 may provide an instant messaging service to a user by exchanging a conversation with another user registered as a real friend through a chat room through a messenger app installed in the user terminal 100.

A user of the user terminal 100 may register an official account corresponding to a specific service as a virtual friend through a messenger app. The official account may mean an account operated by a specific company or organization to provide a specific service. For example, the official account may provide information about a specific service to the user terminal 100 through a chat room established between the official account and the user terminal 100. The user terminal 100 may register a corresponding official account as a virtual friend and receive information on a specific service from the official account. Meanwhile, even if the official account is not registered as a virtual friend, information on a specific service may be provided.

The official account may use the chatbot to provide the user terminal 100 with information about a specific service provided by a specific company and organization. The administrator of the official account creates a chatbot through a user interface related to the bot builder provided by the chatbot server 120, and connects the created chatbot with the official account so as to link information about a specific service (for example, a specific service). Response to the query message related to the). In addition, the official account manager can manage the official account and the chatbot by one-to-one matching through the user interface for the bot builder, and can be extended to link a plurality of chatbots to the official account.

In addition, a user of the user terminal 100 may register a relay chatbot corresponding to a specific service as a virtual friend through a messenger app. The relay chatbot is a chatbot that provides a relay service by relaying between the user terminal 100 and the plurality of chatbots, and the user terminal 100 may receive a response to an inquiry message through the relay chatbot.

Therefore, when the user terminal 100 registers an official account as a virtual friend and the corresponding official account and the relay chatbot are mapped, the user terminal 100 may receive a response to an inquiry message through the relay chatbot linked with the official account, and the relay chatbot itself. If you register as a virtual friend, you can be provided with a response to the query message through the relay chatbot.

The user terminal 100 may provide an instant messaging service to a user by exchanging a chat through a chat room with, for example, a relay chatbot or an official account registered as a virtual friend in a messenger app. In addition, the user terminal 100 may invite a relay chatbot registered as a virtual friend to a chat room established between the user terminal 100 and a real friend.

The user terminal 100 may receive one of a plurality of services provided through a messenger app installed in the user terminal 100 from a user. For example, the plurality of services may include a service provided by an official account and a service provided by a relay chatbot. The user may receive a service provided by the official account or a service provided by the relay chatbot through the chat room by selecting the official account or the relay chatbot as the virtual friend.

When the user terminal 100 selects one service among a plurality of services provided through a messenger app installed in the user terminal 100, the user terminal 100 may transmit service account information corresponding to the selected service to the messenger server 110. . For example, the service account information may include a chatbot ID corresponding to the selected service or an official account ID of the official account corresponding to the selected service.

In addition, when the user terminal 100 receives a query message related to a service selected by the user from the user through a chat room, the user terminal 100 may transmit the corresponding query message to the messenger server 110. The service account information and the query message may be delivered to the chatbot server 120 through the messenger server 110.

The messenger server 110 may provide an instant messaging service that enables other users registered as real friends with the user terminal 100 to exchange conversations in a chat room. In addition, the messenger server 110 may provide an instant messaging service that allows the user terminal 100 to exchange a conversation with a relay chatbot or an official account registered as a virtual friend in a messenger app.

For example, when the messenger server 110 receives a chatbot ID and a query message about a selected service from the user terminal 100, the messenger server 110 may transfer the received chatbot ID and the query message to the chatbot server 120. As another example, when the messenger server 110 receives an official account ID and a query message about a selected service from the user terminal 100, the messenger server 110 receives the official account ID (or a chatbot ID mapped with the received official account ID). And the received query message to the chatbot server 120.

The messenger server 110 may receive a response to the query message derived by the relay chatbot from the chatbot server 120. The messenger server 110 may transmit the received response to a chat room established between the user terminal 100 and the chatbot or a chat room opened between the user terminal 100 and an official account.

The chatbot server 120 receives an official account ID and a query message from the messenger server 110, and when the official account ID and the relay chatbot are mapped, the chatbot server 120 responds to the query message by using a relay chatbot corresponding to the official account ID. Can be derived. In addition, when the chatbot server 120 receives a chatbot ID and a query message from the messenger server 110, the chatbot server 120 may derive a response to the query message using a relay chatbot corresponding to the chatbot ID. In this case, the relay chatbot may derive a response to the query message from at least one chatbot among the plurality of chatbots by relaying a plurality of chatbots and a user terminal connected to the relay chatbot.

The chatbot server 120 may transmit the derived response to the messenger server 110. The response sent to the messenger server 110 may be sent to the user terminal 100.

The relay chatbot proposed by the present invention is a chatbot that provides a relay service between a plurality of chatbots, and selects a chatbot that relays the most appropriate response to a user's inquiry message among the chatbot manager's relay policy (for example, a plurality of chatbots). According to the method, the user's query message can be dynamically delivered to all the chatbots to be the intermediate mediator to derive the most appropriate response to the query message.

Operation of each component included in the instant messaging service providing system of FIG. 1 will be described in more detail with reference to the following drawings.

2 is a block diagram of a chatbot server 120 according to an embodiment of the present invention. Referring to FIG. 2, the chatbot server 120 according to an embodiment of the present invention includes a receiving unit 200, a deriving unit 210, a circulation structure determining unit 220, a bot builder 230 and a transmitting unit 240. ) May be included. However, the configuration of the chatbot server 120 according to an embodiment of the present invention may be made differently from FIG.

When the receiver 200 receives a selection from a user of the user terminal 100 among a plurality of services registered in a messenger app that provides an instant messaging service, the receiver 200 receives service account information corresponding to the selected service from the messenger server 110. Can be received.

For example, when a service provided by a specific official account among a plurality of services is selected by a user of the user terminal 100, the receiver 200 may receive an official account ID corresponding to the selected service from the messenger server 110. Can be.

For another example, when a service provided by a specific chatbot among a plurality of services is selected by a user of the user terminal 100, the receiver 200 may receive a chatbot ID corresponding to the selected service from the messenger server 110. have.

When the user of the user terminal 100 selects a service for the relay chatbot registered in the messenger app, the receiver 200 may receive the corresponding relay chatbot ID from the messenger server 110. For example, when the user of the user terminal 100 selects a certain service among a plurality of services registered in the messenger app, and the chatbot corresponding to the selected specific service is a relay chatbot, the receiver 200 selects the selected service. The corresponding chatbot ID may be received from the messenger server 110. In this case, the relay chatbot may be a chatbot connected to a plurality of chatbots associated with a specific service to interrogate between the user terminal 100 and the plurality of chatbots for a query / response to a specific service.

The receiver 200 may receive a query message input for the service selected by the user from the messenger server 110. For example, when the user selects a music service, the receiver 200 may receive a query message related to the music service from the messenger server 110.

The derivation unit 210 may derive a response to the query message using the relay chatbot corresponding to the relay chatbot ID received from the messenger server 110.

The derivation unit 210 may transmit a query message received from the messenger server 110 to a relay chatbot corresponding to the received relay chatbot ID. At this time, the relay chatbot may derive a response by relaying a plurality of chatbots and user terminals connected to the relay chatbot. The relay chatbot may transmit a query message to at least one chatbot among a plurality of chatbots connected to the relay chatbot. At least one chatbot receiving the query message may derive a response to the query message and transmit the response to the relay chatbot.

For example, the derivation unit 210 derives a chatbot ID of a chatbot to process a query message through a relay chatbot, delivers the derived chatbot ID to a bot DB (not shown), and the bot DB is a previously stored bot-related meta. The bot metadata corresponding to the chatbot ID received from the data may be detected. The derivation unit 310 may receive the bot metadata of the chatbot corresponding to the chatbot ID from the bot DB and derive a response to the query message based on the received bot metadata of the chatbot.

Referring to FIG. 3A for a while, a plurality of services are assigned a relay chatbot ID corresponding to each service, and a service is provided through each relay chatbot corresponding to each assigned relay chatbot ID. For example, suppose the first service is a service provided by Company A's official account. In order to provide the first service, an official account corresponding to the first service and the first relay chatbot 300 are mapped, and queries processed by the plurality of chatbots 302, 304, and 306 connected to the first relay chatbot 300. One response selected by the first relay chatbot 300 among the responses to the message may be provided to the user terminal 100 through the messenger server 110.

As another example, suppose the second service is a service provided by Company B's official account. A query message processed by a plurality of chatbots 306, 310, and 312 that are mapped to an official account corresponding to the second service and a second relay chatbot 308 to provide a second service, and connected to the second relay chatbot 308. One response selected by the second relay chatbot 308 among the responses is provided to the user terminal 100 through the messenger server 110.

Each relay chatbot is connected to a plurality of chatbots that derive a response to a query message of each service, and relays a plurality of chatbots and a user terminal connected to each relay chatbot. For example, referring to FIG. 3A, a first relay chatbot 300 that provides a response to a query message related to a music service may include an external server (eg, a melon) corresponding to a third party that provides a music service. Music server, bucks music server, etc.) connected to the first chatbot 302 and second chatbot 304, and connected to the third chatbot 306 associated with the search engine for providing a search, The chatbots 302, 304, and 306 of the user terminal may be relayed.

At least one chatbot of the plurality of chatbots may be shared with the relay chatbot and other relay chatbots. A chatbot may not be connected to just one relay chatbot, but may be connected to another relay chatbot even if the chatbot is already connected to a specific relay chatbot. Accordingly, in the present invention, different relay chatbots may share a specific chatbot together. For example, referring to FIG. 3A, of the plurality of chatbots 302, 304, and 306 connected to the first relay chatbot 300, the third chatbot 306 associated with the search engine may be configured to answer a query message related to a weather service. It may also be shared with a second relay chatbot 308 that provides a response.

The relay chatbot may derive a response to the query message by relaying another relay chatbot connected to the relay chatbot. For example, referring to FIG. 3B, a third relay chatbot 314 providing a response to a query message related to a food service may select a lower relay chatbot 318 that is in charge of a service for a specific food included in the food service. You can relay. At this time, the third relay chatbot 314 delivers a food-related query message to the lower relay chatbot 318, and the lower relay chatbot 318 is connected to the plurality of chatbots 320 and 322 connected to the lower relay chatbot 318. The derived response may be transmitted to the third relay chatbot 314.

The circulation structure determination unit 220 may determine whether a circulation structure in which different relay chatbots relay each other is generated. At this time, the circular structure determination unit 220 may monitor the traffic between the relay chatbot and the lower chatbot of the relay chatbot. The circulation structure determiner 220 may correspond to an optional element that is not an essential element as a unit configuring the chatbot server 120.

Referring to FIG. 3C, the cyclic structure determination unit 220 is configured such that the cyclic structure in which the subordinate relay chatbots 326 connected to the fourth relay chatbot 324 and the fourth relay chatbot 324 are not generated is relayed. In order to check whether a cycle exists, traffic of transmission and reception of a query message and a response between the fourth relay chatbot 324 and the lower relay chatbot 326 may be monitored. In this case, the circulation structure determination unit 220 may set the relay to be performed only at the fourth relay chatbot 324 which is the upper relay chatbot.

The bot builder 230 may provide a user interface for managing the relay chatbot to the terminal of the administrator. For example, the bot builder 230 may provide a function of designing each relay chatbot corresponding to each service or adding or deleting a relay chatbot through a user interface at the administrator's terminal.

For example, the chatbot manager may access a user interface for managing a relay chatbot provided by the bot builder 230 of the chatbot server 120 through an app or the web. The chatbot manager may create a relay chatbot for relaying a specific service by using a corresponding user interface, and among the plurality of entity chatbots previously stored in the database of the chatbot server 120, the chatbot manager may match at least one chatbot suitable for the chatbot manager's relay policy. You can choose to set up a connection with a relay chatbot.

The transmitter 240 may transmit the derived response to the messenger server 110, and the response may be transmitted from the messenger server 110 to the user terminal 100. Specifically, the transmission unit 240 may transmit a response derived together with the profile information of the relay chatbot corresponding to the service selected by the user and the profile information of the chatbot from which the response is derived, to the messenger server 110, and the response may be a messenger. It may be transmitted from the server 110 to the user terminal 100.

The chatbot's profile information may include the chatbot's homepage (eg, a specific music service homepage if the chatbot is associated with a music service) and link information. At this time, if the user receives a region corresponding to the profile information of the chatbot among the responses output through the user terminal 100, it may be moved to the homepage of the chatbot.

The profile information of the relay chatbot may include, for example, ID information of the relay chatbot and information about a third party server associated with a plurality of chatbots relayed by the relay chatbot. For example, when a user receives an area corresponding to the profile information of the relay chatbot through the user terminal 100, the ID information of the relay chatbot and information on the third party server associated with the plurality of chatbots relayed by the relay chatbot are relayed. You may be taken to a page that provides.

What will be described below relates to a method for the relay chatbot to select a chatbot that can provide the most appropriate response to the user terminal 100 among the plurality of chatbots.

In one embodiment, the relay chatbot may derive scores for the plurality of chatbots and select a chatbot for providing a response to the user terminal 100 according to the derived scores. This corresponds to a non-polling model that can be performed without passing query messages to each chatbot.

For example, referring to FIG. 4A, the relay chatbot inputs a query message to the machine learning model 400 to derive a score for each of the plurality of chatbots connected to the relay chatbot, and according to the derived score, one of the plurality of chatbots. Select a chatbot and send a query message to the selected chatbot. At this time, the chatbot selected by the relay chatbot derives a response to the query message and delivers the derived response to the relay chatbot. Here, the machine learning model 400 may be a model that predicts a probability value capable of processing a query message in each of the plurality of chatbots and derives a score based on the predicted probability value.

The score is the processing rate of the chatbot that previously derived the response provided to the user terminal 100, the processing rate of the chatbot that derived the preferred response of the user terminal 100, or context information about the conversation between the user terminal 100 and the chatbot. It may be a value calculated based on the processing ratio of the chatbot that derives the response that is determined to be the most appropriate response based on the context information, and the processing ratio or the plurality of user terminals It may be a value calculated based on the processing rate of the chatbot that derived the preferred response. For example, the first relay chatbot 300 may derive a score for processing a query message for each of the plurality of chatbots 302, 304, and 306 using the machine learning model 400. When the score of the first chatbot 302 that processes the query message is 50, the score of the second chatbot 304 is 80, and the score of the third chatbot 306 is 60, the first relay chatbot 300 The second chatbot 304 having a high score may be selected, and a query message may be delivered only to the selected second chatbot 304. At this time, the second chatbot 304 may derive the response of the query message received from the first relay chatbot 300 and transfer the derived response to the first relay chatbot 300.

In another embodiment, the relay chatbot may transmit a query message to all of the plurality of chatbots, derive a ranking for responses derived from each chatbot, and select a chatbot for providing a response to the user terminal 100 according to the derived ranking. have. This corresponds to a polling model that can be performed by passing query messages to each chatbot.

For example, referring to FIG. 4B, the relay chatbot delivers a query message to all the chatbots connected to the relay chatbot, derives a ranking of responses to the query messages received from each chatbot, and then selects one response. Can be. In this case, the selected response may be transmitted to the user terminal 100 through the transmitter 240. The ranking for the response may include, for example, a probability distribution of probabilities for the response of the user of the user terminal 100 through a ranking learning algorithm, or context information (context information) about a conversation between the user terminal 100 and the chatbot. Based on the probability distribution of the probability for the response that is determined to be the most reasonable response, and may be determined based on the probability distribution of the probability for the response of the user of the plurality of user terminal. Here, the probability distribution of probabilities for the response to be preferred by the user may be generated by, for example, a response preference score assigned to each of the plurality of responses to the previous plurality of query messages for each of the plurality of services.

For example, in operation S401, the first relay chatbot 300 may transmit the user terminal 100 to each of the first chatbot 302, the second chatbot 304, and the third chatbot 306 connected to the first relay chatbot 300. Forwards the query message received from In step S403, the first relay chatbot 300 receives a first response to the query message from the first chatbot 302, receives a second response to the query message from the second chatbot 304 in step S405, In step S407, a third response to the query message is received from the third chatbot 306. In operation S409, the first relay chatbot 300 may derive a ranking for each of the received first response, second response, and third response. In step S411, the first relay chatbot 300 has a ranking rank of the received first response in the second place, a ranking rank in the received second response in the third place, and a ranking rank of the received third response in one. In the above case, a third response having a high ranking can be selected. In operation S413, the first relay chatbot 300 transmits the third response having the highest ranking to the messenger server 110 through the transmission unit 240, and in operation S415, the messenger server 110 transmits the received response to the user. It may be transmitted to the terminal 100.

In another embodiment, the relay chatbot may select a chatbot for providing a response to the user terminal 100 using a hybrid model combining a non-polling model and a polling model.

For example, referring to FIG. 4C, the relay chatbot selects two or more chatbots based on the processing scores of the query messages in each chatbot using the machine learning model, sends the query messages to the selected two or more chatbots, and selects two or more selected chatbots. The response to be transmitted to the user terminal 100 may be selected based on the ranking of the response of the query message received from the chatbot.

For example, in operation S400, the first relay chatbot 300 may input a query message as a machine learning model to derive a score for each of the plurality of chatbots 302, 304, and 306. In operation S402, the first relay chatbot 300 may have a first chatbot 302 and a second chatbot whose score is equal to or greater than a predetermined score (for example, 60 points) among the plurality of chatbots 302, 304, and 306 according to the derived score. 304) can be selected. In operation S404, the first relay chatbot 300 may transmit a query message to each of the selected first chatbot 302 and the second chatbot 304. In operation S406, the first relay chatbot 300 may receive a first response to the inquiry message from the first chatbot 302, and receive a second response to the inquiry message from the second chatbot 304 in step S408. have. In operation S410, the first relay chatbot 300 derives a ranking with respect to the received first response and the second response. In operation S412, the first response and the second response whose ranking is first are ranked. A high second response can be chosen. In operation S414, the first relay chatbot 300 may transmit the second response with the selected high ranking to the messenger server 110 through the transmitter 240. In operation S416, the messenger server 110 may transmit the received second response to the user terminal 120.

Meanwhile, if the relay chatbot fails to derive a response to the query message through a plurality of chatbots connected to the relay chatbot, the relay chatbot may directly derive a response to the query message itself.

5 is a flowchart of a method for providing an instant messaging service in a chatbot server 120 according to an embodiment of the present invention. Referring to FIG. 5, in operation S501, the user terminal 100 messengers a query message input for a service selected by a user of the user terminal 100 and service account information corresponding to the selected service. May transmit to the server 110. In operation S503, the messenger server 100 may transmit the received query message and service account information to the chatbot server 120. In operation S505, the chatbot server 120 may transmit a query message to at least one chatbot among a plurality of chatbots connected to the relay chatbot by using the relay chatbot corresponding to the service account information. In operation S507, the chatbot server 120 may derive a response to the query message by at least one chatbot. In operation S509, the chatbot server 120 may transmit the derived response to the messenger server 110, and in operation S511, the messenger server 110 may transmit the received response to the user terminal 100.

6 is an exemplary diagram of a response to a query message according to an embodiment of the present invention. For example, referring to FIG. 6, when the user terminal 100 selects a friend corresponding to a weather service, the user terminal 100 may have a conversation related to weather with a relay chatbot corresponding to the selected friend. For example, when the user terminal 100 transmits a query message 601 including 'tell me the weather' to the chatbot server 120 through the messenger server 110, the chatbot server 120 corresponds to a weather service. The response 603 of the chatbot selected through the relay chatbot may be transmitted to the user terminal 100 through the messenger server 110. At this time, the response 703 of the selected chatbot corresponds to any one response selected by the relay chatbot in a response derived from at least one chatbot among a plurality of chatbots connected to the relay chatbot corresponding to the weather service. The response 603 received at the user terminal 100 includes the profile information 605 of the chatbot which derived the response and the profile information 607 of the relay chatbot corresponding to the weather service, and the response (eg, Today's weather can be cloudy etc.)

The profile information 607 of the relay chatbot may include, for example, ID information of the relay chatbot and information about a third party server associated with a plurality of chatbots relayed by the relay chatbot. The chatbot's profile information 605 that derives the response may include link information for moving to the homepage of the chatbot. For example, when the user terminal 100 selects an area 605 corresponding to the chatbot's profile information among the responses 603 output through the messenger app, the user terminal 100 moves to the homepage of the chatbot. If the user terminal 100 selects an area 607 corresponding to the profile information of the relay chatbot, the user terminal 100 is a third party server connected to the ID information of the relay chatbot and the plurality of chatbots relayed by the relay chatbot. You can find a page that provides information about.

One embodiment of the present invention can also be implemented in the form of a recording medium containing instructions executable by a computer, such as a program module executed by the computer. 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. In addition, computer readable media 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.

Although the methods and systems of the present invention have been described in connection with specific embodiments, some or all of their components or operations may be implemented using a computer system having a general purpose hardware architecture.

The foregoing description of the present invention is intended for illustration, and it will be understood by those skilled in the art that the present invention may be easily modified in other specific forms 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 exemplary in all respects and not restrictive. For example, each component described as a single type may be implemented in a distributed manner, and similarly, components described as distributed may be implemented in a combined form.

The scope of the present invention is shown by the following claims rather than the above description, and all changes or modifications derived from the meaning and scope of the claims and their equivalents should be construed as being included in the scope of the present invention. do.

100: user terminal 110: messenger server
120: chatbot server

Claims (4)

In a chatbot server that provides an instant messaging service using a relay chatbot relaying a plurality of chatbots,
A receiving unit for receiving a query message input for a service selected by a user terminal from among a plurality of services and service account information corresponding to the selected service from a messenger server;
A derivation unit for deriving a response to the query message using a relay chatbot corresponding to the service account information; And
Transmission unit for transmitting the response to the messenger server
Including,
The relay chatbot may derive the response by at least one chatbot among the plurality of chatbots by relaying a plurality of chatbots connected to the relay chatbot with the user terminal.
At least one chatbot of the plurality of chatbots is connected to be shared with the relay chatbot and another relay chatbot,
The relay chatbot relays the first chatbot group including the at least one chatbot with the user terminal to derive the response by the first chatbot group.
And the other relay chatbot is different from the first chatbot group and derives the response by the second chatbot group by relaying a second chatbot group including the at least one chatbot with the user terminal. server.
In a chatbot server that provides an instant messaging service using a relay chatbot relaying a plurality of chatbots,
A receiving unit for receiving a query message input for a service selected by a user terminal among a plurality of services and service account information corresponding to the selected service from a messenger server;
A derivation unit for deriving a response to the query message using a relay chatbot corresponding to the service account information; And
Transmission unit for transmitting the response to the messenger server
Including,
The relay chatbot may derive the response by at least one chatbot among the plurality of chatbots by relaying a plurality of chatbots connected to the relay chatbot with the user terminal.
The relay chatbot inputs the query message as a machine learning model to the processing rate of the chatbot which derives the response provided to the user terminal, the processing rate of the chatbot which derives the preferred response of the user terminal, and the conversation between the user terminal and the chatbot. Deriving a score for each of the plurality of chatbots based on at least one of the processing rates of the chatbots that have derived a response based on the context information for the
And selecting a chatbot of the plurality of chatbots according to the derived score, delivering the query message to the selected chatbot, and the selected chatbot deriving a response to the query message.
In a chatbot server that provides an instant messaging service using a relay chatbot relaying a plurality of chatbots,
A receiving unit for receiving a query message input for a service selected by a user terminal among a plurality of services and service account information corresponding to the selected service from a messenger server;
A derivation unit for deriving a response to the query message using a relay chatbot corresponding to the service account information; And
Transmission unit for transmitting the response to the messenger server
Including,
The relay chatbot may derive the response by at least one chatbot among the plurality of chatbots by relaying a plurality of chatbots connected to the relay chatbot with the user terminal.
The relay chatbot transmits the query message to each of the plurality of chatbots, receives a response to the query message from each of the plurality of chatbots, a probability distribution of a probability of a user's preferred response of the user terminal, and the user. A ranking is derived for the received response based on at least one of the probability distributions of the probabilities for the response derived based on the context information of the conversation between the terminal and the chatbot, and the received response is determined based on the derived ranking. Select one response,
The transmitting unit sends the selected one response to the messenger server, chatbot server.
In a chatbot server that provides an instant messaging service using a relay chatbot relaying a plurality of chatbots,
A receiving unit for receiving a query message input for a service selected by a user terminal among a plurality of services and service account information corresponding to the selected service from a messenger server;
A derivation unit for deriving a response to the query message using a relay chatbot corresponding to the service account information; And
A transmission unit for transmitting the response to the messenger server; And
Circular structure determination unit for determining whether or not a circular structure in which the relay chatbot and another relay chatbot relay each other is generated
Including,
The relay chatbot is configured to relay the plurality of chatbots connected to the relay chatbot with the user terminal to derive the response by at least one chatbot among the plurality of chatbots.

Images