KR20110070000A - System and method for providing conversational service - Google Patents

Abstract

PURPOSE: An interactive service providing system and method thereof are provided to reduce a load of a server by integrally managing knowledge DBs of various areas necessary for conversation interpretations. CONSTITUTION: An area classifier(2310) determines an interpretation corresponding a conversion message received from a user terminal among a interpreted areas classified based on kinds of services. A knowledge linking unit(2330) integrally manage information necessary for interpretation of conversation message according to each interpretation area. A history manager(2320) stores history about the conversation message received from the user terminal.

Description

INTERACTIVE SERVICE PROVIDING SYSTEM AND METHOD {SYSTEM AND METHOD FOR PROVIDING CONVERSATIONAL SERVICE}

The present invention relates to a system and method for providing an interactive service, and more particularly, to provide an interactive service for classifying analysis areas based on the characteristics of an IPTV service and integrating and managing analysis rules in each analysis area. System and method.

Recently, with the spread of high-speed communication networks and convergence of broadcasting and communication technologies, IPTV (Internet Protocol TeleVision), which can selectively receive desired contents, has emerged as an issue in the broadcasting industry. Existing broadcasts could only be viewed by the viewers manually, but IPTV is attracting attention as a service that can dramatically increase the convenience and satisfaction of viewers because they can watch the content they want at the desired time. have.

In this IPTV broadcasting service, VOD contents providing service, real time broadcasting service, data broadcasting service and various additional services are provided to the user, and the IPTV broadcasting viewer can watch the broadcasting contents by operating a remote controller and perform detailed service and operation desired by the user. I could ask. However, using a remote controller to select a series of menus and input a command has been overly complicated, causing the user to be reluctant to use the service.

Therefore, in order to solve this problem, a conversation service system has been developed to receive a natural language sentence of a conversational text input by a user, interpret it, and provide a desired service to the user.

However, the current conversation service has a problem in that the application width is limited because it is impossible except for a predetermined range of conversations, and a lot of time and operation are required to interpret and process the conversation command. Since it is not easy to pre-configure the conversation service for all service areas, there is a problem that it is difficult to meet various needs of users in real time.

Furthermore, according to the characteristics of various services of IPTV broadcasting, it was not possible to provide differentiated conversation interpretation and response generation, and thus, there was a problem in that a high quality intelligent conversation service could not be provided to the user.

Therefore, there is an urgent need for a system and method for providing a high quality conversation service that can effectively interpret and respond to user queries by reflecting the types and characteristics of IPTV broadcasting services.

Some embodiments of the present invention provide an interactive service providing system and method capable of reducing the load on a server by integrating and managing a knowledge DB of various areas required for conversation analysis.

In addition, an embodiment of the present invention provides an interactive service providing system and method that can efficiently perform conversation analysis on various services of IPTV broadcasting.

In addition, an embodiment of the present invention provides a system and method for providing an interactive service capable of interpreting a conversation command more accurately by collecting a conversation history for each user.

As a technical means for achieving the above-mentioned technical problem, the 1st aspect of this invention is the area classification which determines the analysis area corresponding to the conversation message received from a user terminal among the analysis areas classified based on the type of IPTV service. A knowledge interlocking unit for integrating and managing information necessary for interpretation of the conversation message for each analysis region, a history management unit for storing a history of the conversation message received from the user terminal, an analysis rule in the determined analysis region, and On the basis of the conversation message history information of the user terminal, it is possible to provide an interactive service providing system including a conversation analysis unit for interpreting the meaning of the received conversation message.

In addition, the second aspect of the present invention, the step of converting the conversation message received from the user terminal, determining the analysis region corresponding to the received conversation message from the analysis region classified based on the type of IPTV service, Receiving rule information necessary for interpretation of the conversation message, corresponding to the determined analysis region, receiving history information of the conversation message received from the user terminal, and based on the rule information and the history information, the reception Comprising a step of interpreting a conversation message, the rule information required for the interpretation of the conversation message, can be provided by the interactive service providing method is classified and managed by the analysis area integrated.

According to the problem solving means of the present invention described above, it is possible to reduce the load on the server by integrating and managing the knowledge DB required for conversation analysis.

Moreover, according to one of the other subject solving means of this invention. Conversation analysis is efficiently performed on various services of IPTV broadcasting, and various feedback such as STB control or additional query is provided to the user.

Moreover, according to one of the another subject solving means of this invention. By collecting the conversation history for each user, it is possible to interpret the conversation commands more accurately, and can be efficiently applied to construct a conversation service that must provide a conversation for several areas at once, such as an IPTV service.

Hereinafter, 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. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. 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.

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

Hereinafter, with reference to FIG. 1, the overall configuration of a chat service according to an embodiment of the present invention will be described.

1 is an overall configuration diagram of a chat service system according to an embodiment of the present invention.

As shown in FIG. 1, a chat service system according to an exemplary embodiment of the present invention includes a user terminal 1000, a chat management system 2000, and an Internet Protocol TV (IPTV) service system 3000. In addition, the conversation management system 2000 may include a conversation interface unit 2100, a conversation control server 2200, and a conversation analysis server 2300.

The user terminal 1000 provides a conversation message related to the IPTV service to the conversation management system 2000 and receives a response to the provided conversation message.

The user terminal 1000 may request or query the provision of a desired service among IPTV services using a conversation message in a natural language format.

The user terminal 1000 includes all terminals capable of transmitting and receiving a communication system 2000, such as a wired / wireless telephone terminal, a mobile communication terminal, a personal computer (PC), a personal digital assistant (PDA), a set-top box (STB), and the like. do.

In addition, the user terminal 1000 may transmit a conversation message in various formats such as voice and text to the conversation management system 2000.

The conversation management system 2000 receives and interprets a conversation message from the user terminal 1000, and provides a response message corresponding to the interpreted conversation message to the user terminal 1000. In addition, the conversation management system 2000 generates a control command corresponding to the interpreted conversation message and provides it to the IPTV service system 3000 to be described later.

The conversation interface unit 2100 converts a conversation message received from the user terminal 1000 into a message of a preset format and provides the message to the conversation management system 2000, and provides a response message received from the conversation management system 2000 to the user terminal. To provide.

The conversation interface unit 2100 may have different types of conversation messages from a user terminal 1000 using a wired / wireless telephone, a service on IP (SoIP), a short message service (SMS), a multi-media message service (MMS), and various web interfaces. Can be received, and the conversation message can be converted into a preset format.

The chat control server 2200 receives a chat message from the chat interface unit 2100, provides the received chat message to the chat interpretation server 2300, which will be described later, and receives a response message from the chat interpretation server 2300. It is provided to the user terminal 1000.

In addition, the conversation control server 2200 generates a control command based on the type of response and the response message provided from the conversation analysis server 2300, and provides the generated control command to the IPTV service system 3000.

Since the control command generated in the conversation control server 2200 is generated based on the meaning and type of the conversation message provided by the user terminal 1000, the user can more accurately select the IPTV service desired by the user through the conversation message. You will be provided.

The conversation interpretation server 2300 receives a conversation message from the conversation control server 2200 and interprets the received conversation message.

The conversation analysis server 2300 may determine which of the preset conversation analysis areas corresponds to the conversation message, and may effectively interpret the conversation message based on the conversation message history of the user terminal 1000.

These interpretation areas are classified according to types and characteristics of IPTV services so that conversation messages can be effectively interpreted. For example, content areas related to VOD content providing service, channel areas related to real-time broadcasting service, and control of STB and Related control areas and the like.

In addition, the conversation interpretation server 2300 determines the type of response message based on the meaning of the interpreted conversation message, and generates a response message suitable for the determined type.

Detailed configuration of the conversation analysis server 2300 will be described later.

The IPTV service system 3000 receives a control command from the conversation control server 2200 and provides a user with a service corresponding to the conversation message based on the received control command.

In addition, the IPTV service system 3000 receives and manages broadcast content from a broadcaster and distributes the content to the STB to provide a broadcast / recording / playback service. The received broadcast content may include related additional information and an EPG (Electronic Program Guide) for the additional information.

The IPTV service system 3000 includes an EPG information (PSIP / PSI) including a broadcasting video and audio signal, data broadcasting data, and a program recommendation service menu through a subscriber aggregation device and a subscriber switch via a router supporting a multicast routing protocol. / SI information) can be transmitted to IP set-top boxes of multiple subscribers by multicasting.

Hereinafter, with reference to FIG. 2, a detailed configuration of the conversation analysis server 2300 according to an embodiment of the present invention will be described.

2 is a detailed block diagram of the conversation analysis server 2300 according to an embodiment of the present invention.

As shown in FIG. 2, the conversation analysis server 2300 according to an embodiment of the present invention may include a region classifier 2310, a history manager 2320, a knowledge linker 2330, an integrated knowledge DB 2340, and the like. The conversation analyzer 2350, the general response inference unit 2360, and the general response generator 2370 are included. In addition, the conversation interpreter 2350 includes a general query interpreter 2351 and a region knowledge reflector 2352.

The area classifier 2310 analyzes a conversation message received from the user terminal 1000, and determines an analysis area corresponding to the conversation message.

The area classifier 2310 may set a plurality of analysis areas in order to more accurately interpret which service the conversation message is used for. The area classifying unit 2310 may determine an analysis region to which the received conversation message belongs, and may enable the conversation analysis unit 2350 to be described later to interpret the conversation message by applying an analysis rule specific to the analysis region.

The analysis area may be classified according to the type and characteristics of the IPTV service, and may be divided into, for example, a content area related to a VOD content providing service, a channel area related to a real time broadcasting service, and a control area related to control of an STB. .

In addition, the area classifier 2310 may analyze words and morphemes included in the conversation message to determine an analysis region to which the conversation message belongs. The area classification unit 2310 may statistically determine an analysis area to which a conversation message belongs, using a classification model such as Hidden Markov Model (HMM), Conditional Random Fields (CRF), and Support Vector Machine (SVM).

For example, when a conversation message “Yes, please record a drama with Kim Hye-ja” is received, the area classification unit 2310 may search for words and morphemes such as “yesterday,” “drama,” “recording,” and “request.” Based on this, it may be determined that such a conversation message belongs to a 'content area'.

The history manager 2320 stores a history of conversations with the user terminal 1000.

The history manager 2320 may store the conversation message received from the user terminal 1000 and the response message generated corresponding to the conversation message for each user terminal 1000.

Such history information may be used to interpret a conversation message, which will be described later, and to generate a response message corresponding thereto. According to an embodiment of the present invention, the conversation interpretation server 2300 may interpret the conversation message of the user terminal 1000 more accurately based on the conversation history with the user terminal 1000 and generate a response message corresponding thereto. It becomes possible.

The knowledge linker 2330 integrates and manages various pieces of information necessary for interpretation of the conversation message and generation of the response message for each analysis area.

The knowledge linker 2330 may classify and manage various types of information stored in the integrated knowledge DB to be described later, and the domain classifier 2310, the general query interpreter 2351, the domain knowledge reflector 2352, and the general response inference The unit 2360 and the general response generator 2370 and the integrated knowledge DB 2340 to be described below may be linked to each other.

In addition, the knowledge linker 2330 may convert or code information such as analysis rules and region knowledge stored in the integrated knowledge DB 2340 and provide the same to the conversation interpreter 2330. The knowledge interlocking unit 2330 may provide interpretation dialogs 2330 by encoding analysis rules, region knowledge, and the like stored in a script form in the form of a C language.

Since the knowledge interlocking unit 2330 integrates and manages analysis rules and area knowledge of various analysis areas, the conversation analysis server 2300 may enable the conversation message server 2300 to interpret the conversation message more quickly and with less load.

The integrated knowledge DB 2340 integrates and stores various pieces of information necessary for interpreting the conversation message and generating the response message.

The integrated knowledge DB 2340 may classify and store various pieces of information necessary for interpreting a conversation message and generating a response message for each analysis area such as a content area, a channel area, and a control area.

In addition, the integrated knowledge DB 2340 may integrate and store information such as analysis rule, region knowledge, inference rule, and response rule for each analysis region.

Since the integrated knowledge DB 2340 can store information such as interpretation rules, domain knowledge, inference rules, and response rules in a script format such as XML, it is possible to effectively integrate and manage the information, and more efficiently. Can be processed.

Furthermore, in the integrated knowledge DB 2340, such rules may be stored as a pair of conditions and corresponding actions, and thus, it may be determined which condition the user's input sentence corresponds to. As a result, it is possible to provide a quick and accurate response to the input and feedback.

The conversation interpreter 2350 interprets the meaning of the conversation message based on an analysis rule in an analysis area corresponding to the conversation message and a conversation message history of the user terminal 1000.

The general query interpreter 2351 interprets the conversation message by using an interpretation rule corresponding to the analysis region to which the conversation message belongs.

The general query interpreter 2351 may request an interpretation rule corresponding to an analysis region to which the conversation message belongs, to the knowledge interlocking unit 2330, and receives an interpretation rule from the knowledge interworking unit 2330 to primarily request a conversation message. Can be interpreted At this time, the analysis rule of the analysis region may be pre-converted into a form that can be received in advance by the knowledge interlocking unit 2330 and interpreted by the general-purpose query analysis unit 2351.

For example, when the general query interpreter 2351 receives a dialogue message “Please, please record a drama with Kim Hye-ja yesterday,” {'date = yesterday = October 21', 'Hye-Jae = celebrity name' , 'User intention = recording reservation', and 'service area = content area'}.

In addition, the general purpose query analysis unit 2351 analyzes the received conversation message based on the conversation history with the user terminal 1000.

The general-purpose query interpreter 2351 receives a conversation message history of the user terminal 1000 from the history manager 2320, which will be described later, and based on the received conversation message history, instructions, pronouns, and the like included in the conversation message may be used. Can understand the exact meaning.

For example, when the general-purpose query interpreter 2351 receives a conversation message "Please request recording of a drama that was recorded yesterday", the general-purpose query analysis unit 2351 is included in the conversation message history of the user terminal 1000 received from the history management unit 2320. Based on this, it is possible to interpret what drama was recorded yesterday.

The area knowledge reflecting unit 2352 analyzes the conversation message using the region knowledge corresponding to the analysis region to which the conversation message belongs.

The area knowledge reflecting unit 2352 may receive information about the area knowledge corresponding to the analysis area to which the conversation message belongs, through the knowledge interlocking unit 2330. The meaning of the conversation message may be determined based on the analysis area provided. Can be interpreted correctly.

For example, when the conversation message is “Please record the drama that Kim Hyeja came out yesterday,” the knowledge linkage unit 2330 may link to the area knowledge corresponding to the “content area”, and the area knowledge reflector 2235 ) Inputs the meaning-interpreted result into {content name = 'mother is horned', main character = 'Kim Hye-ja; Ko So-young; Jang Dong-gun;… ", Airing date =" October 21 ", turn =" 19 times "," paid = 1500 won "} and retrieved this information: {celebrity, protagonist, date, airing date, recording reservation, content By reflecting the knowledge of the domain used in the IPTV area such as the name + turn}, it can be finally interpreted as the meaning of the user who intends to reserve the recording of "Mom is Horny 19 times".

The general response inference unit 2360 determines a type of a response to the conversation message based on the meaning of the interpreted conversation message.

Based on the meaning of the dialogue message, the universal response inference unit 2360 may generate a control command for providing the IPTV service, provide an acknowledgment message prior to the control, or if there is insufficient information in the meaning of the dialogue message. The type of response may be determined, such as whether to provide the user terminal 1000 with a response message. The general response inference unit 2360 may statistically interpret and determine what response to various input forms, or may be determined through a rule or example search. As the rule used for the reasoning rule and the example search, only a rule specific to the "content area" obtained through the knowledge linker 2330 may be used.

In addition, the general response inference unit 2360 may use the history of past conversation messages provided by the history manager 2320 to determine the type of response. That is, the general response inference unit 2360 may determine the type of response in consideration of the case where the 'material' is repeated more than a certain number of times in the past conversation message history or when the information provided by the user is already mentioned in the previous conversation. have.

The general response generator 2370 generates a response message corresponding to the conversation message based on the determined response type.

The general response generator 2370 may generate a response message in the form of a natural language sentence based on the information interpreted from the conversation message of the user and the type of the determined response.

The general response generating unit 2370 generates, for example, “Would you like to record 19 times (Mom Horny, Paid = 1500 Won)” as an acknowledgment message or “Record” as a notification response message for the control result. Reservation completed. ? Mom is horned, paid = 1,500 won, 19 times "can generate a response message. In addition, the response message is provided to the user terminal 1000 through the conversation interface unit 2100.

Meanwhile, the conversation control server 2200 records content based on the generated response message when the information received from the conversation analysis server 2300 is {response type = "control / recording" analysis region = "content"}. The control command may be generated and provided to the IPTV service system 3000.

Hereinafter, an interactive service providing method according to an embodiment of the present invention will be described with reference to FIG. 3.

3 is a detailed flowchart of an interactive service providing method according to an embodiment of the present invention.

Step S300 is a step of converting a conversation message.

In operation S300, a different type of conversation message may be received from the user terminal 1000 using a wired or wireless telephone, a service on IP (SoIP), a short message service (SMS), a multi-media message service (MMS), and various web interfaces. In addition, these conversation messages can be converted into a preset format.

Step S302 is a step of classifying analysis areas.

In operation S302, an area for interpreting a conversation message may be classified based on the type and characteristics of the IPTV service, and the analysis message may be analyzed to determine an analysis region corresponding to the conversation message. In addition, in step S302, an analysis region to which a conversation message belongs can be statistically determined using a classification model such as Hidden Markov Model (HMM), Conditional Random Fields (CRF), and Support Vector Machine (SVM).

Step S304 is a step of receiving analysis rules and area knowledge for each analysis area.

In operation S304, an analysis rule and area knowledge corresponding to the analysis area to which the conversation message belongs may be received through the knowledge linker 2330. In addition, the interpretation rule and the region knowledge of the analysis region may be pre-converted to a form that is received in advance by the knowledge linker 2330 and interpreted by the general-purpose query analysis unit 2351. In addition, since the knowledge linkage unit 2330 integrates and manages information such as analysis rules, domain knowledge, inference rules, and response rules for each classified analysis domain, information such as analysis rules and domain knowledge can be utilized more efficiently. Will be.

Step S306 is a step of receiving conversation history information. In operation S306, a conversation message and a response message transmitted and received with the corresponding user terminal 1000 may be received.

Step S308 is to interpret the conversation message.

In operation S308, the conversation message may be primarily interpreted using analysis rules and region knowledge corresponding to the determined analysis region, and the received conversation message may be secondarily based on the conversation history with the user terminal 1000. Can be interpreted In particular, the conversation history can be used to interpret the exact meaning of an ambiguous expression, such as an directive, in a received conversation message.

Therefore, it is possible to efficiently use the analysis rule of the analysis area to which the conversation content belongs, and to interpret the conversation content more accurately and quickly by reflecting the conversation history.

In operation S310, the type of response corresponding to the conversation message is inferred.

In operation S310, the type of response corresponding thereto may be inferred based on the meaning of the interpreted conversation message. If the interpreted conversation message is a command for controlling the IPTV service system 3000 or STB, the type of response may be inferred as 'control'. Can be inferred from 'the material'. In addition, when the meaning of the interpreted conversation message is to be confirmed once again, the type of response may be inferred as 'confirmation'. The determination in step S310 can be performed more accurately by using the inference rule of the corresponding analysis region provided through the knowledge linker 2330.

Step S312 is a step of determining whether to generate a control command based on the inferred response type.

As a result of the determination in S312, when generating a control command, a response message and a control command are generated, and the generated response message and the control command are transmitted to the user terminal 1000 and the IPTV service system 3000, respectively (step S314).

If the control command is not generated as a result of the determination in S312, a response message is generated and transmitted to the user terminal 1000 (step S316). The response message may be generated in the form of a natural language sentence based on the meaning of the interpreted conversation message and the type of the inferred response, and the control command may be applied by the IPTV service system 3000 based on the generated response message. It may be generated in a preset format.

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 both computer storage media and communication 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. Communication media typically includes computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave, or other transmission mechanism, and includes any information delivery media.

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. It is therefore to be understood that the above-described embodiments are illustrative in all aspects 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.

In addition, the present invention can be efficiently applied to construct a dialogue service that must provide a dialogue for several areas at the same time as an IPTV service.

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.

1 is an overall configuration diagram of a chat service system according to an embodiment of the present invention.

2 is a detailed configuration diagram of a conversation analysis server 2300 according to an embodiment of the present invention.

3 is a detailed flowchart of an interactive service providing method according to an embodiment of the present invention.

* Explanation of symbols on the main parts of the drawings

1000: user terminal 2000: conversation management system

2100: conversation interface unit 2200: conversation control server

2300: conversation interpretation server 3000: IPTV service system

Claims (11)

In the interactive service providing system, An area classification unit for determining an analysis area corresponding to a conversation message received from a user terminal among analysis areas classified based on a type of service; A knowledge interlocking unit that integrates and manages information necessary for interpretation of a conversation message for each analysis area; A history manager which stores a history of conversation messages received from the user terminal; A conversation interpreter that interprets the meaning of the received conversation message based on an analysis rule in the determined analysis region and conversation message history information of the user terminal; Interactive service providing system comprising a. The method of claim 1, A universal response inference unit for determining a type of response message to be generated corresponding to the conversation message, based on the meaning of the interpreted conversation message; A general response generating unit generating a response message corresponding to the conversation message based on the type of the determined response message and the meaning of the interpreted conversation message; More, And the generated response message is converted into a message format of the user terminal and provided to the user terminal. The method of claim 2, A conversation control server generating a control command corresponding to the conversation message based on the type of the response message and the generated response message and providing the generated control command to an IPTV service system; Interactive service providing system further comprising. The method of claim 2, The area classification unit classifies the analysis area into one or more of an area related to a video on demand (VOD) content providing service, an area related to a real-time broadcasting service, and an area related to control of a set-top box (STB). Interactive service delivery system. The method of claim 2, And the knowledge linking unit receives information necessary for interpretation of a conversation message in advance and converts the information in advance into a form that the conversation analysis unit can interpret. The method of claim 5, And information necessary for interpretation of the conversation message includes one or more of interpretation rules, domain knowledge, inference rules, and response rules. The method of claim 2, And the conversation analysis unit uses the conversation message history information to interpret an instruction word and a pronoun included in the conversation message. In the interactive service providing method, Converting a conversation message received from the user terminal; Determining an analysis area corresponding to the received conversation message from among analysis areas classified based on a type of service; Receiving rule information necessary for interpretation of the conversation message, corresponding to the determined analysis region, Receiving history information of a conversation message received from the user terminal; and Interpreting the received conversation message based on the rule information and the history information Including; Rule information necessary for interpretation of the conversation message is classified by the analysis area is integrated management method. The method of claim 8, Determining a type of response message generated in response to the received conversation message, based on the interpreted conversation message; Generating a response message corresponding to the received conversation message based on the determined type of response message and the interpreted conversation message; and Providing the generated response message to the user terminal Interactive service providing method further comprising. The method of claim 9, Generating a control command corresponding to the received conversation message based on the type of the response message and the interpreted conversation message; and Providing the generated control command to an IPTV service system Interactive service providing method further comprising. The method of claim 9, Receiving the rule information, the rule information is received in advance, And converting the received rule information into a format in which the conversation message can be interpreted in advance.

Images