KR20040094234A - User context assay method using interactive interface - Google Patents

Abstract

PURPOSE: A method for analyzing a user situation using an interactive interface is provided to offer a client with information optimized to the client and a communication channel by generating raw data, as the service fit to the situation most collected through the interactive interface is selected/called. CONSTITUTION: An optimal service of a user is selected based on the situation information collected by the ChatBot(S100). The selected service is provided to an information collector by adding profile and context of the user(S110). The first combined service data newly added to the service is transferred to the ChatBot(S120). Various kinds of service information associated with the service are requested to the information collector(S130). The second combined service data newly added to the service is transferred to the ChatBot(S140).

Description

User context assay method using interactive interface

The present invention relates to a user context analysis method using an interactive interface. More specifically, the present invention relates to a technology capable of providing optimized information most suitable for a client and a communication channel by generating raw data by selecting and calling a most suitable service based on context collected through an interactive interface.

Today, the Internet has grown exponentially and contains huge amounts of information.

Such use of the Internet is increasing in importance with the spread of high-speed Internet networks.

Various solutions for providing desired information to specific audiences using the Internet have been proposed, but information sharing is made through passive information listing of information providers rather than users receiving information.

For example, in the case of an information search engine such as Yahoo or Naver, when a user inputs a word related to information that the user wants to find, the web document or homepage including the word is searched and provided to the user.

However, in the conventional information providing method such as the above example, the user needs to input contents necessary for one day from a search site such as Yahoo where the user can search the information, and the user must go through the checking and sorting of the searched information contents. It involves cumbersome work.

On the other hand, mobile terminals are mainly used for the purpose of making calls, and some of them use the wireless Internet to obtain information, but the limitations and economic costs of mobile terminal screens, the limitations of information provided, and one-way limited information providers have prepared in advance. Its use was limited due to the problem of unilateral acceptance of information and the inconvenience of access and navigation.

In order to solve this problem, a patent for a device and method for providing search information to a wireless terminal has been filed (Publication No. 2003-0000808).

The patent has been set up the interests and search terms, the priority of the wireless terminal subscribers to access the Internet, the search terms set by the subscriber is displayed in the Internet search items of the wireless terminal, the search terms to receive information When selected, the present invention relates to an apparatus and a method for receiving information by a wireless terminal.

However, the patent has been inconvenient to set by inputting the interests, search terms, priorities using the wireless terminal, there is a problem to know the correct search word, the information is provided only to the wireless terminal.

In addition, from the user's point of view, it was not interesting to simply enter the characters in the input window.

SUMMARY OF THE INVENTION The present invention has been made to solve the above problems, and an object of the present invention is to select the most suitable service based on the context collected through the interactive interface and to call and generate the raw data to optimize the best information for the client and communication channel It is to provide a user context analysis method using an interactive interface that can provide.

As the technical idea for achieving the above object of the present invention

In the user context analysis method for collecting the user's context information based on the characters entered using the interactive interface, and selecting and calling the most suitable service based on the collected context information,

A first step of selecting an optimal service of a user based on the collected context information in a ChatBot;

A second step of attaching a profile and a context of the user to the selected service and providing the profile to the information collector;

Transmitting a first combined service data newly added to the service to a ChatBot;

Requesting an information collector for various service information associated with the service;

And a fifth step of delivering the second combined service data newly added to the service to a ChatBot.

1 is a block diagram of a user context analysis system using an interactive interface according to the present invention.

2 is a flowchart illustrating a user context analysis process using an interactive interface according to the present invention.

3 is a diagram illustrating a context mining table according to an embodiment of the present invention.

<Description of Symbols for Major Parts of Drawings>

10: user terminal 20: Connection Manager

30: Dispatcher 40: Interface Data Processor

50: ChatBot 55: Context mining table

60: Context / Profile 65: User Propensity DB

70: system infrastructure (Infra) 75: LSB DB

80: Information Collector 90: CP content

Hereinafter, with reference to the accompanying drawings, the configuration and operation of the embodiment of the present invention will be described in detail.

1 is a block diagram of a user context analysis system using an interactive interface according to the present invention.

1, a user terminal 10 for inputting an interactive character using a user character; Connection manager 20 for connection status management and log management; Dispatcher 30 for classifying and transmitting messages; An interface data processor 40 for input / output management, separation, and character interworking; A ChatBot 50 having a context mining table 55 for chat and input sentence preprocessing, dialogue rule application and learning; A context / profile 60 having a user propensity DB 65 and including demographic information such as age, gender, address, and location information time information; A system infrastructure (Infra) 70 having an LSB DB 75 for smooth operation of the system; An Information Collector 80 including a Profile Collector, a Context Collector, and a Service Invocator; CP content 90 for providing various contents.

The user terminal 10 is an interactive interface terminal device through a wired / wireless internet network and includes various mobile terminals such as a PC, a virtual machine (VM), and an IMT-2000.

The connection state management function in the connection manager 20 includes 1) creating and maintaining a Transmission Control Protocol (TCP) connection when a virtual application connects to a server. 2) Do not exceed a certain number of connections to each server. 3) transmits the setting (character, item, version, etc.) of the terminal to the state management. 4) It handles connection functions such as chat and character / item download. 5) When connecting, deliver the content or message to download from D / L on connect table. The function of log management is to generate a billing log when billing is required, and to generate a log for all input / output.

The dispatcher 30 determines a header of a received packet for classification and transmission for each message, and delivers a message to a processor that will process the packet.

The interface data processor 40 serves as input / output management, parsing, and character interworking. The parsing function converts input text data into a form suitable for chatbot input and separates commands from the input data. Pre-processing. In addition, the character linkage function serves to map the skin and the movement of the character in consideration of the likelihood to the response of the ChatBot (50) to deliver to the output manager.

The ChatBot 50 allows general conversation, learning, and voting in a natural language with a user, and performs a function by interworking with another module for each service.

The basic function is to: 1) emit a matching response to the forwarding input from the I / F data processor (40). 2) Exact matching and inference matching are used for pattern matching with AIML Rule, and Exact matching is used for pattern matching with DB Rule. 3) If there is a request for content (news, weather, humor, horoscope, dictionary) during the conversation, the desired content is provided. 4) Each character has its own background and episode. 5) You can learn the answer to a specific question from the user, you can get a user vote for a specific answer. 6) Provide additional services during the conversation, such as, for example, kung kung game, sending email, sending SMS. 7) Send and receive data to and from I / F dataProcessor (40) through TCP connection.

Here, the context mining table 55 includes a serial number (ID) for a service to be performed as shown in FIG. 3, information about a service to be performed (HTTP, Web service), a service rule determined by the system, and Information on the behavior of formatting rules that converted the generated data is provided.

The LSB DB 75 of the system infrastructure (Infra 70) provides intelligent location information by providing a user's current location information when using weather information.

The Profile Collector in the Information Collector 80 collects the customer's profile and delivers it at the request of ChatBot 50, and provides customer subscription information, terminal information, service usage information, and customer characteristic information. Do it.

The Context Collector collects contexts such as the location and time of the customer, delivers them to the ChatBot 50 request, and provides the location and current time of the customer through LBS linkage.

The Service Invocator converts the content to be linked to the form requested by ChatBot 50, manages and provides it, and provides information such as weather, fortune telling, news, dictionary and humor.

2 is a flowchart of a user context analysis method using an interactive interface according to the present invention.

Through the dialog input window of the user terminal 10, dialog information such as "I'm hungry" is input. The input conversation information is transmitted to the I / F data processor 40 via the Connection Manager 20 and the Dispatcher 30.

At this time, the dispatcher 30 analyzes the header of the conversation information. In addition, the I / F data processor 40 analyzes input / output packets with the ChatBot 50, inquires user information, and transmits the information to the ChatBot as an appropriate input pattern. Process the response and return it to the user.

Thereafter, through the parser, the extracted conversation information, for example, "hungry" is extracted, and the user's profile in the ChatBot 50 is based on the extracted conversation information. Context, user input information, user propensity, and usage history.

Here, the ChatBot 50 performs general conversation, learning, and voting in natural language with the user, and performs a function by interworking with another module for each service.

Then, it is determined whether a service is suitable for a user based on the recognized context information.

Subsequently, the following process collects the user's context information based on the characters entered using the interactive interface, and analyzes the user's context for selecting and calling the most suitable service based on the collected context information. .

First, the ChatBot 50 selects an optimal service suitable for a user based on the collected context information. At this time, the ChatBot 50 uses the Context Mining table 45 as shown in FIG. 3 to refer to a "restaurant recommendation service" corresponding to an optimal service for user context information, for example, "hungry." It will be provided (S100).

Next, the user inputs and attaches the profile and the context of the selected service to the information collector 80 (S110).

At this time, the information collector 80 executes information on the service selected from the CP content 90, receives the generated data, and delivers the first combined service data newly added to the service to the ChatBot 50. (S120).

Next, the ChatBot 50 requests the information collector 80 for various service information associated with the service, that is, a reservation service, a coupon service, a map information, and a friend search service (S130).

At this time, the information collector 80 executes the information on the service selected from the CP content 90, receives the generated data and delivers the second combined service data newly added to the service to the ChatBot 50. It is made (S140).

In the present invention, a transformation language such as XSLT (extensible Stylesheet Language Transformation) is used to use the result of the first combined service as an input of the second combined service. In addition, only a desired value may be extracted from the result of the first combined service and the result of the second combined service to form one result data.

As described above, according to the present invention, by selecting and calling the most suitable service based on the context information of the user collected through the interactive interface, it is possible to provide optimized information most suitable for the client and communication channel by generating raw data. have.

Claims (5)

In the user context analysis method for collecting the user's context information based on the characters entered using the interactive interface, and selecting and calling the most suitable service based on the collected context information, A first step of selecting an optimal service of a user based on the collected context information in a ChatBot; A second step of attaching a profile and a context of the user to the selected service and providing the profile to the information collector; Transmitting a first combined service data newly added to the service to a ChatBot; Requesting an information collector for various service information associated with the service; And a fifth step of delivering the second combined service data newly added to the service to a ChatBot. The method of claim 1, wherein the optimal service of the user selected in the first step is a format obtained by converting a serial number of a service to be performed, information on a service to be performed, a service rule determined by the system, and generated data. A method of analyzing a user context using an interactive interface, using a context mining table containing information on formatting rules. The method of claim 1, wherein the service information associated with the service selected in the fourth step includes a reservation service, a coupon service, a map information, and a friend search service. The interactive interface of claim 1, wherein a transformation language such as extensible stylesheet language transformation (XSLT) is used to enable the result of the first combined service to be used as an input of the second combined service in the fifth step. User context analysis method. The user context analysis using the interactive interface of claim 1, wherein only the desired value is extracted from the first combined service result of the third step and the second combined service result of the fifth step to form one result data. Way.