KR0173921B1 - User Information Input Processing and Knowledge-Based Call Access Control in Intelligent Electronic Secretary Service - Google Patents

Abstract

The present invention relates to a user information input processing method and a knowledge-based call connection control processing method for performing functional call connection processing through knowledge-based inference by receiving user information, and receiving a user's schedule or call-related instructions. Knowledge-based call access control that selects and executes appropriate call processing using user information and knowledge base rules stored in user information database after verifying and supplementing with grammatical error checking and semantic error checking. In order to provide a processing method, a first step (51 to 54) for receiving a schedule or call-related instructions; A second step 55 to 57 for performing grammatical error checking; And a third step (58 to 61) of performing semantic error checking, the first step (71 to 74) of inferring the state of the called party and the called party service characteristic; Second steps 75 to 80 for inferring call instructions, caller status, and calling party service characteristics; And the third step (81 to 85) of extracting a call processing method or performing call processing through resolving contradiction, increasing efficiency and convenience as if supported by a personal assistant, and at the network level, call completion rate and service This has the effect of improving satisfaction.

Description

User Information Input Processing and Knowledge-Based Call Access Control in Intelligent Electronic Secretary Service

1 is a structural diagram of an electronic assistant agent to which the present invention is applied.

2 is a diagram illustrating an embodiment of a service characteristic according to a user state.

3 is a diagram illustrating an embodiment of a call processing method according to service characteristics.

4 is an exemplary diagram of user information.

5 is a flowchart illustrating an embodiment of a user information input process according to the present invention.

6 is a flowchart illustrating an example of a knowledge-based call access control process according to the present invention.

* Explanation of symbols for main parts of the drawings

11: user information input subsystem 12: user information database

13: Knowledge Base 14: Service Control Subsystem

The present invention relates to a user information input processing method and a knowledge-based call access control processing method for receiving intelligent user input and performing intelligent call access processing through knowledge-based inference in an intelligent electronic assistant service.

An electronic secretary agent is a kind of software object or application process that performs instructions and control of a service on behalf of a user.

In the future, it is expected that more diverse and complex communication services will emerge according to user's needs. However, from the user's point of view, the amount of knowledge about the communication and services required to use these advanced services will increase dramatically, resulting in paradox avoidance of use due to complex usage procedures. Intelligent electronic assistant service using artificial intelligence technology is needed to increase the usefulness and efficiency and to make personalized service universal.

Accordingly, the present invention has been made to solve the above requirements, and in implementing an electronic secretary agent that provides an intelligent electronic secretary service, it imposes a knowledge processing capability on the terminals of telephone users and mutually The purpose of the present invention is to provide a user information input processing method and a knowledge-based call access control processing method that can provide users with convenience related to call processing.

In the present invention for achieving the above object, in the user information input processing method applied to the electronic assistant agent system, after performing the authentication process for the user to confirm the user registration and input the schedule or call-related instructions Receiving first step; Performing a grammatical disability check on the input user information; And a third step of storing the user information in the user information database after performing the semantic fault check on the input user information.

In addition, the present invention, in the knowledge-based call access control processing method applied to the electronic assistant agent system, after receiving a call connection request message, searching the schedule to infer the status of the called party, Inferring a called party service characteristic; Retrieving call instructions, inferring caller status, and inferring calling service characteristics using the knowledge base; And a third step of extracting a call processing method having mutually suitable service characteristics by comparing the called party service characteristic and the calling party service characteristic, and performing a corresponding call processing according to the existence of a suitable call processing method.

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

1 is a structural diagram of an electronic assistant agent to which the present invention is applied and includes a user information input subsystem 11, a user information database 12, a knowledge base 13, and a service control subsystem 14. .

The electronic secretary agent consists of a user information input subsystem 11 and a service control subsystem 14.

The user information input subsystem 11 receives a user's schedule or call-related instructions, verifies and supplements it, and stores the information in the user information database 12. These functions are performed in the user interface module and the user information ban module. Interaction with the user through the user interface is made of a personal computer or a general telephone. Depending on the type of terminal, a graphic interface, a speech processing, or a digit input method according to voice guidance may be used.

The service control subsystem 14 is a communication control module that is in charge of communication with the counterpart electronic secretary agent, an inference control module that determines an appropriate call processing method using user information and rules of the knowledge base, and a call processing method selected as a result of inference It consists of a call processing module that controls the switch and the like to perform the.

Two types of information are stored in the electronic secretary agent, one of which is a user information database 12 that holds relatively dynamically changing information such as user schedules and instructions, and static information such as knowledge of services for intelligent reasoning. There is a knowledge base 13 having.

2 is a diagram illustrating an embodiment of a service characteristic according to a user state.

There are two types of user's statuses: In a Hurry and Normal. In the case of the called party, there are three kinds of status: Busy, Normal, and Absent.

The service characteristics are not delayed and delayed depending on whether or not delay is possible, and classified into two-way, one-way, and no-connection according to the call type.

There is an appropriate service characteristic according to the user's state, and there is an inappropriate service characteristic. This content is coded into knowledge about the service and stored in the knowledge base 13.

3 is a diagram illustrating an embodiment of a call processing method according to service characteristics.

Call handling methods include Normal Connection, which means a real-time two-way call, Call Forwarding, which changes the called party's receiving location, and Call Redirection, which connects someone other than you. ), Call Interupt forcing a call from the called party who is currently talking to a third party and requesting a call, and re-calling automatically attempting to connect to the calling party when the called party ends the call. Back), voice mail for storing the message of the calling party so that it can be searched and listened later, and close connection for rejecting the connection request and terminating the call connection.

The characteristics of the necessary services are determined according to the situation in which the call request occurs, and there are processing methods that can be taken according to the characteristics of each service. Knowledge about this also exists in the knowledge base 13, and these contents can be modified by a change of the knowledge base 13 or a learning function.

4 is an exemplary diagram of user information.

The schedule represents the location of the user's time zone, and when the knowledge that can infer the user's state is applied according to the place where the user is located, the user's state can be determined by the current time.

The call instruction is inputted in the form of a rule when a particular caller wants special treatment. In the example, when the user moves to another place, Mr. P's call connection request is handled as a call transfer, and the call connection request of the other person indicates an instruction to use the voice mailbox.

5 is a flowchart illustrating an embodiment of a user information input process according to the present invention.

First, by performing a necessary authentication procedure for the user by a password checking method (51) to confirm the user registration (52), if the user is not confirmed that the registration is rejected by the announcement and terminates the input (53).

However, if the user is registered, the user receives a schedule or call-related instruction through the user interface (54) and checks whether there is a grammatical obstacle in the input content (55) to determine whether there is a failure (56). For example, in the case of call instructions, some of the instructions may be omitted. If call forwarding is requested, the grammatical disability check should be found and requested supplementation if the call forwarding number is not specified. This grammatical knowledge is determined in advance, so check it.

It determines whether there is a grammatical disorder (56), if a disorder occurs, it is requested to supplement through interaction with the user and reflects the result (57) and then judges the presence of the disorder again (56) all grammatical disorders are corrected and If it does not exist, it checks whether there is no semantic disorder (58) and determines whether there is a disorder (59). Semantic disorders occur when there is a contradiction between several instructions. For example, if there is already an instruction to refuse call connection when there is a meeting with the user information shown in Fig. 4, the processing of Mr. P phone between 10 and 12 o'clock should be forwarded or voiced. It is ambiguous whether it is necessary to process a mailbox or the like. The step of checking whether the newly inputted user information is inconsistent with the existing information is called a semantic disorder checking step.

Determining the existence of a semantic disorder (59), if a disorder occurs, requires supplementation through interaction with the user and reflects the result (60) and then judges the existence of the disorder again (59). If not present, the entered user information is stored in the user information database 12 and the process is terminated (61).

6 is a flowchart illustrating an embodiment of a knowledge-based call connection control process according to the present invention.

First, when the communication control module receives the call connection request message (71), it interprets it and delivers it to the inference control module. The message includes the calling party's name, job title, the called party's relationship, the called party's name, the importance of the call, the calling party's preferred calling method, and so on.

The inference control module identifies the called party from the message and retrieves the corresponding schedule from the user information (72). By comparing the current time and the schedule, the user's location is extracted and the recipient's status is inferred by applying the relevant knowledge (73). Using the knowledge of the appropriate service characteristics and the inappropriate service characteristics for each user state stored in the knowledge base 13, the service characteristics necessary for the called party and the service characteristics to be avoided are discussed (74).

Next, retrieve the relevant call instruction based on the caller's name (75). The caller's status is inferred according to the calling party's position, the importance of the call, the instruction (76). It is determined whether additional information is needed (77), and if necessary, a message for requesting additional information is sent to the calling electronic secretary agent by using the call control module (78). If a response message to a request for additional information is received from the calling party (79) or no additional information is required, the sender can use the knowledge of the appropriate service characteristics and the inappropriate service characteristics for each user state stored in the knowledge base 13 to the sender. The necessary service characteristics and service characteristics to be avoided are inferred (80).

Next, a call processing method having mutually suitable service characteristics is extracted by comparing a called party service characteristic and a calling party service characteristic (81) to determine whether a suitable call processing method exists (82) when only one call processing method exists. Instructs the call processing module to execute the method to execute (85).

As a result of the determination, if there are a plurality of call processing methods, the call processing module is executed to compare the called party's preference information with the caller's preference information, select a call processing method having the highest preference (83), and perform the corresponding method. (85).

If a call processing method conforming to one of the service characteristics has an unsuitable service characteristic to the other party, and if there is no appropriate call processing method, a conflict resolution step may be performed (84). If the method is selected, this is done and terminated (85). In the contradiction solution, there is a method of applying the largest priority rule using a priority value predefined in the rules in the knowledge base 12 and a method of giving a user a choice. Meta information about the rule itself, such as priority adjustment for resolving contradictions, may be reinforced by the learning function.

As described above, the present invention, after verifying and supplementing the user's schedule (schedule) and call-related instructions through the grammatical disorder test and semantic disorder test, and stores in the user information database and the rules of the knowledge base ( By selecting and conducting appropriate call processing using rules, users can handle their business very conveniently and efficiently as if they are supported by a personal assistant when they perform their work. There is an effect that can be significantly improved. In particular, the present invention can build a knowledge-based communication network when it is combined with a communication service such as Universal Personal Telecommunication (UPT) by intelligentizing the terminal in accordance with the evolving communication network to distribute the call processing function to the terminal. That has a very excellent effect.

Although the technical idea of the present invention has been described in detail according to the above preferred embodiment, it should be noted that the above-described embodiment is for the purpose of description and not of limitation. In addition, those skilled in the art will understand that various embodiments are possible within the scope of the technical idea of the present invention.

Claims (11)

A user information input processing method applied to an electronic assistant agent system, comprising: a first step of receiving a schedule or call-related instruction after confirming user registration by performing an authentication procedure for a user; Performing a grammatical disability check on the input user information; And a third step of storing user information in a user information database after performing a semantic fault check on the input user information. 4. The method of claim 3, wherein the first step comprises: a fourth step of confirming user registration by performing a necessary authentication procedure for the user using a password checking scheme; A fifth step of rejecting and terminating the input through guide broadcast if it is not confirmed that the user is registered in the fourth step; And a sixth step of receiving a schedule or call-related instruction through the user interface when the user is registered in the fourth step. 4. The method of claim 3, wherein the second step comprises: a fourth step of determining whether there is a disorder by checking whether there is a grammatical disorder in the input content; A fifth step of requesting supplementation through the interaction with the user and reflecting the result if the obstacle exists in the fourth step, and then proceeding to the fourth step; And a sixth step of performing the third step if the failure does not occur in the fourth step or if the grammar disorder is corrected and there is no failure. 4. The method of claim 3, wherein the third step comprises: a fourth step of determining whether a disorder exists by examining whether there is no semantic disorder; A fifth step of requesting supplementation through the interaction with the user and reflecting the result if the obstacle exists in the fourth step, and then proceeding to the fourth step; And a sixth step of storing the inputted user information in a user information database and terminating if a failure does not occur in the fourth step or if all of the semantic obstacles are corrected and there is no failure. In the knowledge-based call access control processing method applied to an electronic secret agent agent system, the call connection request message is input, the schedule is searched to infer the status of the called party, and then the receiver's service characteristics are inferred using the knowledge base. A first step of doing; Retrieving call instructions, inferring caller status, and inferring calling service characteristics using the knowledge base; And a third step of comparing a called party service characteristic with a calling party service characteristic to extract a call processing method having mutually suitable service characteristics, and performing a corresponding call processing according to the existence of a suitable call processing method. Based Call Access Control Processing Method. 8. The method of claim 7, wherein the first step comprises: a fourth step of receiving a call connection request message and identifying a called party from the message to retrieve a corresponding schedule from user information; A fifth step of inferring the called party's state by extracting the user's location by comparing the current time and the schedule and applying related knowledge; And a sixth step of inferring the service characteristics necessary for the called party and the service characteristics to be avoided by using knowledge of suitable service characteristics and inappropriate service characteristics for each user state stored in the knowledge base. Way. 8. The method of claim 7, wherein the second step comprises: a fourth step of retrieving related call instructions according to the caller and name, and determining whether additional information is needed after inferring the caller status; If additional information is needed in the fourth step, after receiving a response message by transmitting a message for requesting the additional information, using knowledge about suitable service characteristics and inappropriate service characteristics for each user state stored in the knowledge base, A fifth step of inferring service characteristics required by the caller and service characteristics to be avoided; And in case the additional information is not needed in the fourth step, inferring the service characteristics necessary for the caller and the service characteristics to be avoided by using knowledge about the appropriate service characteristics and the inappropriate service characteristics for each user state stored in the knowledge base. A knowledge-based call connection control processing method comprising a sixth step. 10. The method of claim 9, wherein the caller state inference of the fourth step comprises inferring using the calling party's position, importance of the call, and instructions. . 8. The method of claim 7, wherein the third step comprises: a fourth step of determining whether a suitable call processing method exists by extracting a call processing method having mutually suitable service characteristics by comparing a called party service characteristic and a calling party service characteristic; A fifth step of performing a corresponding call processing method if only one call processing method exists in the fourth step; A sixth step of comparing the called party's preference information with the caller's preference information when the call processing method exists in the fourth step to select and perform the call processing method having the highest preference; And a seventh step of resolving the contradiction and finally selecting and performing one call processing method if there is no appropriate call processing method in the fourth step. 12. The method of claim 11, wherein the contradiction resolution process of the seventh step uses a priority value predefined in rules in the knowledge base, and uses a meta information that is reinforced by a learning function. A knowledge based call access control processing method characterized by giving a user a choice by applying a rule. 9. The method of claim 7 or 8, wherein the message includes the calling party name, the title, the relationship with the called party, the called party name, the importance of the call, and the calling method preference priority of the calling party. Based Call Access Control Processing Method.