Introduction    |    Members    |    Research    |    Resources    |    Service    |    Information
  Research Overview
  Research Area
- Natural Language Processing
- Ontology
- Semantic Web
  Major Projects
- COREONTO
+ Project Introduction
+ Project Background
+ Project Detail
+ Project Result
- BORA
+ Project Introduction
+ Project Background
+ Project Detail
+ Project Result
- KORTERM
+ Project Introduction
+ Project Background
+ Project Detail
+ Project Result
 
HOME > Research >     
 
Cognitive Science

The goals and scope of research
It is intended to present the goals and expected results of research in terms of input, output and processing. . The ultimate scenario of research is part of an artificial brain, to be used for the artificial assistant. Figure 1 is a visual representation of the artificial brain.



Figure 1. An artificial brain model

This project is targeted for a function of the artificial assistant that furnishes answers customized for the level of the user and the intended use in response to the request for specialized knowledge. This will be implemented through an "autonomous mental" using the control device of the artificial brain.

From the viewpoint of the autonomous mental, problems may be summarized as three: (i) the problem of the accumulation of knowledge, (ii) the problem of the search of knowledge, and (iii) the problem of the reorganization of knowledge.

Autonomous mental
A high-level cognitive inference model, a goal of this study, is an autonomous mental capable of processing knowledge such as learning and inference through the mimicking of human behavior patterns and interactions with humans. The autonomous mental is the core component of the artificial assistant that is capable of using human-like abilities to assist humans with necessary tasks while interacting with humans. The initial model of the autonomous mental is configured with the control unit having a framework similar to that of an intelligent agent (http://www.fipa.org). An intelligent agent is capable of spontaneity, learning, interactivity, and mobility. In this respect, the control unit of the autonomous mental is defined as follows, based on the general capabilities of an intelligent agent. Here, mobility is replaced with portability for the control unit of the autonomous mental.

  - Spontaneity: It sets its own goals, acts and learns on its own.

  - Learning: It acquires knowledge, learns new facts and use them to draw inferences.

  - Interactivity: It has a command of languages to interact with humans or other autonomous mentals.
    Languages
    employed by the interactive agent include, among other things, the Agent Communication Language (ACL),
    which is a formal language.

  - Portability It is capable of distinguishing knowledge in its own possession, delivering it, and porting part of
    knowledge to other autonomous mentals.

The autonomous mental grows by learning. The growth of the autonomous mental may be considered in two aspects. One is the growth of the control unit that learns knowledge, determines behavior, and engage in interactions, and the other is the growth of knowledge per se. The control unit is an apparent capability and the growth of knowledge is unapparent. The growth of these two aspects in a manner to complement each other allows the mental to grow into one that could act with more knowledge.

For example, the growth of spontaneity is when learning of user intentions based on the recognition of the user results in the mental being capable of spontaneously performing tasks that previously required user directions.

Learning could be achieved by the growth of the internal knowledge system. The growth of the internal knowledge system will increase the scope of knowledge attainable from knowledge sources, which in turn enables the mental to discover facts that have not been learned by making inferences with internal knowledge and to correct any conflicts with the existing knowledge system by learning from the user.

The growth of interactivity will allow the mental to gradually build up more information on the user through the interaction with the user, present the information desired by the user in a more prompt manner, and make suggestions and directions to the user.

The growth of portability will result from the capability of finding and resolving any conflicts with existing knowledge over the course of communicating knowledge with other mentals. The autonomous mental identifies any semantic conflicts and then either transfers its own knowledge or learns other mental knowledge.