University of Houston
Department of Computer Science

In partial fullfillment of the Requirements for the Degree of
Master of Science


Dipesh Valia
will defend his thesis

A Subsystem for Learning Relation


Abstract

The project, A Learning Program ( ALPS), has been initiated by Dr Kam-Hoi Cheng to develop a program capable of learning quantitative knowledge, and applying them to solve problems. This project develop knowledge and infer new knowledge from existing knowledge.
The Relation learning subsystem is a part of the ALPS learning system. It concentrates on learning both comparative and superlative relationships among objects, such as heavier, and heaviest. This system assumes that the binary relation "larger than" that relates two numbers is already understood by the system. The system is designed to teach relations "heavier", "heaviest", "largest", based on this basic "larger than" relation. An interface is provided to demonstrate that the relations have been taught properly. Based on the learned relations, user can enter information about individual input items and the relations among them. Based on all this input information, the system can answer questions that use any relation known by the system.
The design of the system is done using object oriented paradigms and the solution is expressed in UML. The system has been implemented in C++. The system performs consistency checks while creating the relation instances and the questions provided by a user.


Date: Wednesday, December 1, 1999
Time: 10:00AM
Place: 550-PGH


Faculty, students, and the general public are invited
Thesis Advisor: Dr. Kam-Hoi Cheng