ML & KBANN Algorithms

Hybrid Machine Learning and Knowledge-Based Neural Network Tools in C++

Short Description

Development of Machine Learning and KBANN Algorithms.

Contact

Professor Ian Cloete
Dept. of Computer Science
University of Stellenbosch
Private Bag X1
Matieland 7602
SOUTH AFRICA

phone: +27-21-8084210
fax: +27-21-8084416
email: ian@cs.sun.ac.za

World Wide Web

NN

Description

Hybrid Machine Learning and Knowledge-Based Neural Network Tools in C++

Machine Learning methods aim to construct a domain theory from data, while Knowledge-Based Neural Networks incorporate domain knowledge into a connectionist network. We are developing a toolbox of methods for the integration of machine learning methods, knowledge-based systems and artificial neural networks.

The main topics of this research project are:

• Methods for incorporating prior knowledge into feed-forward and recurrent connectionist architectures
• Exploration of relationships between symbolic Machine Learning algorithms, Knowledge Acquisition, Knowledge-Based Systems and Knowledge-Based Neural Networks
• Rule extraction and refinement
• Knowledge and topology revision
• Neural expert systems
• Implementation issues: Modularity, Parallel and Distributed implementations, etc.
• Applications