Digital Library

cab1

 
Title:      IMPLEMENTING SEMANTIC WEB AGENTS THAT LEARN UPON EXPERIENCE: A MACHINE LEARNING MODULE FOR THE CWM RULE SYSTEM
Author(s):      Ossi Nykänen
ISBN:      972-99353-6-X
Editors:      Nuno Guimarães and Pedro Isaías
Year:      2005
Edition:      1
Keywords:      Semantic Web, Machine Learning, Decision Trees, Rule Systems.
Type:      Full Paper
First Page:      409
Last Page:      417
Language:      English
Cover:      cover          
Full Contents:      click to dowload Download
Paper Abstract:      This article describes a study and demonstrates an implementation of decision tree learning algorithms in the context of the CWM rule system. Potential applications include implementing Semantic Web (SW) agents that are able to learn concepts and classifications from the Web. Effectively, this enables systems that learn upon experience. Decision Tree (DT) learners are well-known and robust empirical Machine Learning (ML) algorithms that yield classification rule systems over instances described in terms of attributes. The CWM rule system provides a practical technical framework of implementing SW rule systems. The DT module introduced in this article defines a set of functions, integrated with the CWM rule system, for training, serializing, and publishing decision trees, and using them for making predictions. The potential of decision tree learning algorithms in applications is demonstrated by an empirical study of predicting math students' grades, based on factual data. In the abstract sense, this article provides a concrete strategy for integrating various kinds of machine learning algorithms with the rule systems of the Semantic Web.
   

Social Media Links

Search

Login