Authors:
Nicolas Cebron
;
Fabian Richter
and
Rainer Lienhart
Affiliation:
University of Augsburg, Germany
Keyword(s):
Decision trees, Counterexamples, Machine learning, Data mining, Decision making.
Related
Ontology
Subjects/Areas/Topics:
Applications
;
Classification
;
Economics, Business and Forecasting Applications
;
Inductive Learning
;
Instance-Based Learning
;
Pattern Recognition
;
Theory and Methods
Abstract:
While it is well accepted in human learning to learn from counterexamples or mistakes, classic machine learning algorithms still focus only on correctly labeled training examples.We replace this rigid paradigm by using complementary probabilities to describe the probability that a certain class does not occur. Based on the complementary probabilities, we design a decision tree algorithm that learns from counterexamples. In a classification problem with K classes, K 1 counterexamples correspond to one correctly labeled training example. We demonstrate that even when only a partial amount of counterexamples is available, we can still obtain good performance.