Abstract: | This paper presents a new attribute-based learning algorithm, TS. Different from ID3, AQ11 and HCV in strategies, this algorithm operates in cycles of test and split. It uses those attribute values which occur only in positive examples but not in negative examples to discriminate positive examples against negative examples in a straightforward manner and chooses the attributes with the least number of different values to split example sets. TS is natural, easy to implement, and polynomial in time complexity.
|