Data Mining: Concepts and Techniques
Data Mining: Concepts and Techniques
Pattern Recognition and Machine Learning (Information Science and Statistics)
Pattern Recognition and Machine Learning (Information Science and Statistics)
Principles and Theory for Data Mining and Machine Learning
Principles and Theory for Data Mining and Machine Learning
Fuzzy set-based microarray data analysis techniques for interesting block identification
FUZZ-IEEE'09 Proceedings of the 18th international conference on Fuzzy Systems
| Hi-index | 0.00 |
It is sometimes useful to get easily-interpretable linguistic patterns when classes are needed to be discriminated in a numeric data set. This paper is concerned with a new method to construct decision trees which encode linguistic discriminating rules for zero-centered numeric data. A zero-centered data set could be prepared by normalizing a data set of which all attributes are measured by the same kind of techniques or sensors. In order to get linguistic patterns, the proposed method quantizes the continuous numeric domains into three intervals, labeled with low, neutral, and high, respectively. It allows the negation to the interval labels in the construction of decision trees so that more simpler rules could be obtained. The method was applied to a classification problem of microarray data to show its applicability. The experiment showed that the proposed method could produce linguistic classification rules comparable to the decision tree induction algorithm C4.5.