Data mining: practical machine learning tools and techniques with Java implementations
Data mining: practical machine learning tools and techniques with Java implementations
Knowledge Discovery and Measures of Interest
Knowledge Discovery and Measures of Interest
Generating Accurate Rule Sets Without Global Optimization
ICML '98 Proceedings of the Fifteenth International Conference on Machine Learning
An Analysis of Quantitative Measures Associated with Rules
PAKDD '99 Proceedings of the Third Pacific-Asia Conference on Methodologies for Knowledge Discovery and Data Mining
Selecting the right interestingness measure for association patterns
Proceedings of the eighth ACM SIGKDD international conference on Knowledge discovery and data mining
Evaluation of rule interestingness measures with a clinical dataset on hepatitis
PKDD '04 Proceedings of the 8th European Conference on Principles and Practice of Knowledge Discovery in Databases
Using Information-Theoretic Measures to Assess Association Rule Interestingness
ICDM '05 Proceedings of the Fifth IEEE International Conference on Data Mining
A Rule Evaluation Support Method with Learning Models Based on Objective Rule Evaluation Indexes
ICDM '05 Proceedings of the Fifth IEEE International Conference on Data Mining
Evaluation of rule interestingness measures in medical knowledge discovery in databases
Artificial Intelligence in Medicine
A data analysis approach for evaluating the behavior of interestingness measures
DS'05 Proceedings of the 8th international conference on Discovery Science
Hi-index | 0.00 |
In order to support rule evaluation procedure of human experts, objective rule evaluation indices, such as accuracy, coverage, support and other interestingness measures have been developed. However, the relationship between their values and real human evaluations has not been clarified. In this paper, we compared the availability of sorting of composed objective rule evaluation indices to that of each index. To compose objective rule evaluation indices, we used Principle Component Analysis to a dataset of their values for rule sets from 32 UCI common datasets. By using a rule set with the real human evaluation for the meningitis dataset, we performed a comparison of a sorting availability to determine the human evaluations between the composed objective rule evaluation indices and each single index. The result shows that the composed indices perform equally by comparing to the best single indices based on the sorting avalibality.