원문정보
초록
영어
There is ineffective classification problem in application of K-means clustering algorithm in massive data cluster analysis. This paper presents a K-means algorithm based on generalization threshold rough set optimization weight. Firstly, utilize attribute order described method, using the average distance calculation with Laplace method to optimize the generalization threshold of fuzzy rough set , then the Euclidean distance metric is used in the calculation of the similarity of K-means algorithm, introducing the variation coefficient into the cluster analysis, clustering the Euclidean distance weighted K-means algorithm totally based on data, finally, combine the rough set algorithm based on the generalization threshold optimization and K-means clustering algorithm, applied to medical and health data classification. The K-means algorithm based on generalization threshold rough set optimization weight presented by this paper has a better effect on medical and health data classification.
목차
1. Introduction
2. Advantages and Disadvantages of K-Means Clustering Algorithm
3. A K-Means Algorithm Based on Generalization Threshold Optimization Rough Set
3.1 A Rough Set Based on Generalization Threshold Optimization
3.2 K-Means Algorithm Based on Weighted Euclidean Distance
3.3 Improved K-Means Algorithm Based on Rough Set Optimization
4. Algorithm Performance Simulation
5. Conclusion
Acknowledgment
References