Apply Rough Set Theory into the Information Extraction The Application of the Clustering

Abstract

Clustering has always been an important subject in data mining, and it has been applied in various domains. Constrained clustering has been an emerging issue over the last few years. Its main idea is applying constraints to the process of clustering to decrease the running time and cost to expectantly promote the quality of clustering. Because clustering is a combinative optimization question, there are some problems such as NP-Hard work and deciding the number of clusters. This paper proposes a constrained clustering technique combining Rough Set theory and Genetic Algorithm into the clustering. We also developed the prototyping system and performed experiments to prove the effectiveness and compare it with other clustering techniques, such as Genetic Algorithm-based clustering and Self-organizing Maps. Finally, the results showed our approach is actually better than other methods.