Earticle

초록

영어

Clustering of class labels can be generated automatically, which is much lower quality than labels specified by human. If the class labels for clustering are provided, the clustering is more effective. In classic document clustering based on vector model, documents appear terms frequency without considering the semantic information of each document. The property of vector model may be incorrectly classified documents into different clusters when documents of same cluster lack the shared terms. To overcome this problem are applied by the knowledge based approaches. However, these approaches have an influence of inherent structure of documents on clustering and a cost problem of constructing ontology. In addition, the methods are limited to cluster suitable text document clustering from in exploding big text data on Cloud environment. In this paper, we propose a big text document clustering method using terms of class label and semantic feature based Hadoop. Class label term can well represent the inherent structure of document clusters by non-negative matrix factorization (NMF) based Hadoop. The proposed method can improve the quality of document clustering which uses the class label terms and the term weights based on term mutual information (TMI) with WordNet at a little cost. It also can cluster the big data size of document using the distributed parallel processing based on Hadoop. The experimental results demonstrate that the proposed method achieves better performance than other document clustering methods.

목차

Abstract
 1. Introduction
 2. Hadoop Framework
 3. Non-negative Matrix Factorization
 4. Proposed Document Clustering Method
  4.1. Preprocessing
  4.2. Extracting Class Label Terms
  4.3. Computing Term Weights by TMI based on WordNet
  4.4. Clustering Document using Similarity
 5. Experiments and Evaluation
 6. Conclusion
 References

키워드

저자

• Yong-Il Kim [ Honam University, South Korea ]
• Yoo-Kang Ji [ DongShin University, South Korea ]
• Sun Park [ GIST, South Korea ] Corresponding author

참고문헌

발행기관

간행물

표지보기 관심저널 등록

• 간행물명

  International Journal of Software Engineering and Its Applications

• 간기

  월간

• pISSN

  1738-9984

• 수록기간

  2008~2016

• 등재여부

  SCOPUS

• 십진분류

  KDC 505 DDC 605