한국정보처리학회:학술대회논문집 (Proceedings of the Korea Information Processing Society Conference)
- 한국정보처리학회 2004년도 추계학술발표논문집(상)
- /
- Pages.459-462
- /
- 2004
- /
- 2005-0011(pISSN)
- /
- 2671-7298(eISSN)
유전자 알고리즘 기반 적응 군집화 알고리즘
An Adaptive Clustering Algorithm Based on Genetic Algorithm
- Park Namhyun (Dept. of Information Communication, Gwangju Institute of Science and Technology) ;
- Ahn Chang Wook (Dept. of Information Communication, Gwangju Institute of Science and Technology) ;
- Ramakrishna R.S. (Dept. of Information Communication, Gwangju Institute of Science and Technology)
- 발행 : 2004.11.01
초록
This paper proposes a genetically inspired adaptive clustering algorithm. The algorithm automatically discovers the actual number of clusters and efficiently performs clustering without unduly compromising cluster purity. Chromosome encoding that ensures the correct number of clusters and cluster purity is discussed. The required fitness function is desisted on the basis of modified similarity criteria and genetic operators. These are incorporated into the proposed adaptive clustering algorithm. Experimental results show the efficiency of the clustering algorithm on synthetic data sets and real world data sets.
키워드