Analysis method of big data based on K-means algorithm. Issue 1 (May 2021)

• Record Type: Journal Article
• Authors: Zhang, Yichi; Shu, Tao; Zhang, Pengxiang; Xu, Yan
• Is Part Of: Journal of physics. Volume 1939:Issue 1(2021)
• Publication Date: 2021-05
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/1939/1/012083
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: Efficient and accurate clustering of massive data is the basic requirement of deep excavation for internal knowledge of data in the information era. Firstly, in the distributed parallel processing framework of Hadoop, the Map Reduce model is applied to operate the K-means clustering algorithm, then the genetic algorithm is utilized to transform the clustering problem of K-means algorithm into the solution of extremum to solve the randomness confirmed in cluster center. After testing, this method has achieved the ideal effect in terms of clustering efficiency and clustering precision, and plays certain role in deeply excavating the internal knowledge rule of data.