Similarity measure design on big data

Abstract

Clustering algorithm in big data was designed, and its idea was based on defining similarity measure. Traditional similarity measure on overlapped data was illustrated, and application to non-overlapped data was carried out. Similarity measure on high dimension data was obtained through getting information from neighbor data. Its usefulness was proved, and verified by calculation of similarity for artificial data example. © 2013 Springer Science+Business Media Dordrecht.