ON FUZZY NEIGHBORHOOD BASED CLUSTERING ALGORITHM WITH LOW COMPLEXITY

The main purpose of this paper is to achieve improvement in thespeed of Fuzzy Joint Points (FJP) algorithm. Since FJP approach is a basisfor fuzzy neighborhood based clustering algorithms such as Noise-Robust FJP(NRFJP) and Fuzzy Neighborhood DBSCAN (FN-DBSCAN), improving FJPalgorithm would an important achievement in terms of these FJP-based meth-ods. Although FJP has many advantages such as robustness, auto detectionof the optimal number of clusters by using cluster validity, independency fromscale, etc., it is a little bit slow. In order to eliminate this disadvantage, by im-proving the FJP algorithm, we propose a novel Modied FJP algorithm, whichtheoretically runs approximately n= log2 n times faster and which is less com-plex than the FJP algorithm. We evaluated the performance of the ModiedFJP algorithm both analytically and experimentally.