• DEVI B
• VANII G

Data mining techniques are used mainly of the clustering and document categorization, for calculating data points into identity related groups such that the points that belong to the similar group are additional similar than the points belonging to different clusters. Detection of anomalies or outliers points is a very significant issue in a high dimensional unsupervised datasets. Clustering has been employed by numerous researchers to solve such problems and choosing amount of K clusters without information loss. These issues of the existing techniques are making them inappropriate for selecting an amount of clusters in real world dataset. To overcome this problem without requiring choosing K value based clustering, this paper presents an improved Fuzzy based reverse neighbour clustering with KNN classification outlier mapping (FRNC-KNNOM) algorithm that can be easily allocated number of clusters in dynamic manner in MATLAB R2013a simulation. The proposed method follows a modified Fuzzy clustering and KNN Classification algorithm with reverse neighbour outlier mapping to discover the distance between data points. Finally, the proposed FRNC-KNNOM algorithm performs the Reverse Neighbour search and clusters process in dynamic manner.

Data Mining, RNN, Fuzzy, KNN, Classification, Outlier

"FRNC-KNNOM: AN INMPROVED FUZZY BASED REVERSE NEIGHBOUR CLUSTERING WITH KNN CLASSIFICATION OUTLIER MAPPING ALGORITHM", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 5, page no.304-309, May-2019, Available :http://www.jetir.org/papers/JETIR1905J45.pdf

"FRNC-KNNOM: AN INMPROVED FUZZY BASED REVERSE NEIGHBOUR CLUSTERING WITH KNN CLASSIFICATION OUTLIER MAPPING ALGORITHM", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 5, page no. pp304-309, May-2019, Available at : http://www.jetir.org/papers/JETIR1905J45.pdf