Clustering of Data using Affinity Algorithm
B Rachana · International Journal of Engineering Research and · 2019
The term Big Data is used for denoting the collection of datasets that are extremely large and complex making it difficult to process using traditional data processing applications.The datasets clustering has become a challenging issue in the field of big data.The most widely used procedure to identify clusters is known as k-means.The k-means algorithm finds clusters with the least inertia for a given k.A drawback of this k-means is that if k is not known.This paper presented a new algorithm called affinity propagation which is based on the passing of the message between data points.The number of clusters to be determined or estimated before running the algorithm is not required in this proposed affinity algorithm.