Analyzing Formation Of K Mean Clusters Using Similarity And Dissimilarity Measures
Kompal Ahuja, Ajmer Saini · 2013
Abstract: Measurement of similarity and dissimilarity between two data objects is a challenging problem for data mining. Mainly the clustering algorithms use distance function between the data points to define dissimilarity measure to form clusters but normalized scaler product can also be used as similarity function to form clusters among data points. This paper presents the results of an experimental study of some common k mean clustering techniques. In particular, we compare the two main approaches to k mean clustering. ( we used Euclidean distance as dissimilarity measure and cosine similarity.)