Categorical Data Clustering using Cosine based similarity for Enhancing the Accuracy of Squeezer Algorithm
Sree Ranjani Rajendran, S. Anitha Elavarasi, J. Akilandeswari · 2012
Clustering categorical data is the major challenge in data mining. Direct comparison of categorical data is not possible as in numerical data, understanding the province of categorical data help to form good quality clusters. This paper presents an algorithm Enhanced Squeezer, which incorporates Data Intensive Similarity Measure for Categorical data (DISC) in Squeezer algorithm. DISC measure, cluster data by understanding domain of the dataset, thus clusters formed are not purely based on frequency distribution as many similarity measures do. Clusters formed using Enhanced Squeezer algorithms are intensive and accurate.