A Study on Performing Clusters in Transactional Data with Different Sizes and Shapes
Data mining uses various clustering algorithm for grouping similar objects. One of the most popular algorithm for clustering is density based clustering algorithm, which clusters are of widely differing sizes, densities and shapes when the data contains large amounts of noise and outliers. Many of these issues become even more important when the data is of very high dimensionality, such as text, time series and sequence data. In this paper, the authors present a novel clustering technique, which can solve mentioned issues significantly. They show that their algorithm is intuitive, easy to state and analyze than traditional methods on data set: transactional data. As a result, the algorithm can effectively find the behaviours in transactional data i.e. in banking system.