Comparison of Hierarchical and Non-Hierarchical Clustering Algorithms

As a result of wide use of technology, large data volumes began to emerge. Analyzes on this big size data and obtain knowledge in it very difficult with simple methods. So data mining methods has risen. One of these methods is clustering. Clustering is an unsupervised data mining technique and groups data according to similarities between records. The goal of cluster analysis is finding subclasses that occur naturally in the data set. There are too many different methods improved for data clustering. Performance of these methods can be changed according to data set or number of records in dataset etc. In this study we evaluate clustering methods using different datasets. Results are compared by considering different parameters such as result similarity, number of steps, processing time etc. At the end of the study methods are also analyzed to show the appropriate data set conditions for each method.