An analytical study on clustering methods to solve cluster challenges

Finite Uncontrollable data growth in various fields has raised concerns about data storage and analysis. Increasing the volume, variety and velocity of data growth has created problems in analyzing and optimized extracting the knowledge in data using traditional data management tools. Hence, there is need for new methods that can overcome the analytic problems by increasing data. Data clustering is one of the most popular methods for dividing a group of unlabeled data into a subgroup (cluster), so that data instances in a cluster similar to each other and the others belonging to other clusters are vary. Clustering methods suffer from two basic problems: selection of the number of clusters and choice of the starting mid-point for each cluster. Hence, in this paper, we will present an analytical review of data clustering methods to solve these challenges Methods.

Keywords:

Data mining , Association rules mining , Adaptive Support threshold , Attractive items extracting