From Data Points to Data Curves: A New Approach on Big Data Curves Clustering

摘要

In the new era of IoT, enormous real-values datasets are produced daily. Time series created by smart devices, financial data, weather analysis, medical applications, traffic control etc. become more and more important in human day life. Analyzing and clustering these time series or in general any kind of curve could be critical. In the current paper, a new methodology (BD2C) is presented, which applies text mining and pattern detection techniques in order to cluster curves according to their shape. Several experiments have been conducted on artificial and real datasets in order to present the accuracy, efficiency and rapid discovery of the best possible clustering that the proposed methodology can achieve.