How to implement a big data clustering algorithm
Michele Ianni, Elio Masciari, Giuseppe M. Mazzeo, Carlo Zaniolo · 2019
The current era of Big Data [7] has forced both researchers and industries to rethink the computational solutions for analyzing massive data. In fact, a great deal of attention has been devoted to the design of new algorithms for analyzing information available from Twitter, Google, Facebook, and Wikipedia, just to cite a few of the main big data producers. Although this massive volume of data can be quite useful for people and companies, it makes analytical and retrieval operations really time consuming due to their high computational cost. A possible solution relies upon the possibility to cluster big data in a compact but still informative version of the entire data set. Obviously, such clustering techniques should produce clusters (or summaries) having high accuracy. Clustering algorithms could be beneficial in several application scenarios such as cybersecurity, user profiling and recommendation systems, to cite a few.