Computational Statistics Research Talk - Saied Amiri

Title: Revisiting the clustering methods with the application for the BIG DATA

Abstract:

According to the Moore’s and Kryder’s laws, the computational power and information storage increase exponentially, so researchers and workers need more sophisticated and advanced methods for managing and interrogating Big Data.

In this talk, we review important aspects of using machine learning for the Big Data analytics. We discuss the applicability of clustering methods to draw the inference of complex data, non-convex dataset. We also provide a series of arguments to justify the steps in the stages of our clustering methods. We consider ongoing specific techniques likely to achieve the clustering technique for Big data. Finally, we discuss the innovative statistical computing strategies to represent, model, analyze and interpret Big Data using the ensemble approach.