Abstract
Machine learning approaches are widely used in many branches of science and design, including discourse acknowledgment, image categorization, and language processing. Basically, handling huge amounts of data is restricted by a few limitations of traditional data management approaches. Additionally, in order to analyse big data continually with high accuracy and efficiency, new and complex algorithms in light of machine and deep learning approaches are required. In this commitment, we examine Apache Spark MLlib 2.0's growing set of work from a computational standpoint. This library is distributed, versatile, open source, and independent of stage. To explicitly investigate the qualitative and quantitative properties of the stage, we run a few verifiable machine learning experiments. We also discuss upcoming initiatives and highlight recent advances in big data machine learning research.