Machine Learning for Mining Big Data: A Review

Abstract

Development of Big Data is virtually transforming our lifestyle. It is also ac- celerating industrial growth through process optimization, insight discovery and improved decision making. The massive scale of big data exceeds the processing and analytic capacity of conventional database systems within an acceptable time frame. Researchers rely on the ability to extract values from such massive data through new data analytics principle; machine learning is at its core because of its ability to learn from data and provide data driven insights, decisions, and predictions. In this study, a review investigation is undertaken for exploring application of machine learning techniques in Big Data analytics of various sectors. We have reviewed 45 papers in the area of machine learning and Big Data analytics involving various sectors such as transportation, healthcare, energy, education, supply chain man- agement, etc. Characteristics and difficulties of Big Data management are reviewed with focus on relevant solutions in order to develop overview for future researchers. We have explored the benefits of machine learning and different machine learning models. We have discussed Big Data systems and programming that is used to process and mine Big Data, and we have also given an overview of how Big Data can be manipulated to generate knowledge. Finally, the conclusion along with future recommendations is provided as a research direction underscoring the importance of inventing more machine learning algorithms, tools and techniques to be prepared for processing and mining Big Data. This review study will provide a head start for future researchers in scanning the appropriate machine learning tools and techniques for mining Big Data.