An Empirical Analysis Data Mining Frameworks—An Overview

Abstract

In recent years, a massive breakthrough has attracted interest, dramatically altering the way companies produce, conduct, and operate data and analytics structures. Hadoop, NoSQL, and the cloud have ushered in a new age of scale-out, flexible, and real-time computation, resulting in the development of new data processing, collection, and analytical knowledge to enable advanced machine learning and optimization techniques. Over the last decade, many developments in data processing and development have occurred, including big data systems, cloud services, data lakes, optimization, self-service, data collation, machine learning, and many others. Today’s data management systems imitate the old data warehousing and BI systems on the outskirts, but with new data management concepts and updated processes. The need to modernize data processing architecture is widespread, as shown by the amount of architecture consultancy inquiries. As a result, this research explores a simple path to massive data handling framework as well as various popular file system architectures for handling big data for data mining. Furthermore, major frameworks for data mining are investigated and compared.