[9] A BIBLIOMETRIC METHOD TO REVIEWING MACHINE LEARNING FOR BIG DATA ANALYTICS INVESTIGATING THE DIMENSION DATABASE

Submitted: Oct 14, 2024 , Revised: Oct 30, 2024 , Accepted: Nov 10, 2024 https://doi.org/10.56815/IJMRR.V3I4.2024/103-115

Abstract

Big data has become more popular as a subject of study in recent years, and its rise has been widespread. The purpose of this work is to use text mining-based analytic techniques along with bibliometrics to capture the scientific structure and subject progression of big data research. From the Dimensions database, bibliographic information on big data journal papers published between 2024 and 2014 was gathered and examined. The findings indicate a notable increase in publications since 2014. The main journals, most referenced papers, most prolific authors, nations, and institutions are all highlighted by the study's findings. Second, an original method for locating and examining key research themes in big data papers was put forward. Each cluster of keywords was identified as a topic once it had been clustered. Additionally, to track the thematic progression, the papers were split into four sub-periods. The topic mapping indicates that big data analytics—which includes techniques, tools, supporting infrastructure, and applications—dominates big data research. Security and privacy are two other essential components of big data research. Big data is mostly generated by social networks and the Internet of things, and cloud computing's resources and services make big data processing and administration much easier