Exploring Unconventional Sources in Big Data: A Data Lifecycle Approach for Social and Economic Analysis with Machine Learning

Abstract

This study delves into the realm of leveraging unconventional sources within the domain of Big Data for conducting insightful social and economic analyses. Employing a Data Lifecycle Approach, the research focuses on harnessing the potential of linear regression, random forest, and XGBoost techniques to extract meaningful insights from unconventional data sources. The study encompasses a structured methodology involving data collection, preprocessing, feature engineering, model selection, and iterative refinement. By applying these techniques to diverse datasets, encompassing sources like social media content, sensor data, and satellite imagery, the study aims to provide a comprehensive understanding of social and economic trends. The results obtained through these methods contribute to an enhanced comprehension of the intricate relationships within societal and economic systems, further highlighting the importance of unconventional data sources in driving valuable insights for decision-makers and researchers alike.