Networks form an integral part of our daily lives, comprising structures known as nodes connected via ties. These networks exhibit a complex interplay of random and structured features. We encounter networks regularly in our daily activities, whether it is logging into websites, commuting by train, going to work, or attending school. Networks are prevalent in various academic disciplines and practical applications, spanning biology, physics, engineering, social science, and numerous other fields. The early 21st century has witnessed a rapid expansion of techniques facilitating human interaction, significantly impacting behavior and social bonding in both urban and rural settings. The availability and accessibility of extensive population datasets have spurred ongoing research into understanding and shaping human inclinations and sociological phenomena in unconventional ways. This review delves into the nuanced aspects of social physics, exploring the rationale behind choosing this research topic, various conceptual frameworks, categorizations, and distinctions between machine learning and deep learning. The article briefly outlines the advantages of leveraging big data in social physics, highlighting its transformative potential. Lastly, the authors provide insights into several applications of social physics in subsequent sections, offering a glimpse into its real-world implications.