During my B.Tech years, I completed two research internships focused on data-driven analysis of real-world systems: one in social network analytics and one in public policy data. These experiences strengthened my foundations in applied data science, research methodology, and translating analysis into insight.

1. Estimating Retweet Likelihood Using Topological Metrics
Summer Research Internship — IIT Patna, 2019

Developed a binary classification model to predict retweet likelihood using only graph-based features, without textual content. Working with the Higgs Twitter dataset from the Stanford SNAP repository, I constructed social and retweet graphs and engineered topological features such as degree centrality, clustering coefficient, and PageRank.

The project demonstrated that network structure alone can effectively predict information propagation, achieving strong classification accuracy and offering insights into how topology influences social behavior.

TOOLKIT: Network science, graph theory, feature engineering, statistical modeling, research documentation

2. Data Analysis of Government Social Welfare Schemes
Internship Project — Digital Govt. Research Centre, Patna, 2018

Conducted an exploratory data analysis of Bihar Government welfare schemes using the e-Labharthi portal dataset for Arwal district. The analysis examined demographic patterns, rural–urban disparities, documentation gaps, and banking preferences across three social security programs.

The study uncovered critical implementation insights—such as high rural participation, gender distribution trends, and systemic gaps in digital documentation—and translated complex administrative data into actionable policy recommendations for improving outreach, infrastructure, and service delivery.

TOOLKIT: Exploratory data analysis, demographic analysis, public data insights, data-driven recommendations

-> Explore more details on these two projects in the attached document.