Swastika Mohapatra

Built credit risk models using XGBoost to help capture defaulters in the US small business market using a dataset of 18MM records. Evaluated model performance using metrics such as Gini and default capture.

Wrote production level python web-scrapers to scrape data from IFCN certified fact-checking websites in India (text and multimedia content within articles), which was then pushed into a mongo database. Was a part of a 3-person team working on a project titled ‘standardized GitHub metrics’ to consult github on how to release platform engagement data(this data informs global tech maturity, resilience, digital readiness, etc). This project was under Github's social impact team. Analyzed the data being scraped by the various fact-checking scrapers used at Tattle.

Was a part of the team at the Aspirational districts program under NITI Aayog. Wrote a python script to identify the socio-economic indicators wherein a given district is based on real time data collected from the aspirational districts (having 81 features for 112 districts collected from April 2018 onwards). This script would help the chief secretaries of states to identify the areas where they need to pay more attention. Analyzed district level health infrastructure and human resources data of India to estimate inputs required for augmenting health infrastructure in the aspirational districts as per the Indian public health standard norms. Analyzed covid caseload, vaccination coverage, and district-level testing centers data to help provide insights on the situation of the covid-19 pandemic in the aspirational districts to assist in resource distribution planning.

Worked on the import-export food industry data of India to analyze which food products produced in India have a potential demand in the upcoming years in various countries such as Venezuela, Costa Rica, Honduras and other South American countries. Also carried out competitive market analysis for Tender Leaf to understand which areas the company can work on to be at par with its competitors.

Worked on a research-based project under two professors of University of Delhi. The study aimed at examining the association of various socio-demographic factors with prevalence of anxiety and depression among students. Logistics regression analysis was performed. Research Paper - https://journalppw.com/index.php/jppw/article/view/16512