I am a masters student in the Data Science Track at Institute of Computational and Mathematical Engineering (ICME). I completed my undergraduate degree in Computer Science and Engineering at VIT University, Tamil Nadu, India. I enjoy problem solving and analytical thinking. My hobbies include reading, photography and travelling.
Education & Certifications
B.Tech, VIT University, Vellore, Tamil Nadu, India, Computer Science and Engineering (2015)
Margot Gerritsen, Master's Program Advisor
Current Research and Scholarly Interests
Data Science, Information Retrieval, Machine Learning and Data Mining
Temporal Summarization- Sequential update summarization, L3S Research Center, Leibniz University
TREC’s Temporal Summarization Track provides an opportunity to evaluate systems seeking to generate temporal summaries which provide users with timely updates during an event like protests or hostage crises. I adopted the following approach to select update sentences for each hour: First I retrieved the top-m documents for the query to account for prevalence, then I selected the top-q summary-worthy sentences using a learning to rank model, LambdaMart. For training LambdaMart, I used features like variations of unigram frequencies, presence of numerical entities, sumbasic, mutual information and query overlap. Next, I removed redundancy using Minhash based clustering and finally filtered by information gain to select novel updates. I tuned the parameters for best performance using a few topics and evaluations from the 2014 TREC TS task coupled with judgments of my own. I successfully generated three runs which were validated and submitted by L3S for the 2015 task (team id: l3sattrec15).
Association rule mining for quantitative medical data, VIT University
Vellore, Tamil Nadu, India
Member Technical, DE Shaw India Private Limited (July 2015 - June 2016)
Research Intern, L3S Research Center, Leibniz University (January 2015 - June 2015)