ATTN: Rahul Bhagat
360 Huntington Avenue
Boston, MA 02115-5000
- Natural language processing
- PhD in Computer Science, University of Southern California
- MS in Computer Science, University of Southern California
- BE in Computer Science, Visvesvaraya National Institute of Technology, India
Rahul Bhagat is a senior manager for applied science at Amazon Inc. He has over 15 years of experience leading science, engineering, and product development in industry and academia. His primary interests lie in developing products and algorithms that solve real-world problems through advances in natural language processing, personalization, and information retrieval. Bhagat pioneered a new area in personalized recommendations called “Repeat Purchase Recommendations” by inventing the “Buy it Again” recommendations on Amazon. These recommendations significantly simplified the shopping experience for Amazon customers by making it easy for them to shop for everyday essentials, like groceries and diapers, that they tend to buy over and over again.
His interests include query understanding, document understanding, personalized recommendations, large-scale text mining, information extraction, and applied semantics. Bhagat has been at Amazon since 2009 and currently leads a team of managers, scientists, and engineers focused on improving Amazon Search using advanced natural language processing. During his time at Amazon, Bhagat and his teams have impacted and improved all the major search and discovery features at Amazon including Amazon search, Amazon home page, Amazon product details page, and Amazon checkout experiences, as well as generated over a billion dollars in incremental sales through controlled experimentation.
While pursuing a PhD at the University of Southern California, Bhagat worked under Dr. Eduard Hovy with a thesis on learning paraphrases from natural language text. He spent summer 2007 and summer 2008 in the Google Research group. Before that, he spent two years as a senior programmer analyst at the Information Sciences Institute working with Dr. Eduard Hovy.
Bhagat has published more than 10 papers in top-tier peer-reviewed conferences and journals. He has eight issued patents and more than 10 patents pending. He has also served as a program committee member and reviewer for top-tier conferences including Association for Computational Linguistics, North American Association for Computational Linguistics: Human Language Technology, International Joint Conference on Artificial Intelligence, Empirical Methods in Natural Language Processing, and European Association for Computational Linguistics.
What are your research interests?
My primary research is on using natural language processing to improve product search. My research path has certainly evolved. While I spent a large part of my research and industry career working on standard applications of natural language processing — like information extraction and dialogue — the applications of natural language processing to product search has been a new and rewarding experience. My research evolved in this direction because I realized there is a huge opportunity to improve product search through advances in natural language processing.
What courses/subjects do you teach?
I teach a graduate course on Natural Language Processing (CS 6120).