Rahul Bhagat is the Head of Multilingual Search Quality in Amazon Search. Rahul has been at Amazon for over 10 years and manages teams in Seattle and Bangalore focused on improving Amazon Search using advanced Natural  Language Processing. Rahul pioneered a new class of personalized  recommendations called "Repeat Purchase Recommendations" by inventing the "Buy it Again" recommendations on Amazon. Rahul  has published 11 papers in top-tier peer reviewed conferences and  journals, has 8 issued patents, and over 10 patents pending.  Before joining Amazon, Rahul earned a PhD in Computer Science at the University of Southern California. His primary  interests lie in developing products and algorithms that solve real  world problems through Natural Language Processing, Personalization, and  Information Retrieval.
 

Summary
Rahul Bhagat is the Head of Multilingual Search Quality in Amazon Search. Rahul 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 (NLP), Personalization,  and Information Retrieval (IR). Rahul 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 (e.g., groceries, diapers, etc.) 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. Rahul has been at  Amazon since 2009 and currently leads a team of managers, scientists,  and engineers focused on improving Amazon Search using advanced NLP.  During his time at Amazon, Rahul 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, Amazon  checkout experiences, among others and generated billions of dollars in incremental sales through controlled experimentation.
 

Prior to Amazon, Rahul earned a PhD in Computer Science at the University of Southern California  with a thesis on Learning Paraphrases from  (natural language) Text. During his PhD, he spent two summers as an intern in the Google Research group. Before that, he spent was a Sr. Programmer Analyst at the Information Sciences Institute.  Prior to that, he obtained a Master's (MS) degree in Computer Science  from the University of Southern California and Bachelor's (BE) degree  also in Computer Science from the Visvesvaraya National Institute of  Technology in Nagpur, India.
 

Rahul  has published 11 papers in top-tier peer reviewed conferences and  journals, has 8 issued patents, and over 10 patents pending. He has also  served as a Program Committee member and reviewer for top tier  conferences including Association for Computational Linguistics (ACL), North American Association for Computational Linguistics: Human Language Technology (NAACL-HLT), International Joint Conference on Artificial Intelligence (IJCAI), Empirical Methods in Natural Language Processing (EMNLP), and European Association for Computational Linguistics (EACL).
 

Research Interests
Natural Language Processing, Personalization, Text Mining, Information Extraction, Social Media Data Mining, Query Reformulation, Applied Semantics, Question Answering, Information Retrieval, and Machine Learning.
 

Publications
Rahul Bhagat, Srevatsan Muralidharan, Alex Lobzhanidze, and Shankar Vishwanath.
Buy it Again: Modeling Repeat Purchase Recommendations.
In the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD). 2018. [download]
 

Rahul Bhagat and Eduard Hovy.
What Is a Paraphrase?
Computational Linguistics, Volume 39, Issue 3, Page 463-472. 2013. [download
 

Rahul Bhagat.
Learning Paraphrases from Text.

PhD Dissertation, University of Southern California. 2009. [download]
 

Rahul Bhagat, Eduard Hovy, and Siddharth Patwardhan.
Acquiring Paraphrases from Text Corpora.
In the International Conference on Knowledge Capture (KCap). 2009. [download]
 

Rahul Bhagat and Deepak Ravichandran.
Large Scale Acquisition of Paraphrases for Learning Surface Patterns.
In the Association for Computational Linguistics (ACL). 2008. [download]
 

Partha Pratim Talukdar, Joseph Reisinger, Marius Pasca, Deepak Ravichandran, Rahul Bhagat, and Fernando Pereira.
Weakly-Supervised Acquisition of Labeled Class Instances using Graph Random Walks.
In the Empirical Methods in Natural Language Processing (EMNLP). 2008. [download]
 

Rahul Bhagat, Patrick Pantel and Eduard Hovy.
LEDIR: An Unsupervised Algorithm for Learning the Directionality of Inference Rules.
In the Empirical Methods in Natural Language Processing (EMNLP). 2007. [download]
 

Rahul Bhagat and Eduard Hovy.
Phonetic Models for Generating Spelling Variants.
In the International Joint Conference on Artificial Intelligence (IJCAI). 2007. [download]
 

Patrick Pantel, Rahul Bhagat, Bonaventura Coppola, Timothy Chklovski and Eduard Hovy.
ISP: Learning Inferential Selectional Preferences.
In the North American Association for Computational Linguistics/Human Language Technology (NAACL/HLT). 2007. [download]
 

Rahul Bhagat, Anton Leuski and Eduard Hovy.
Shallow Semantic Parsing despite Little Training Data.
In the ACL/SIGPARSE 9th International Workshop on Parsing Technologies, Vancouver, B.C., Canada. 2005.
 

David Traum, William Swartout, Jonathan Gratch, Stacy Marsella, Patrick Kenney, Eduard Hovy, Shri Narayanan, Ed
Fast, Bilyana Martinovski, Rahul Bhagat, Susan Robinson, Andrew Marshall, Dagen Wang, Sudeep Gandhe and Anton
Leuski.
Demo Submission: Virtual Humans for non-team interaction training.
In the ACL/ISCA 6th SIGdial Workshop on Discourse and Dialogue. 2005.

Anton Leuski, Douglas W. Oard, and Rahul Bhagat.
eArchivarius: Accessing collections of electronic mail.
In ACM Special Interest Group on Information Retrieval (SIGIR). 2003.

Patents
Rahul Bhagat.
Foreign Language Translation using Product Information.

CN 104145270. 2018. [link]
US 9684653. 2017. [link]
JP 5948445. 2016. [link]

 

Rahul Bhagat, Michael Brundage, and Daniel Parshall.
Adjusting Search Result Interface based upon Query Language.
US 9626431. 2017. [link]
US 8949107. 2015. [link]

Grant Emery, Rahul Bhagat, Brian Cameros, Bejamin Cohen, Logan Dillard, Yongweng Liang, Scott Mongrain, Michael Quinn, Eli Rosofsky, and Adam Sanders. 

Video Classification

US  9615136. 2017. [link ]

  

Rahul Bhagat.

Generating Suggested Search Queries.
US 9098569. 2015. [link]
 

Rahul Bhagat, Brian Cameros, and Srikanth Thirumalai.
Graph Based Semantic Analysis of Items.
US 8832091. 2014. [link]

Program Committee and Reviewer
* 2019: Association for Computational Linguistics (ACL)
* 2019: Empirical Methods in Natural Language Processing (EMNLP)
* 2019: International Joint Conference on Artificial Intelligence (IJCAI)
* 2018: Association for Computational Linguistics (ACL)
* 2018: North American Association for Computational Linguistics: Human Language Technology (NAACL-HLT)
* 2017: European Association for Computational Linguistics (EACL) workshop on Ethics in Natural Language Processing
* 2016: Empirical Methods in Natural Language Processing (EMNLP)
* 2013: International Joint Conference on Artificial Intelligence (IJCAI)
* 2012: North American Association for Computational Linguistics: Human Language Technology (NAACL-HLT)
* 2010: Corpus-Based Approaches for Paraphrasing and Nominalization (CBA)
* 2009: North American Association for Computational Linguistics: Human Language Technology (NAACL-HLT)
* 2009: European Association for Computational Linguistics (EACL)
* 2009: Empirical Methods in Natural Language Processing (EMNLP)
* 2008: North American Association for Computational Linguistics: Human Language Technology (NAACL-HLT)