Srinivas Vadrevu

Srinivas Vadrevu (Sree)
Principal Applied Scientist Lead
Microsoft Bing
Sunnyvale, CA
sree47--at--yahoo-com

 

Dr. Srinivas Vadrevu is a Principal Applied Scientist Lead at Bing Microsoft. He is part of the Bing Core Ranking team, working on web search ranking, document understanding and bing prediction engine. His work focuses on natural language understanding of Web data, information extraction and working with large-scale machine learning systems. Previously, he built question answering systems that powered factual question answers in Bing. Prior to Microsoft, Srinivas was Senior Research Scientist at Yahoo! Labs. During his time at Yahoo!, he worked on ranking related problems such as Web search ranking, cross-domain ranking for international countries and ranking and recommendation of related entities and videos. During this time, he authored several publications at venues such as WWW, CIKM and KDD conferences and Machine Learning Journal. His recent work on object ranking powers entity search experiences at various parts within Yahoo!. Previously he worked on information extraction and particularly on extracting structured data from Web pages, where he published a number of papers. He received his Ph.D. in Computer Science from Arizona State University and received his M.S. in Computer Science from University of Minnesota, Duluth. He completed his B.Tech. in Computer Science from Velagapudi Ramakrishna Siddhartha Engineering College (VRSEC), Vijayawada, Andhra Pradesh, India.


Research

  • Information Extraction and Information Retrieval from Web: extraction of semi-structured data from Web pages, Web-scale information retrieval, learning to rank automatic transformation of HTML pages into semi-structured XML documents, separation and extraction of metadata from Web pages.
  • Web Content Mining: automated ontology mining from Web pages, merging/mapping ontologies, metadata and instance mining from domain specific Web pages.
  • Machine Learning and Data Mining: efficient training of artificial neural networks and statistical learning methods for handling large data sets.

Publications

Journals, Magazines and Book Chapters

Refereed Conferences

  • Srinivas Vadrevu, Emre Velipasaoglu. Identifying Primary Content from Web Pages and its Application to Web Search Ranking. In The 20th International World Wide Web Conference (WWW 2011), Hyderabad, India, 2011.
  • Changsung Kang, Srinivas Vadrevu, Ruiqiang Zhang, Roelof van Zwol, Lluis Garcia Pueyo, Nicolas Torzec, Jianzhang He, Yi Chang. Ranking Related Entities for Web Search Queries. In The 20th International World Wide Web Conference (WWW 2011), Hyderabad, India, 2011.
  • Srinivas Vadrevu, Choon Hui Teo, Suju Rajan, Kunal Punera, Byron Dom, Alex Smola, Yi Chang, Zhaohui Zheng. Scalable Clustering of News Search Results. In The Fourth International Conference on Web Search and Data Mining (WSDM 2011), Hong Kong, February, 2011.
  • Bo Long, Yi Chang, Srinivas Vadrevu, Shuang-Hong Yang, Zhaohui Zheng. Ranking with auxiliary data. In The 19th ACM Conference on Information and Knowledge Management (CIKM 2010), pp 1489-1492, Toronto, Canada, October, 2010.
  • Olivier Chapelle, Pannagadatta Shivaswamy, Srinivas Vadrevu, Kilian Q. Weinberger, Ya Zhang, Belle Tseng. Multi-Task Learning for Boosting with Application to Web Search Ranking. In The Proceedings of the 16th international conference on Knowledge discovery and data mining (SIGKDD 2010), pp 1189-1198, Washington DC, USA, July 2010.
  • Bo Long, Sudarshan Lamkhede, Srinivas Vadrevu, Ya Zhang, Belle Tseng. A Risk Minimiza-
    tion Framework for Cross-Domain Learning. In The 18th ACM Conference on Information and Knowledge Management (CIKM 2009), Hong Kong, China, November, 2009.
  • Srinivas Vadrevu, Anne Ya Zhang, Belle Tseng, Gordon Sun, Xin Li. Identifying Regional
    Sensitive Queries in Web Search. In The 17th International World Wide Web Conference (WWW 2008), Beijing, China, April 2008.
  • Fatih Gelgi, Srinivas Vadrevu, Hasan Davulcu. Fixing Weakly Annotated Web Data Using
    Relational Models. In The Seventh International Conference on Web Engineering (ICWE 2007),
    Como, Italy, July 2007.
  • Fatih Gelgi, Srinivas Vadrevu, Hasan Davulcu. Relational Model Based Annotation of Web
    Data. In The Fifth Atlantic Web Intelligence Conference (AWIC 2007), Fontainebleau, France,
    June 2007.
  • Fatih Gelgi, Srinivas Vadrevu, Hasan Davulcu. Scuba Diver: Subspace Clustering of Web Search Results. In The 3rd International Conference on Web Information Systems and Technologies (WEBIST 2007), Barcelona, Spain, March 2007.
  • Syed Toufeeq Ahmed, Srinivas Vadrevu, Hasan Davulcu. DataRover: An Automated System for Extracting Product Information from Online Catalogs. In The 4th Atlantic Web Intelligence Conference 6th International Conference on Web Information Systems (AWIC 2006), Beer Sheva, Israel, June 2006.
  • Srinivas Vadrevu, Fatih Gelgi, Saravanakumar Nagarajan, Hasan Davulcu. METEOR: Metadata and Instance Extraction from Object Referral Lists on the Web. In The First Online Metadata and Semantics Research Conferences (MTSR 2005), Approaches to Advanced Information Systems, 2005.
  • Srinivas Vadrevu, Fatih Gelgi, Hasan Davulcu. Semantic Partitioning Web Pages. In The 6th International Conference on Web Information Systems Engineering (WISE 2005), New York City, NY, November 2005. (Acceptance Rate: 12%)
  • Fatih Gelgi, Srinivas Vadrevu, Hasan Davulcu. Improving Web Data Annotations with Spreading Activation. In The 6th International Conference on Web Information Systems Engineering (WISE 2005), New York City, NY, November 2005. (Acceptance Rate: 12%)
  • Srinivas Vadrevu, Saravanakumar Nagarajan, Fatih Gelgi, Hasan Davulcu. Automated Metadata and Instance Extraction from News Web Sites. In The 2005 IEEE/WIC/ACM International Conference on Web Intelligence (WI 2005), France, September 2005. (Acceptance Rate: 18%)
  • Hasan Davulcu, Srinivas Vadrevu, Saravanakumar Nagarajan, Fatih Gelgi. METEOR: Metadata and Instance Extraction from Object Referral Lists on the Web. In The 14th International World Wide Web Conference (WWW 2005), Chiba, Japan, 2005. (Poster Presentation)
  • Hasan Davulcu, Srinivas Vadrevu, Saravanakumar Nagarajan. OntoMiner: Bootstrapping Ontologies from Overlapping Domain Specific Web Sites. In The Thirteenth International World Wide Web Conference (WWW 2004), New York, 2004. (Poster Presentation)
  • Hasan Davulcu, Srinivas Vadrevu, Saravanakumar Nagarajan. OntoMiner: Bootstrapping and Populating Ontologies from Domain Specific Web Sites In First International Workshop on Semantic Web and Databases, September 2003, Berlin, Germany.
  • Srinivas Vadrevu. Efficient Neural Network Training Using Subsets of Very Large Datasets. Masters' Thesis, University of Minnesota Duluth, Computer Science Department, August 2002.
  • Srinivas Vadrevu and Masha Sosonkina. Network Visualization Tool: User Manual. Technical Report, UMD TR 02-01, University of Minnesota Duluth, Computer Science Department, February 2002. (NVT is currently used as a demonstration tool in the Computer Networks course in Computer Science Department at University of Minnesota Duluth. The distribution of NVT tool can be foundas NVT.tar.gz)

Professional Work Experience

  • Intern in Windows Live Search at Microsoft Corp. (Summer 2006): Developed tools and methodology to analyze the quality of the index of Windows Live Search based on the static rank distribution of the pages. This involved collection and analysis of the appropriate metrics for documents in the entire index data and and identify possible solutions to make the index generation pipeline more efficient.
  • Intern in MSN Search at Microsoft Corp. (Summer 2005): Developed a tool that measures of the effect of query rewriting on the relevance of the results in a search engine. The tool automatically suggests the rewriting options with synonyms and co-occurring terms with the query terms that improve the relevance of the results.
  • Research Scientist in Data Mining Research Group at Humana Inc. (Summer 2004): Developed a system that visualizes and analyzes the disease conditions of health insurance members based on their medical and pharmacy claims. Devised Bayesian Vector Autoregression (BVAR) model to forecast claim costs and health conditions. This system helps in identifying the potential sick customers that can be helped by a dedicated personal nurse program.

Other Interests

I founded a Web site called 'Vaidika Vignanam', http://www.vignanam.org which aims to gather the vast spiritual and devotional literature of India in a single place in all of the Indian languages.