Siddharth Jonathan J.B.

Department of Computer Science,
Stanford University.

Bio

Siddharth Jonathan co-founded Infoaxe, a Search Engine for your Web Memory along with Vijay Krishnan. The Infoaxe(alpha) is now live! Check it out! Read more about Infoaxe here.


Jonathan graduated with a Masters with Distinction in Research in Computer Science from Stanford University. He was awarded the Christofer Stephenson Memorial Award for Graduate Research for Best Masters Thesis at Stanford University for his work on SpotSigs: Near Duplicate Detection in Web Pages with Dr.Andreas Paepcke. Martin Theobald, Andreas and I subsequently published SpotSigs in SIGIR'08. Martin and Greg Linden wrote some very informative blog posts on SpotSigs, here and here.

At Stanford, Jonathan worked with Dr.Andreas Paepcke and Prof.Hector Garcia Molina of the Stanford InfoLab on problems of Context Driven Ranking for Search, Passage retrieval for Web Search, Supervised Web Page Classification and Near Duplicate Detection in Web pages(SpotSigs).

Jonathan's primary focus of research has been applications of Machine Learning for learning & optimizing Ranking functions & algorithms for Web Search. He spent the Summer of 06 with the Yahoo! Search Relevance and Ranking Group where his work on exploiting regional cues from the query and document combined with click stream mining to offer better regionalization of results resulted in an improvement of nearly 3% in the search relevance of the Yahoo! Search Engine. His research on making use of query context in the Ranking algorithm to improve search relevance also showed gains of nearly 2%. Jonathan also worked briefly at Powerset(recently acquired by Microsoft), where he was one of the scientists behind the initial ranking algorithms powering the Natural Language Search Engine. At Powerset, Jonathan worked on developing & improving the ranking function by exploiting various features derived from deep linguistic processing, semantic analysis, traditional keyword based analysis & Web graph information.

Prior to the InfoLab at Stanford, Jonathan worked with Dr.Renate Fruchter of the  Project Based Learning Lab where he helped design and build an Enterprise search engine, InfoMem which could work with hierarchically  structured data and had a ranking function and search result visualization system which was tailored for this purpose. InfoMem was introduced and demoed at the MediaX Conference at Stanford University.

Jonathan's undergraduate degree was from SVCE,(Anna University), Madras in Computer Science and Engineering where he graduated at the top of his graduating class of 2005. He is also one of 15 recipients in India of the Bharat Petroleum Scholarship for Graduate Education. His undergraduate research publications spanned applications of Artificial Intelligence for autonomous navigation in cars, medical diagnosis and applications of Swarm Intelligence for task allocation in Computational Grids.

Jonathan is currently the Corporate Liaison for the Stanford IEEE Chapter. He has also been keenly involved in numerous entrepreneurship events at Stanford. When he's not thinking about Web Search (and sometimes when he is) he occasionally writes on his blog, Searching for A.I.

Current Research Interests

  • Information Retrieval
  • Web Search & Search Personalization
  • Natural Language Processing
  • Applications of Machine Learning and Probabilistic Models to improve Search Relevance/Ranking in Information Retrieval.
  • Text Mining
  • Some of my Courses at Stanford

  • Text Retrieval and Web Search (Prof. Chris Manning, Dr.Prabhakar Raghavan)
  • Natural Language Processing (Prof. Chris Manning)
  • Machine Learning (Prof. Andrew Ng)
  • Probabilistic Models in Artificial Intelligence (Prof. Daphne Koller)
  • Computational Methods in Data Mining (Prof. Sepandar Kamvar)
  • Transaction Processing and Distributed Databases (Prof. Hector Garcia-Molina)
  • Database System Principles (Prof. Hector Garcia-Molina)
  • Algorithms in Biology (Prof. Serafim Batzoglou)