Information Retrieval | HCI | Archaeology | Travel | Photography | Running | Fun

During my time at Google, I worked with the Search Quality Team to improve the Google search results and the user experience on the Google front page. I contributed to the development of Sitelinks - the sublinks that often appear below the top Google search result. These links lead to important pages on a site and allow users to quickly find the information they want.

Website Summarization and Titles

I worked on a system that automatically generates concise and relevant website titles for Sitelinks. Original website titles are often long, irrelevant and spammy. Several usability studies have shown that users prefer shorter titles. My project aimed to produce concise titles that summarize the contents of webpages. In addition to the usual information extraction and NLP challenges, I had to develop algorithms that scaled for the web. Publication is forthcoming.


Similar Blogsearch

I worked on a 20% project at Google to build a recommendation engine for blogs. The project aimed to suggest new blogs to users based on the blogs they already read. We experimented with clustering algorithms that grouped blogs based on their content, link structure, and subscription data.


Webmaster Console

The Google Webmaster Console provides users with feedbacks about their sites. I developed a pipeline that analyzes site contents and identifies problems that lead to poor search results. I also worked on a feature to enable webmasters to remove Sitelinks from their search result listings. Check out my blogpost.