Olena Medelyan

Natural Language Processing Consulting

I worked with the blog search, news search and sentiment analysis teams. I implemented a method for multilingual concept-based indexing of blog posts and news articles, including a server application and a web demo. Technologies used: MapReduce, Google Apps. Supervisor: Kerry Hannan, New York office

I worked on mining semantic relations from Wikipedia at the NLP group with Michael Strube.

The company specializes in error-tolerant search in structured data. I was implementing experimental search features that use linguistic knowledge. I also wrote a server-sided query refinement web application and designed an algorithm for extracting synonyms and collocation from query logs (published at ESSLLI 2004).

I was working as a student research assistant (HiWi) in the Computational Linguistics group and in the department of Medical Informatics. This was an on and off, part time position during my studies. I implemented a language and topic specific web crawler, published at LREC 2006, and I also was involved in development of a cross-language information retrieval system for the medical domain.

Natural Language Processing, Information Retrieval, Text Mining

Computational Linguistics, Computer Science