XMLFind was a search prototype for finding relevant sections within scientific articles in the INEX Document Collection, rather than returning only whole articles.
Context
The INEX Document Collection contained XML versions of more than 12,000 scientific articles published in several IEEE journals between 1995 and 2001. XMLFind explored focused search over the articles’ XML structure.
My role and contribution
Principal investigator. I supervised the student team that developed the prototype during Project Information Retrieval, a one-month intensive course at the University of Amsterdam. The students built it on top of the XML retrieval engine I had developed as part of my PhD research.
How it works
XMLFind provided a search interface to the INEX collection and returned the most relevant sections within the articles for a query.
Outcome
The student team won the best-project prize in 2005. XMLFind later served as the basis for Wikiii, which adapted focused search to Wikipedia.
Technology
PHP, Lucene
Related publications
. xmlfind. Project Information Retrieval Report (P6). 2005. [N/A]
. What do Users Think of an XML Element Retrieval System?. Advances in XML Information Retrieval and Evaluation: Fourth Workshop of the INitiative for the Evaluation of XML Retrieval (INEX 2005). Lecture Notes in Computer Science. 2006. [more]
. Focused Information Access using XML Element Retrieval. SIKS Dissertation Series. 2006. [more]