SAWUS: Siena's Automatic Wikipedia Update System
Abstract
The National Institute of Standards and Technology (NIST) has been running an annual Text Retrieval Competition and Conference (TREC) since 1992. This is a premier conference that offers researchers in the field of Computational Linguistics the opportunity to showcase their work and compare their results against other leading researchers. Our Siena research team participated in the TREC Knowledge Based Acquisition (KBA) Track which was offered for the first time in 2012. The objective of this track is to drive research into automatic acquisition of knowledge such as automatically updating Wikipedia by utilizing online news. Specifically our team of researchers developed a system that filters a stream of content for information that should be included on a given Wikipedia page. It was not yet clear how traditional Information Retrieval (IR) techniques perform for this task therefore we began with a baseline test using current state of the art IR techniques. We then went on to experiment with query expansion building a module that utilized Wikipedia Infoboxes to add terms to our query. This module was incorporated with our IR component to create SAWUS. Four submissions were sent to NIST to undergo a formal evaluation.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 01, 2012
- Accession Number
- ADA581303
Entities
People
- Carl Tompkins
- Sharon G. Small
- Zachary Witter
Organizations
- Siena College