Paper: Learning a Lexical Simplifier Using Wikipedia

ACL ID P14-2075
Title Learning a Lexical Simplifier Using Wikipedia
Venue Annual Meeting of the Association of Computational Linguistics
Session Main Conference
Year 2014

In this paper we introduce a new lexical simplification approach. We extract over 30K candidate lexical simplifications by identifying aligned words in a sentence- aligned corpus of English Wikipedia with Simple English Wikipedia. To apply these rules, we learn a feature-based ranker us- ing SVM rank trained on a set of labeled simplifications collected using Amazon?s Mechanical Turk. Using human simplifi- cations for evaluation, we achieve a preci- sion of 76% with changes in 86% of the examples.