Learning a Lexical Simplifier Using Wikipedia

Learning a Lexical Simplifier Using Wikipedia - scientific work related to Wikipedia quality published in 2014, written by Colby Horn, Cathryn Manduca and David Kauchak.

Overview

In this paper authors introduce a new lexical simplification approach. Authors extract over 30K candidate lexical simplifications by identifying aligned words in a sentencealigned corpus of English Wikipedia with Simple English Wikipedia. To apply these rules, authors learn a feature-based ranker using SVMrank trained on a set of labeled simplifications collected using Amazon’s Mechanical Turk. Using human simplifications for evaluation, authors achieve a precision of 76% with changes in 86% of the examples.