On the Contribution of Discourse Structure on Text Complexity Assessment

摘要

This paper investigates the influence of discourse features on textcomplexity assessment. To do so, we created two data sets based on the PennDiscourse Treebank and the Simple English Wikipedia corpora and compared theinfluence of coherence, cohesion, surface, lexical and syntactic features toassess text complexity. Results show that with both data sets coherence features are more correlatedto text complexity than the other types of features. In addition, featureselection revealed that with both data sets the top most discriminating featureis a coherence feature.