Kaljahi, Rasoul, et al. "Foreebank: Syntactic analysis of customer support forums." Conference on Empirical Methods in Natural Language Processing (EMNLP). 2015.

This paper introduces a novel treebank of English and French technical forum content annotated for grammatical errors and phrase structure. This dual annotation enables an empirical assessment of how grammatical errors impact parsing performance. The study finds that while parsing corrected versions of the forum sentences is marginally easier, grammatical errors are not the primary factor contributing to the difficulty of parsing this type of text.