Features for automatic discourse analysis of paragraphs

In this paper, we investigate which information is useful for the detection of rhetorical (RST) relations between (Multi-) Sentential Discourse Units ((M -)SDUs)-text spans consisting of one or more sentences-within the same paragraph. In order to do so, we simplified the task of discourse parsing to a decision problem in which we decided whether an (M-)SDU is either rhetorically related to a preceding or a following (M-)SDU. Employing the RST Treebank (Carlson et al. 2003), we offered this choice to machine learning algorithms to­ gether with syntactic, lexical, referential, discourse and surface features. Next, the features were ranked on the basis of (1) models established by the classification algorithms and (2) feature selection metrics. Highly ranked features that predict the presence of a rhetorical relation are syntactic similarity, word overlap, word similarity, continuous punctuation and many reference features. Other features are used to introduce new topics or arguments: time references, proper nouns, definite articles and the word further.