Form and function in evaluative language

MyBook is a cheap paperback edition of the original book and will be sold at uniform, low price.
This Chapter is currently unavailable for purchase.

In this paper we summarise current research in the analysis of evaluative language,both from a linguistic and a NLP perspective, and offer a description ofthe approaches that have been proposed, to focus on the use of text corpora as acommon resource employed by practitioners of both disciplines. We then presentsome of the work carried out by our research team, Tecnolengua, that hasconcentrated on the construction of a domain-neutral, linguistically-motivatedsentiment analysis tool, which makes extensive use of lexical resources of varioustypes. Among such resources, the inclusion of multiword expressions in ourdatabase is a key factor in the improved performance of Sentitext, as they playan integral role in the creation of the context rules we use to account for thosecases in which the assigned valence of an individual word is modified by its surroundinglinguistic environment.


This is a required field
Please enter a valid email address