A tool for polarity classification of human affect from panel group texts

Abstract

We introduce an explorative tool for affect analysis from texts. Rather than the full range of emotions, feelings, and sentiment, our system is currently restricted to the positive or negative polarity of phrases and sentences. It analyses the input texts with the aid of a affect lexicon that specifies among others the prior polarity (positive or negative) of words. A chunker is used to determine phrases that are the basis for a compositional treatment of phraselevel polarity assignment. In our current experiments we focus on phrases that are targeted towards persons, be it the writer (I, my, me,.), the social group including the writer (we, our,.) or the reader (you, your,). We evaluate our system with standard data (customer reviews). We also give initial results from a small corpus of 35 texts taken from a panel group called 'I battle depression'.