Are Emotions Detected Differently within First-Person Narratives ?

Emotion classification within text has been attempted many times through lexical and implicit features. It remains to be seen if the extracted features from personal narratives can provide a base model for emotion classification. This paper explores the ability of traditional lexical and syntactic features used in previous emotion classification work toward personal narratives. The results are compared to a model developed from NonNarratives and the results of Balahur et al. (2002), finding that Non-Narrative models produce the best results when tested within the Non-Narrative domain, but Narrative models produce the best generalized results.