Contextual stance classification of opinions: A step towards enthymeme reconstruction in online reviews

Pavithra Rajendran, Danushka Bollegala, Simon Parsons

Abstract: Enthymemes, that are arguments with missing premises, are common in natural language text. They pose a challenge for the field of argument mining, which aims to extract arguments from such text. If we can detect whether a premise is missing in an argument, then we can either fill the missing premise from similar/related arguments, or discard such enthymemes altogether and focus on complete arguments. In this paper, we draw a connection between explicit vs. implicit opinion classification in reviews, and detecting arguments from enthymemes. For this purpose, we train a binary classifier to detect explicit vs. implicit opinions using a manually labelled dataset. Experimental results show that the proposed method can discriminate explicit opinions from implicit ones, thereby providing encouraging first step towards enthymeme detection in natural language texts.