Customer emotion detection by emotion expression analysis on adverbs

The growing interests of affective computing (AC) and its implementation in e-business demand researchers to explore the applications of AC techniques in detecting customer emotions. Recent studies have shown that language is powerful in conveying emotions, and a good understanding of major language features bears great implications for customer relationship management. However, the effect of adverbs in text emotion prediction has only been briefly mentioned in a few related works, but never thoroughly studied. This paper addresses the gap by investigating how to detect customer emotions through analyzing the adverbs in emotion expressions. In particular, we develop a Japanese adverb emotion corpus, analyze emotion usage of adverbs, and further derive an adverb-emotion lexicon and its rule base. Utilizing these resources, we design and perform experiments to classify emotions in sentences so as to evaluate the effectiveness of different adverb features including adverbs, adverb emotions, and adverb emotion rules. Our experiments and analysis show the great impact of adverbs on emotion expressions, which can be applied to assist e-businesses in improving customer related processes.