Interactive POS-aware network for aspect-level sentiment classification

Existing aspect-level sentiment-classification models completely rely on the learning from given data-sets. However, these are easily misled by biased samples, resulting in learning some ill-suited rules thatlimit their potential. The information of some specific part-of-speech (POS) categories often indicates theword sentiment polarity, which can be introduced as prior knowledge to facilitate prediction of themodel. Accordingly, we propose an interactive POS-aware network (IPAN) that explicitly introducesthe POS information as reliable guidance to assist the model in accurately predicting sentiment polarity.We distinguish the information of different POS categories using a POS-filter gate and reinforce the fea-tures extracted from adjectives, adverbs, and verbs via a POS-highlighting attention mechanism. Thisenables the model to concentrate on the words that contain significant sentiment orientations and toobtain the most practical learning experience. To emphasize the target information, we construct atarget-context gate that enables the interaction of the target information with contexts; consequently,the model considerably focuses on target-related sentiment features. The experiments onSemEval2014 and Twitter datasets verify that our IPAN consistently outperforms the current state-of-the-art methods