tailieunhanh - Báo cáo khoa học: "Using Anaphora Resolution to Improve Opinion Target Identification in Movie Reviews"

Current work on automatic opinion mining has ignored opinion targets expressed by anaphorical pronouns, thereby missing a significant number of opinion targets. In this paper we empirically evaluate whether using an off-the-shelf anaphora resolution algorithm can improve the performance of a baseline opinion mining system. We present an analysis based on two different anaphora resolution systems. | Using Anaphora Resolution to Improve Opinion Target Identification in Movie Reviews Niklas Jakob Iryna Gurevych Technische Universitat Darmstadt Technische Universitat Darmstadt HochschulstraBe 10 64289 Darmstadt HochschulstraBe 10 64289 Darmstadt http people Abstract Current work on automatic opinion mining has ignored opinion targets expressed by anaphorical pronouns thereby missing a significant number of opinion targets. In this paper we empirically evaluate whether using an off-the-shelf anaphora resolution algorithm can improve the performance of a baseline opinion mining system. We present an analysis based on two different anaphora resolution systems. Our experiments on a movie review corpus demonstrate that an unsupervised anaphora resolution algorithm significantly improves the opinion target extraction. We furthermore suggest domain and task specific extensions to an off-the-shelf algorithm which in turn yield significant improvements. 1 Introduction Over the last years the task of opinion mining OM has been the topic of many publications. It has been approached with different goals in mind Some research strived to perform subjectivity analysis at the document or sentence level without focusing on what the individual opinions uttered in the document are about. Other approaches focused on extracting individual opinion words or phrases and what they are about. This aboutness has been referred to as the opinion target or opinion topic in the literature from the field. In this work our goal is to extract opinion target - opinion word pairs from sentences from movie reviews. A challenge which is frequently encountered in text mining tasks at this level of granularity is that entities are being referred to by anaphora. In the task of OM it can therefore also be necessary to analyze more than the content of one individual sentence when extracting opinion targets. Consider this example sentence Simply put it s unfathomable that this movie