Opinion Target Extraction Using Word-Based Translation Model

Proceedings of the 2012 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning, pages 1346–1356, Jeju Island, Korea, 12–14 July 2012. c⃝2012 Association for Computational Linguistics Opinion Target Extraction Using Word-Based Translation Model Kang Liu, Liheng Xu, Jun Zhao National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, Beijing, 100190, China {kliu, lhxu, jzhao}@nlpr.ia.ac.cn Abstract This paper proposes a novel approach to extract opinion targets based on word- based translation model (WTM). At first, we apply WTM in a monolingual scenario to mine the associations between opinion targets and opinion words. Then, a graph- based algorithm is exploited to extract opinion targets, where candidate opinion relevance estimated from the mined associations, is incorporated with candidate importance to generate a global measure. By using WTM, our method can capture opinion relations more precisely, especially for long-span relations. In particular, compared with previous syntax-based methods, our method can effectively avoid noises from parsing errors when dealing with informal texts in large Web corpora. By using graph-based algorithm, opinion targets are extracted in a global process, which can effectively alleviate the problem of error propagation in traditional bootstrap-based methods, such as Double Propagation. The experimental results on three real world datasets in different sizes and languages show that our approach is more effective and robust than state-of-art methods. 1 Introduction With the rapid development of e-commerce, most customers express their opinions on various kinds of entities, such as products and services. These reviews not only provide customers with useful information for reference, but also are valuable for merchants to get the feedback from customers and enhance the qualities of their products or services. Therefore, mining opinions from these vast amounts of reviews becomes urgent, and has attracted a lot of attentions from many researchers. In opinion mining, one fundamental problem is opinion target extraction. This task is to extract items which opinions are expressed on. In reviews, opinion targets are usually nouns/noun phrases. For example, in the sentence of “The phone has a colorful and even amazing screen”, “screen” is an opinion target. In online product reviews, opinion targets often are products or product features, so this task is also named as product feature extraction in previous work (Hu et al., 2004; Ding et al., 2008; Liu et al., 2005; Popescu et al., 2005; Wu et al., 2005; Su et al., 2008). To extract opinion targets, many studies regarded opinion words as strong indicators (Hu et al., 2004; Popescu et al., 2005; Liu et al., 2005; Qiu et al., 2011; Zhang et al., 2010), which is based on the observation that opinion words are usually located around opinion targets, and there are associations between them. Therefore, most pervious methods iteratively extracted opinion targets depending upon the associations between opinion words and opinion targets (Qiu et al., 2011; Zhang et al., 2010). For example, “colorful” and “amazing” is usually used to modify “screen” in reviews about cell phone, so there are strong associations between them. If …

Content truncated to 500 words