V3: Unsupervised Aspect Based Sentiment Analysis for SemEval2015 Task 12

Proceedings of the 9th International Workshop on Semantic Evaluation (SemEval 2015), pages 714–718, Denver, Colorado, June 4-5, 2015. c⃝2015 Association for Computational Linguistics V3: Unsupervised Aspect Based Sentiment Analysis for SemEval-2015 Task 12 Aitor Garc´ıa-Pablos, Montse Cuadros Vicomtech-IK4 research center Mikeletegi 57 San Sebastian, Spain {agarciap,mcuadros}@vicomtech.org German Rigau IXA Group Euskal Herriko Unibertsitatea, San Sebastian, Spain german.rigau@ehu.es Abstract This paper presents our participation in SemEval-2015 task 12 (Aspect Based Senti- ment Analysis). We participated employing only unsupervised or weakly-supervised ap- proaches. Our attempt is based on requiring the minimum annotated or hand-crafted con- tent, and avoids training a model using the provided training set. We use a continuous word representations (Word2Vec) to leverage in-domain semantic similarities of words for many of the involved subtasks. 1 Introduction The continuous growing of textual content on the In- ternet has motivated an important research on find- ing automatic ways of processing and exploiting this valuable source of information. That is one of the reasons why sentiment analysis has become a very active research field during the last decade (Pang and Lee, 2008; Liu et al., 2012; Zhang and Liu, 2014). Sentiment analysis aims to detect and classify the polarity of sentiments expressed in a text. The gran- ularity of this classification goes from the overall polarity of full documents to paragraphs, sentences or, as in Aspect Based Sentiment Analysis (ABSA), the sentiment about precise aspects being opinion- ated (Hu and Liu, 2004) (Popescu and Etzioni, 2005) (Wu et al., 2009) (Zhang et al. , 2010). In this paper we describe our participation in SemEval-2015 task 121 (Pontiki et al., 2015), which is about ABSA. We have participated in all subtasks 1http://alt.qcri.org/semeval2015/task12/ employing unsupervised or weakly supervised ap- proaches. The rest of the paper is structured as follows. Sec- tion 2 introduces the SemEval-2015 task 12 compe- tition and provided datasets, and a brief introduction about how we have approached the different slots. Sections 3, 4 and 5 describe more in detail the em- ployed techniques. Section 6 shows the results of the evaluation, and finally section 7 summarizes the conclusions. 2 Our approach SemEval2015 task 12 was about ABSA. The task was divided into 3 slots. Slot 1 was about classify- ing review sentences into entity-attribute pairs, be- ing the entity a main aspect of the reviewed item (e.g. food, drinks, location) and the attribute a par- ticular facet of that aspect (e.g. food-quality, food- price, etc.). Slot 1 runs on two domains, restaurants and laptops. Slot 2 was about detecting explicit men- tions to aspect-terms that are being reviewed (e.g. service in ”The service was attentive”). Slot 2 runs only on restaurants domain. Slot 3 was about detect- ing the polarity/sentiment for the given gold entity- attribute pairs in sentences (see slot 1). Slot 3 was meant for restaurants and laptops domain, plus an additional hidden domain (i.e. revealed in the last moment and with no training data available) which resulted to be about hotels. Two training datasets were provided. The first …

Content truncated to 500 words