Sentiment Analysis of Conditional Sentences

Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing, pages 180–189, Singapore, 6-7 August 2009. c⃝2009 ACL and AFNLP Sentiment Analysis of Conditional Sentences Ramanathan Narayanan Dept. of EECS Northwestern University ramanathan.an@gmail.com Bing Liu * Dept. of Computer Science Univ. of Illinois at Chicago liub@cs.uic.edu Alok Choudhary Dept. of EECS Northwestern University alokchoudhary01@gmail.com Abstract This paper studies sentiment analysis of condi- tional sentences. The aim is to determine whether opinions expressed on different topics in a conditional sentence are positive, negative or neutral. Conditional sentences are one of the commonly used language constructs in text. In a typical document, there are around 8% of such sentences. Due to the condition clause, sentiments expressed in a conditional sentence can be hard to determine. For example, in the sentence, if your Nokia phone is not good, buy this great Samsung phone, the author is posi- tive about “Samsung phone” but does not ex- press an opinion on “Nokia phone” (although the owner of the “Nokia phone” may be nega- tive about it). However, if the sentence does not have “if”, the first clause is clearly nega- tive. Although “if” commonly signifies a con- ditional sentence, there are many other words and constructs that can express conditions. This paper first presents a linguistic analysis of such sentences, and then builds some super- vised learning models to determine if senti- ments expressed on different topics in a condi- tional sentence are positive, negative or neu- tral. Experimental results on conditional sen- tences from 5 diverse domains are given to demonstrate the effectiveness of the proposed approach. 1 Introduction Sentiment analysis (also called opinion mining) has been an active research area in recent years. There are many research directions, e.g., senti- ment classification (classifying an opinion doc- ument as positive or negative) (e.g., Pang, Lee and Vaithyanathan, 2002; Turney, 2002), subjec- tivity classification (determining whether a sen- tence is subjective or objective, and its associated opinion) (Wiebe and Wilson, 2002; Yu and Hat- zivassiloglou, 2003; Wilson et al, 2004; Kim and Hovy, 2004; Riloff and Wiebe, 2005), fea- ture/topic-based sentiment analysis (assigning positive or negative sentiments to topics or prod- uct features) (Hu and Liu 2004; Popescu and Et- zioni, 2005; Carenini et al., 2005; Ku et al., 2006; Kobayashi, Inui and Matsumoto, 2007; Titov and McDonald. 2008). Formal definitions of different aspects of the sentiment analysis problem and discussions of major research direc- tions and algorithms can be found in (Liu, 2006; Liu, 2009). A comprehensive survey of the field can be found in (Pang and Lee, 2008). Our work is in the area of topic/feature-based sentiment analysis or opinion mining (Hu and Liu, 2004). The existing research focuses on solving the general problem. However, we argue that it is unlikely to have a one-technique-fit-all solution because different types of sentences ex- press sentiments/opinions in different ways. A divide-and-conquer approach is needed, e.g., fo- cused studies on different types of sentences. This paper focuses on one type of sentences, i.e., conditional …

Content truncated to 500 words