Book contents
- Frontmatter
- Contents
- Preface
- Acknowledgments
- 1 Introduction
- 2 The Problem of Sentiment Analysis
- 3 Document Sentiment Classification
- 4 Sentence Subjectivity and Sentiment Classification
- 5 Aspect Sentiment Classification
- 6 Aspect and Entity Extraction
- 7 Sentiment Lexicon Generation
- 8 Analysis of Comparative Opinions
- 9 Opinion Summarization and Search
- 10 Analysis of Debates and Comments
- 11 Mining Intentions
- 12 Detecting Fake or Deceptive Opinions
- 13 Quality of Reviews
- 14 Conclusions
- Appendix
- Bibliography
- Index
4 - Sentence Subjectivity and Sentiment Classification
Published online by Cambridge University Press: 05 June 2015
- Frontmatter
- Contents
- Preface
- Acknowledgments
- 1 Introduction
- 2 The Problem of Sentiment Analysis
- 3 Document Sentiment Classification
- 4 Sentence Subjectivity and Sentiment Classification
- 5 Aspect Sentiment Classification
- 6 Aspect and Entity Extraction
- 7 Sentiment Lexicon Generation
- 8 Analysis of Comparative Opinions
- 9 Opinion Summarization and Search
- 10 Analysis of Debates and Comments
- 11 Mining Intentions
- 12 Detecting Fake or Deceptive Opinions
- 13 Quality of Reviews
- 14 Conclusions
- Appendix
- Bibliography
- Index
Summary
As discussed in the previous chapter, document-level sentiment classification is too coarse for practical applications. We now move to the sentence level and look at methods that classify sentiment expressed in each sentence. The goal is to classify each sentence in an opinion document (e.g., a product review) as expressing a positive, negative, or neutral opinion. This gets us closer to real-life sentiment analysis applications, which require opinions on sentiment targets. Sentence-level classification is about the same as document-level classification because sentences can be regarded as short documents. Sentence-level classification, however, is often harder because the information contained in a typical sentence is much less than that contained in a typical document because of their length difference. Most document-level sentiment classification research papers ignore the neutral class mainly because it is more difficult to perform three-class classification (positive, neutral, and negative) accurately. However, for sentence-level classification, the neutral class cannot be ignored because an opinion document can contain many sentences that express no opinion or sentiment. Note that neutral opinion often means no opinion or sentiment expressed.
One implicit assumption that researchers make about sentence-level classification is that a sentence expresses a single sentiment. Let us start our discussion with an example review:
I bought a Lenovo Ultrabook T431s two weeks ago. It is really light, quiet and cool. The new touchpad is great too. It is the best laptop that I have ever had although it is a bit expensive.
The first sentence expresses no sentiment or opinion as it simply states a fact. It is thus neutral. All other sentences express some sentiment. Sentence-level sentiment classification is defined as follows:
Definition 4.1 (Sentence sentiment classification): Given a sentence x, determine whether x expresses a positive, negative, or neutral (or no) opinion.
As we can see, like document-level sentiment classification, sentence-level sentiment classification also does not consider opinion or sentiment targets. However, in most cases, if the system is given a set of entities and their aspects, the sentiment about them in a sentence can just take the sentiment of the sentence.
- Type
- Chapter
- Information
- Sentiment AnalysisMining Opinions, Sentiments, and Emotions, pp. 70 - 89Publisher: Cambridge University PressPrint publication year: 2015
- 1
- Cited by