Sentiment Analysis

Sentiment Analysis

A. M. Abirami (Thiagarajar College of Engineering, India), A. Sheik Abdullah (Thiagarajar College of Engineering, India), A. Askarunisa (KLN College of Information Technology, India), S. Selvakumar (G. K. M. College of Engineering and Technology, India) and C. Mahalakshmi (Thiagarajar College of Engineering, India)
Copyright: © 2017 |Pages: 13
DOI: 10.4018/978-1-5225-2031-3.ch009
OnDemand PDF Download:
Available
$37.50
Current Special Offers
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

It requires sophisticated streaming of big data processing to process the billions of daily social conversations across millions of sources. Dataset needs information extraction from them and it requires contextual semantic sentiment modeling to capture the intelligence through the complexity of online social discussions. Sentiment analysis is one of the techniques to capture the intelligence from Social Networks based on the user generated content. There are more and more researches evolving about sentiment classification. Aspect extraction is the core task involved in aspect based sentiment analysis. The proposed modeling uses Latent Semantic Analysis technique for aspect extraction and evaluates senti-scores of various products under study.
Chapter Preview
Top

Types Of Sentiment Analysis

Sentiment Analysis is one of the applications of Information Extraction (IE). It is defined as the automatic extraction of subjective content from digital text and predicting the subjectivity as positive or negative. The types are illustrated in Figure 1.

Figure 1.

Sentiment analysis approach

978-1-5225-2031-3.ch009.f01

Sentiment analysis is performed at various levels as follows:

  • •

    Document level classification

  • •

    Sentence level classification

  • •

    Aspect or Feature level classification and Summarization

    • o

      Named Entity Recognition

    • o

      Feature level analysis

Document Level

The main task of this level is to classify whether positive or negative sentiment expressed in the whole document. For example, given a product review, the system determines whether the review expresses an overall positive or negative opinion about the product. This level of analysis applicable for single entity not for multiple entities

Sentence Level

The task of this level is to identify whether each sentence expresses positive, negative or neutral opinion. This is related to subjectivity analysis. All these sentence level sentiment classifications can be summarized to get overall opinion about the product.

Complete Chapter List

Search this Book:
Reset