Aspect-Based Sentiment Analysis of Online Product Reviews

Aspect-Based Sentiment Analysis of Online Product Reviews

Vinod Kumar Mishra (Bipin Tripathi Kumaon Institute of Technology, India) and Himanshu Tiruwa (Bipin Tripathi Kumaon Institute of Technology, India)
DOI: 10.4018/978-1-5225-2031-3.ch010
OnDemand PDF Download:
List Price: $37.50


Sentiment analysis is a part of computational linguistics concerned with extracting sentiment and emotion from text. It is also considered as a task of natural language processing and data mining. Sentiment analysis mainly concentrate on identifying whether a given text is subjective or objective and if it is subjective, then whether it is negative, positive or neutral. This chapter provide an overview of aspect based sentiment analysis with current and future trend of research on aspect based sentiment analysis. This chapter also provide a aspect based sentiment analysis of online customer reviews of Nokia 6600. To perform aspect based classification we are using lexical approach on eclipse platform which classify the review as a positive, negative or neutral on the basis of features of product. The Sentiwordnet is used as a lexical resource to calculate the overall sentiment score of each sentence, pos tagger is used for part of speech tagging, frequency based method is used for extraction of the aspects/features and used negation handling for improving the accuracy of the system.
Chapter Preview

Applications Of Sentiment Analysis

Sentiment occurs in almost all human activity because they describe the human behaviors. Whenever the human want to take a decision they want to know about others opinion. In real words, all business organization and all politicians want another opinion about their product and political condition even individual consumer also wants to know the opinion of the other user before purchasing any new product. In the past, whenever we want to take any decision we discuss it with our family member and friends for personel decisions and for taking any decision for an organizing we conducted the surveys. These traditional method are very time consuming and always does not give a good result.Application as a sub-component technology:Sentiment analysis plays an important role for enabling one technology for other system e.g. it cannot recommend those systems to be acquired with other if the systems contain more negative review. Detection of flame in the email and other type of communication is another use of sentiment classification. An online system, it helpful to take only the positive material in their system avoids all negative ads and content to be added to the online system and also helpful to the improve the online system and also helpful in the citation analysis, resulted increases the human computer interaction.

Key Terms in this Chapter

Maximum Entropy Classifier: This classifier determine the most likely class for a document set it convert the labelled document set into a vector using encoding and with the help of encoded vector we calculate the weight of a document and combine to get the result.

Naive Bayes Classifier: It is a simple probability based classifier based on bayes theorem with strong independence assumption it compute the posterior probability of a class based on the distribution of the word in the document it ignore the position of the word in the document.

Decision Tree Classifiers: In decision tree classifiers we provides the training data in the hierarchical form and for dividing the data we used the conditions on attributes and to perform the spilt we use the similarity based multi-attribute split.

Support Vector Machine (SVM): It determine the linear separator by constructing hyperlanes that separates the cases that belong to different categories. Hyperlanes provide best separation between the class and it takes the maximum margin for separation because the normal distance between any data point is largest.

Rule-Based Classifiers: Rule based classifier is a sets of rule for classification and uses term absence and presence for the classification. In this classifier we define some criteria to generate the rule and theses rule are generates at the training time.

Bayesian Network (BN): Bayesian network iis the directed acyclic graph in which the nodes represent the random variable and the edge represent the conditional dependencies. it specified the joint probability distribution over all the variable.

Neural Network (NN): Neural network is combination of neuron. Neural network is used for both classification and prediction. Neural network contain three layers for computation first is input layer second is hidden layer and last is output layer.

Complete Chapter List

Search this Book: