A Survey on Classification Techniques for Feature-Sentiment Analysis

Abstract

As use of internet and its application are growing exponentially; the e-commerce business i.e. online purchase is proportionately swelling in the world. The e-commerce websites and similar service providing websites are providing a rich variety of product and service to be sold. As the quality of service and product/goods has much effect on its sell, the websites nowadays tends to have public opinion on the product in the form of feedback; we can name it as reviews. These reviews provide much information about the service/product as the customers are encouraged to write their reviews cum assessments about the product, more precisely saying, customer writes their view about product’s specifications or product’s features. These unrestricted or restricted opinions from public can then be considered by the customers and vendor to make the required design/engineering/production changes to the product to upsurge its quality. The Feature Mining along with Sentiment Analysis techniques can be applied to achieve product’s feature and public opinion on these features. Here in this paper we are interestingly motivated by the scenario as discussed above. We had a survey on the different methods cum techniques that can be usually used to extract products/service features and categorizing those feature along with the sentiment classification on the determined features which is part of Machine learning. The public opinions can be classified as positive, negative and neutral sentimentalities. Research area ‘Data Mining’ has proven its importance with its rich set of Machine Learning Algorithms which in turn can be used as Sentiment or Opinion Classifier. After evaluating feature-sentiment techniques, we then studied the feature classification/categorizing by using its overall sentiment and influence on the product/service sell.
DOI: 10.17762/ijritcc2321-8169.150796