Tool For Review Analysis Of Product

Today people buy products online rather than manual shopping. These e-commerce websites let their customers write product feedback and review in the form of rating. The review given by these customers helps the company personnel of the product understand where their product stands in the market. At the same time help other fellow customers decide whether this product is suitable for them to buy. Reviews can be fraudulent in a way that they demote the product or advertising different products which might misdirect the customer about the quality of the product. Hence, we have worked towards developing a tool which will classify the reviews as fake or genuine and provide it to the user. The proposed tool will operate on the principle of the Weighted Ensemble Classifier. The said ensemble classifier uses Support Vector Machine (SVM), Naive Bayes (NB) and K-Nearest Neighbor (KNN) classifiers. Experimental results show that the proposed ensemble classifier is efficient in the fake review detection task.