Estimating the Helpfulness and Economic Impact of Product Reviews

Abstract

The rapid growth of the internet, the ability of users to create and publish product content has created active electronic communities that provide wealth of that information. A large number of reviews for a single product may also make it harder for individuals and evaluate the true underlying quality of a product. We reexamine the impact of reviews that product sales and how different factors affect social outcomes. Our approach explores multiple aspects of review text, such as subjectivity levels, various measures of readability and extent of spelling errors to identify text-based features. We examine multiple reviewer-level features such as average of past reviews and identify the measure of reviewers that are displayed to next review. Econometric analysis reveals the extent of subjectivity, readability, in formativeness, correctness in review matters and subject sentence are negatively associated with product sales. These all give the aware of product usefulness. Then reviews are rated by using "random forest-based classifiers". These are more informative by other users. We examine three broad feature categories: "reviewer-related" features, "review subjectivtty" features and "review readability" features. These three features sets results in a statistically performance. Using this text mining and product modeling techniques how prevent the text of reviews affect product sales and the perceived helpfulness of these reviews.