Online reviews have become the most critical part of decision-making for online purchases. Online reviews are in the form of unstructured data have both positive and negative impacts. First of all the consumers are getting the real experience of the peers about the product or services, which helps them make intelligent decisions about the product or service. The rise in the e-commerce sites has provided the open platform to the consumers to write their opinions which is also led to the big data challenge for both online businesses and consumers. Due to the information overload, sometimes consumers cannot go through all the reviews to make their purchase decisions. For example, averagely ranked product in Amazon.com have several hundred reviews. whereas for a popular product, reviews in thousands. In such situations, it is very difficult to read all the reviews before making their decisions. It would be very helpful to consumers if they had the high level of visibility of the few helpful reviews which helps to make better decisions. This is traditionally done just by asking simple questions “Was this review helpful to you?” and placing “Thums up” and “Thums down” buttons. And, helpfulness is predicted by counting the number votes received by each review. But, this approach will not work efficiently because the useful review posted recently buried beneath even before they get the votes. Reviews which are exposed to consumers attracts more votes and remaining reviews will attract fewer votes. As a result of this consumer, decision-making is mostly influenced by a number of votes rather than when the review is posted and the context of the review. Hence, it is necessary to develop methods to find the useful review’s and order them accordingly in order to mitigate the Mathew effect. which implies that top reviews gain more helpfulness votes since they expose to the consumers and lower reviews gain fewer votes because of less exposure cite{Wan2015}.

x

Hi!
I'm Joan!

Would you like to get a custom essay? How about receiving a customized one?

Check it out