Open AccessMachine Learning Methods for Predicting the Popularity of Forthcoming Objects
Author(s) -
Gulab Sah,
Subhra Rajat,
Sunit Kumar Nandi
Publication year - 2019
Publication title -
international journal of innovative technology and exploring engineering
Language(s) - English
Resource type - Journals
ISSN - 2278-3075
DOI - 10.35940/ijitee.b1104.1292s19
Subject(s) - product (mathematics) , computer science , quality (philosophy) , popularity , set (abstract data type) , new product development , marketing , data science , operations research , business , engineering , mathematics , philosophy , geometry , epistemology , programming language , psychology , social psychology
Now a day, product ratings are very much essential for the product available online so that customers can view a product's actual rating before they are going to buy it. This is only the primary source of information for a product, and it is also essential for manufacturers, retailers to improve product quality in terms of production and sale.A rating can make it easy for consumers to figure out how much they enjoy the product. Now in case of new arrival products which have not been used by any customers or any users, the ratings not available online. We have tried to find ratings for new arrival products in this research work by identifying the quality of that product, which will assist customers before buying it. We have also examined different method that will predict the rating of the newest arrival product based on product features, description, information that are available on the e-commerce platform like Amazon, Flipchart. To achieve the defined goal, we have worked on existing data that are available for products already arrived in the market and already used by a customer. The main objective of this research is to help the customer who is going to purchase new arrival products. This is done by comparing different existing Machine Learning methods with the help of the existing data set. We have tried to find out the best method among the existing Machine learning methods and applied that method to predict the rating of the newest arrival product based on the available features.