Price forecasting in e-commerce using neural network approach

Abstract

The advancement of internet and better communication technology has led to the development of e-commerce. E-commerce is buying and selling of products or services over the computer network. With the incorporation of e-commerce in daily life, enterprises no longer rely on traditional business tactics. They have to cope up with dynamic and uncertain electronic environment, especially when developing cost planning. In such a situation prices are calculated dynamically through an aggressive bidding process depending on the market position, competitor policy and customer preferences. The capability to predict the price of any product is a crucial task. This paper introduces neural network approach of predicting future value of a product. Basics of neural network, different activation function and different network architecture are also explored. Training with different datasets taken from different online shopping website show the potential of proposed algorithm for forecasting price in the competing and uncertain environment.

Keywords—neural network;  price forecasing; styling; e-commerce