Analysis of Online Discount Sales and Price Optimization Using Cognitive Learning

Prices of products are usually static during discount sales, even when customers' demand for a particular product exceeds the available stock. As such, the economic principle that states, ―if demand exceeds supply, there will be scarcity and prices will naturally go high‖, is not satisfied. However, this research focuses on analyzing online discount sales and optimizing pricing decisions based on the wealth of data deposited in a retailer's database using cognitive learning. Customers information as they visit the Inspired Network's website during the online discount sales, are considered as input to the system. A database is created to record and store information of customers who visit the website. It is based on these information that the retailer monitors and analyzes his customers shopping pattern, competitive price, inventory, profit margin and other array of factors needed to improve sales, thus, as such prices can change. A binary purchasing decision ―Change (indicating a 1)/No Change (indicating a 0)‖ is utilized to effect the change based on the number of demand placed on the product(s). Thus, if the demand for a particular product is very high during discount sales, its price increases while high sales will still be achieved and if the demand for the product is low, its price might be further reduced. The input datasets are subjected to training, validation and testing in order to achieve a valid prediction.