Published Paper Details:

The growth of online shopping is rapid as people buy goods over the internet by selecting items and paying for them on the internet. This has helped make shopping easier and quicker. With the popularisation of online shopping, a large amount of data about users' online activities is available on the web. Such data could be used for understanding user behaviour and classifying online customers.The proposed system aims to extract the patterns of the online shopping behaviour of the users and to display them in graphical form. These graphical representations help the administrators in the decision making process by identifying the products that are frequently purchased and the products that are in demand at the moment. It combines multiple information sources and simulates decision-making by using the statistical features of graphical hidden state model.The proposed model has achieved a performance improvement of about 30% on million-click datasets. The application is a web platform on the internet where users can view and buy products. User behaviour is tracked in sequence and then analysed and graphically displayed. This allows administrators to make decisions more quickly and efficiently. The graphical approach simplifies the visualisation of data, so administrators can quickly gain useful insights by watching patterns in customer behaviour.