Knowledge Based Shopping Prediction Through E-Commerce Application (P-0571) (MFN 4816)

Abstract

Searching for the desired product sometimes become very difficult as sometimes the amount of available products in e-commerce web based application is enormous in number so users face the difficulty of information overload .For the ease of customers to locate the products they would like to purchase we proposed Knowledge Based shopping Prediction recommendation system that enhances a solution to the information overload through knowledge learned from the behavior of consumers that is mostly based on customer Clicks Patterns, product’s Comments, product Ratings, Product Cart and customer Wish list proposed a knowledge based shopping prediction for an e-commerce website. Basket analysis technique used to able seeks to understand the purchase behavior of customers. The information gathered can then be used for purposes of cross-selling and up-selling, in addition to influencing sales promotions, loyalty programs, store (website. design, and discount plans. Shopping trend Analysis using Business Intelligent Tool in which made different dynamic graphs (fusion charts. for the admin. This application is helpful for many organizations and is used by large firms/industries around the world to improve business, increase sale & promotions by identifying with the help of charts where the business is performing well, where underperforming, provide evidence to inform decision making, use historical data to analyze trends and improve business. To guide every single user that browse the electronic store website through the often-overwhelming task of locating products they will like. As to increase the ” loyalty of the customers” , build a “value-added relationship” .This is very important since the competition is just one click away. Application will help to propose different recommendations to the personalized, non-personalized users by maintaining a user profile. Hybrid Approach is used that combines two or more recommendation techniques to get better recommendation and to overcome the weaknesses of individual technique. Here we combine both CF (collaborative filtering. and CNF (content based filtering. .Used basic PHP. Supported database MYSQL used at backend. Data Mining Algorithm that is KNN (K Nearest Neighbor. implemented in finding the shopping prediction on the statistical data stored in the database. (Item based + user based KNN. . Have done also done. In the end done Unit level testing (control flow graph, data flow graph. , scenarios based testing and GUI testing.