Strongly correlated high-utility item-sets recommendation in e-commerce applications using EGUI-tree over data stream

The product recommendation feature in e-commerce applications plays a vital role in helping customers choose the right items. By analysing their purchase history, the system recommends relevant items. Although customers have the freedom to choose from the suggestions, if they do purchase them, it benefits both buyers and sellers. Sellers experience increased sales, while buyers save time searching for products. To meet these requirements, a recommendation system should prioritise items with High Utility (HU) and strong correlation. HU items generate high profits, and strongly correlated items have a higher probability of being selected. Here, since up-to-date data in the data stream are got, the stream of purchase transactions is mined using the sliding window technique to extract such item-sets. To store the immediately processed results, a tree like structure termed Extended Global Utility Item-sets Tree is used. Experimental results on real-time data prove the proposed algorithm is optimistic and scalable.