Real-Time Product Recommendations for Smart Shopping Using Lattice Structure

The IoT-enabled smart shopping cart system incorporates a recommendation engine to enhance customer experience through real-time, personalized product suggestions. The system uses Most Generalization Association Rules (MGARs) mined from Frequent-Closed-Itemset Lattices (FCIL) to address challenges such as redundancy, computational inefficiency, and lack of scalability in traditional recommendation systems. By dynamically analyzing cart contents and historical purchase data, the system provides accurate and relevant recommendations, ensuring improved customer satisfaction and operational efficiency. Experimental results show that FCIL-based MGAR mining is significantly faster, reducing computation time from 0.0449s to 0.0018s at 1% minimum support. FCIL also generates fewer rules (220 vs. 1,287) and reduces memory consumption to 0.05 KB compared to 63.59 KB in FCI, ensuring better scalability and accuracy. These findings confirm the system's efficiency, making it ideal for real-time smart shopping applications