Publication

Abstract : Smartwatches have become adaptable instruments for ongoing health monitoring in recent years, and their ability to provide real-time physiological data holds the promise of revolutionizing healthcare. Understanding and projecting the cost of smartwatches is essential for consumers and businesses in a world where the market for these devices is proliferating. With the brand and model as crucial factors, this effort attempted to address the problem of precisely predicting smartwatch pricing. Four machine learning models are investigated in this research to create prediction solutions: Gradient Boosting ((MSE): 23487.86, (R2): 0.27), Decision Tree Regression ((MSE): 94115.03 (R2): −1.91), Linear Regression ((MSE): 40830.96, (R2): 0.00) and Random Forest model (MSE): 272942.21 (R2): 0.03. Customers will be able to make better-informed judgments about what to buy and get the most out of their investment. With the help of these predictive models, producers and merchants can set pricing that is competitive and specifically catered to each brand and model. These models, beyond price, provide market intelligence that can guide positioning and strategy in the ever-evolving smartwatch industry.