Abstract: The rapid growth of the laptop industry has led to a wide variation in product specifications and prices, making price prediction a challenging task. Consumers often struggle to identify whether a laptop is fairly priced, while retailers face difficulties in determining competitive pricing strategies. To address this challenge, this study proposes an integrated machine learning approach for laptop price prediction using Linear Regression, Decision Trees, and Random Forests.

The dataset consists of key laptop features such as brand, processor type, RAM size, storage capacity, graphics card, display characteristics, and operating system, which significantly influence the overall price. Before model development, preprocessing steps such as data cleaning, feature encoding, and normalization are performed to ensure consistency and accuracy.

Three predictive models are applied and compared:
Linear Regression provides a baseline by establishing a linear relationship between features and price.
Decision Trees capture non-linear relationships and offer rule-based interpretability.
Random Forests, as an ensemble method, combine multiple decision trees to enhance accuracy and reduce overfitting.

The performance of these models is evaluated using metrics such as Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and R² Score. The results indicate that Random Forest outperforms the other models, achieving higher accuracy and robustness, while Linear Regression and Decision Trees provide valuable interpretability and feature insights.

This integrated approach demonstrates the effectiveness of combining multiple machine learning techniques for price prediction tasks. The findings can assist consumers in making informed purchase decisions, e-commerce platforms in optimizing pricing strategies, and manufacturers in competitive market analysis. Furthermore, this research highlights the potential of machine learning in real world pricing applications and lays the foundation for future exploration with advanced models such as Gradient Boosting and Deep Learning.

Downloads: | DOI: 10.17148/IARJSET.2025.121033