Smartphone Price Range Prediction Using Machine Learning-Based Classification Models

Abstract

The rapid expansion of the smartphone market has introduced a wide variety of devices with diverse hardware configurations and price segments. As manufacturers continue to release new models with varying technical specifications, estimating the appropriate price category has become increasingly complex for both consumers and businesses. An automated prediction system can simplify this process by learning the relationship between smartphone specifications and their corresponding market price ranges. This study proposes a machine learning-based classification framework for smartphone price range prediction using hardware characteristics as predictive features. The proposed approach utilizes a publicly available smartphone dataset containing twenty technical attributes describing battery capacity, memory, processor characteristics, display properties, camera specifications, and connectivity options. Before model development, the dataset undergoes preprocessing to remove duplicate entries, verify data consistency, handle missing information, and normalize numerical features through standardization. Six supervised machine learning algorithms were investigated, namely Logistic Regression, Decision Tree, Random Forest, Extra Trees Classifier, K-Nearest Neighbors, and Support Vector Machine. Their predictive performance was evaluated using identical training and testing datasets with accuracy as the primary evaluation criterion. Comparative analysis demonstrated that Logistic Regression achieved the highest classification accuracy of 97.5%, providing an effective balance between prediction performance, computational efficiency, and model simplicity. To demonstrate the practical applicability of the proposed model, the selected classifier was integrated into a Streamlit-based web application that enables users to enter smartphone specifications and receive an instant prediction of the expected price category. The developed system illustrates that carefully preprocessed structured data combined with classical supervised learning techniques can provide reliable and efficient smartphone price range prediction. The proposed framework can assist consumers during purchasing decisions and support retailers and manufacturers in preliminary pricing analysis while serving as a scalable foundation for future intelligent pricing systems.

Introduction

This study presents a machine learning-based framework for smartphone price range prediction using hardware specifications. With the rapid growth of smartphone models and features, manually estimating a device's price category has become difficult. The research applies supervised machine learning to classify smartphones into predefined price ranges based on technical specifications such as RAM, storage, processor speed, battery capacity, display quality, camera features, and connectivity.

A comprehensive preprocessing pipeline—including data validation, duplicate removal, missing value treatment, and feature standardization—was applied to a publicly available dataset containing approximately 2,000 smartphone records. Six supervised classification algorithms were trained and compared under identical conditions: Logistic Regression, Decision Tree, Random Forest, Extra Trees Classifier, K-Nearest Neighbors (KNN), and Support Vector Machine (SVM). Their performance was evaluated using accuracy, precision, recall, F1-score, and confusion matrix.

Among the evaluated models, Logistic Regression achieved the highest accuracy of 97.5%, outperforming more complex ensemble methods. The findings demonstrate that proper preprocessing and feature scaling enable a simple linear classifier to effectively predict smartphone price categories with high accuracy and low computational cost.

The best-performing model was deployed as a Streamlit-based web application, allowing users to input smartphone specifications and receive real-time price range predictions. This deployment illustrates the practical application of machine learning for assisting consumers in smartphone selection and supporting retailers in pricing decisions.

Conclusion

This paper presented a machine learning framework for predicting smartphone price ranges based on their technical specifications. The study addressed the challenge of estimating smartphone price categories by utilizing supervised learning techniques instead of traditional manual comparison methods. A structured workflow consisting of data preprocessing, feature standardization, model training, comparative evaluation, and web-based deployment was developed to ensure accurate and reliable prediction. The smartphone dataset was carefully prepared through duplicate removal, validation, missing value treatment, and feature scaling before model development. Six supervised classification algorithms—Logistic Regression, Decision Tree, Random Forest, Extra Trees Classifier, K-Nearest Neighbors, and Support Vector Machine—were trained and evaluated using the same experimental conditions. Their performance was compared using standard classification metrics to identify the most suitable prediction model. Among the investigated algorithms, Logistic Regression achieved the highest classification accuracy of 97.5%, demonstrating that a relatively simple linear classifier can effectively model the relationship between smartphone hardware specifications and predefined price categories when appropriate preprocessing techniques are applied. In addition to providing high prediction accuracy, the selected model offers low computational complexity, fast inference, and ease of deployment, making it well suited for practical applications. To demonstrate real-world applicability, the trained model was integrated into a Streamlit-based web application. The application enables users to enter smartphone specifications through an intuitive interface and obtain instant predictions of the expected price category. This deployment illustrates how machine learning models can be transformed into practical decision-support tools that assist consumers in selecting suitable smartphones and help retailers perform preliminary pricing analysis. Overall, the proposed framework demonstrates that classical supervised learning techniques remain highly effective for structured classification tasks. The combination of systematic preprocessing, comparative model evaluation, and lightweight deployment provides a reliable and efficient solution for smartphone price range prediction.

References

[1] M. Cetin and Y. Koc, \"Mobile Phone Price Class Prediction Using Different Classification Algorithms with Feature Selection and Parameter Optimization,\" Proc. International Conference, pp. 1–6, 2021. [2] K. S. Kalaivani, N. Priyadharshini, S. Nivedhashri, and R. Nandhini, \"Predicting the Price Range of Mobile Phones Using Machine Learning Techniques,\" Proc. International Conference on Intelligent Computing, pp. 1–6, 2021. [3] E. Güvenç, G. Cetin, and H. Koçak, \"Comparison of KNN and DNN Classifiers Performance in Predicting Mobile Phone Price Ranges,\" International Journal of Computer Applications, vol. 183, no. 15, pp. 15–22, 2021. [4] F. Pedregosa et al., \"Scikit-learn: Machine Learning in Python,\" Journal of Machine Learning Research, vol. 12, pp. 2825–2830, 2011. [5] A. Géron, Hands-On Machine Learning with Scikit-Learn, Keras& TensorFlow, 3rd ed. Sebastopol, CA, USA: O\'Reilly Media, 2022. [6] Q. Liang, \"Mobile Phone Price Prediction: A Comparative Study among Four Models,\" Applied and Computational Engineering, vol. 48, pp. 212–220, 2024. [7] J. Lu, \"Prediction on Mobile Phone Price Range Using Classification Models,\" Science and Technology of Engineering, Chemistry and Environmental Protection, vol. 1, no. 10, pp. 45–53, 2024. [8] S. Agalayil, N. Shajan, and K. N. Shreyas, \"Prediction of Mobile Phone Prices Using Machine Learning,\" in Proceedings of the 9th International Conference on Machine Learning Technologies (ICMLT), 2024. [9] Y. Chen, \"Prediction of Different Types of Mobile Phone Prices Based on Machine Learning Models,\" Highlights in Science, Engineering and Technology, vol. 92, pp. 275–279, 2024. [10] G. S. Parveen, A. Khalique, R. Islam, and I. Hussain, \"Prediction of Mobile Phone Price Using Machine Learning Classifiers,\" Journal of Artificial Intelligence Research & Advances, vol. 11, no. 2, pp. 101–108, 2024. [11] R. Wahyusari and N. Azizah, \"Predicting Smartphone Prices Based on Key Features Using Random Forest and Gradient Boosting Algorithms in a Data Mining Framework,\" International Journal for Applied Information Management, vol. 5, no. 2, pp. 73–85, 2025. [12] X. Li, \"Smartphone Price Prediction Using Decision Tree and Support Vector Regression,\" in Proceedings of the 5th International Conference on Signal Processing and Machine Learning, 2025. [13] T. Hastie, R. Tibshirani, and J. Friedman, The Elements of Statistical Learning, 2nd ed. Springer, 2009. [14] C. M. Bishop, Pattern Recognition and Machine Learning. Springer, 2006. [15] I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning. MIT Press, 2016.

Copyright

Copyright © 2026 Amreen Jahan, Tulasi Miryala. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.