Prediction of Different Types of Mobile Phone Prices based on Machine Learning Models

Abstract

In contemporary society, mobile communication technology is advancing rapidly. People's lives have become more convenient, significantly driving the development of human society. As mobile communication introduces many new and practical technologies, people are increasingly inseparable from it. Mobile phones, as a form of mobile communication, have gradually integrated into people's lives and become a crucial and irreplaceable product. With the continuous popularization of smartphones and the increasing richness of mobile content, choosing a cost-effective phone has become a crucial decision for many. This article is based on hardware data and prices from a dataset comprising 2000 already-released smartphones. Machine learning techniques, specifically the K-Nearest Neighbors (KNN) model and linear regression model, are employed to predict the prices of these smartphones. The study reveals that both the K-Nearest Neighbors (KNN) model and the Linear Regression model can be utilized for predicting smartphone prices. However, the KNN model exhibits a slightly higher accuracy. Specifically, when K is set to 16, the accuracy of the KNN model reaches 93.3%. The Linear Regression model achieves an accuracy of 91.3%.