Comparing the Performance of Four Machine Learning Algorithms in Prediction

Abstract

To investigate and compare the performance of different machine learning algorithms in forecasting, this paper presents a case study on predicting industrial land use. We conduct a benchmarking study employing four representative algorithms: Linear Regression, Random Forest, Naïve Bayes, and Artificial Neural Networks (ANN). The historical data of Shandong Province from 2001 to 2022 was established as the dataset for this study, in which GDP, FAI(Fixed Asset Investment), and IOVAS(Industrial Output Value Above Scale) were set as independent variables, while ILV(Industrial Land Volume)was used as the dependent variable, and four machine learning algorithms, utilized the dataset to complete the training and testing of the models, and finally, this study provides the accuracy of these four machine learning algorithms in industrial land use prediction.