Evaluation and Comparison of Ten Machine Learning Classification Models Based on the Mobile Users Experience

Machine learning prediction is a statistical prediction model based on computer algorithms, which has been very popular with the development of AI. It is the most commonly used tool in big data prediction and data mining. To establish an effective machine learning model and get better prediction accuracy, we must choose a suitable algorithm based on actual problems and then fully improve the model. This article lists ten commonly used machine learning models and train and fit these classification models in Python based on the data set provided by the 2022 MathorCup Big Data Competition. Meanwhile, by establishing feature engineering and choosing the best hyperparameter combination, we got the best performance of the model’s prediction. The result shows that among the ten Machine learning models, the CatBoost Classifier had the highest accuracy 0.7849, while the XGBoost Classifier had the highest macro-F1 score 0.4898; after hyperparameter tuning, the best macro-F1 score of XGBoost Classifier could attain 0.5216.