Annual Academic Achievement Level Prediction Based on Machine Learning

With the continuous development and progress of machine learning related technologies, this technology is increasingly being used by people from all walks of life to solve various problems encountered in the industry. This paper combines machine learning technology with the field of student management in this application context, and proposes a machine learning based annual academic achievement level prediction method. The method first uses the academic achievements exported from the academic administration system as the data source; Then use visual charts to preliminarily visualize the raw data, try to observe the data, and draw simple conclusions; Next, the machine learning model (using the random forest algorithm in this article) was trained, validated, and tested using the training set, validation set, and test set of the data. The optimal model parameter was found to be 17 decision trees, and the model scored 1.0 in the training set, 0.8667 in the validation set, and 1.0 in the test set; Finally, through actual testing, it was found that the model accurately predicted the results for the new samples obtained, and it can be widely used in the future.