Using Feature Importance And Decision Tree To Analyse Examination Score

It is a very important part that using machine learning and data mining methods toanalyze the examination score and further extract the useful knowledge. The currentstudies on examination score analysis can be regarded as the design to a supervisedlearning system, that is to say, classifying the examination score data with the specificdata mining methods. The main limitation of the current studies is that these methods allneglect the underlying information hidden in the examination score data, e.g., thedependence between different courses and the impact of this dependence on thecomprehensive evaluation. So we need to improve the defect of existing examinationscore analysis methods,This thesis presents the following three main contributions based on the examinationscores deriving from the10different adult education projects. Firstly, preprocess theexamination score data by using the feature selection based on mutual informationtechnology so as to determine the maximally dependent courses with comprehensiveevaluation and meanwhile reduce the teaching time of weak correlation. Secondly,conduct the data mining based on the preprocessed datasets and extract the relatedlearning rules in order that the students can make the learning project as convenient aspossible. Finally, carry out the detailed experiments to compare the performance of threedifferent decision tree algorithms. The experimental results show that the proposedstrategy is feasible and effective. And, it plays a very important role for students in thelearning project making and the improvement of examination score.