Assessing　students＇　code　quality　is　a　major　workload　for　teachers,　making　research　on　automated　systems　crucial.　We　developed　an　AI-based　system　to　automate　code　quality　assessment,　providing　immediate　feedback　to　students　and　reducing　teachers＇　burden.　We　evaluated　multiple　machine　learning　models,　including　Multilayer　Perceptron　(MLP),　k-　Nearest　Neighbors　(k-NN),　Naive　Bayes　(NB),　Support　Vector　Machine　(SVM),　and　Convolutional　Neural　Networks　(CNN).　Results　show　that　the　MLP　model　demonstrates　the　best　performance,　with　an　accuracy　of　0.903,　and　our　system　can　effectively　identify　code　quality　issues　and　improve　the　efficiency　of　educational　feedback.　©　2024　IEEE.