现代信息科技2026，Vol.10Issue(4)：24-31,8.DOI:10.19850/j.cnki.2096-4706.2026.04.005
基于CNN-BiLSTM-ATT混合模型的高校高考录取分数预测研究
College Entrance Examination Admission Score Prediction in Colleges and Universities Based on CNN-BiLSTM-ATT Hybrid Model
摘要
Abstract
The prediction of college entrance examination admission scores in Colleges and Universities is of great significance to candidates,parents,and educational institutions.However,this prediction work is challenging due to the influence of multiple factors such as examination difficulty,enrollment strategies of Colleges and Universities,and the scale of candidates.Therefore,this paper proposes a hybrid model based on CNN-BiLSTM-ATT to predict the admission scores of Colleges and Universities.The model first uses Convolutional Neural Network(CNN)to extract the local features of admission scores in Colleges and Universities,then learns the long-term dependency in the time series through Bidirectional Long-Short Term Memory(BiLSTM),and finally introduces the Attention Mechanism(ATT)to enhance the focus on key year data to improve prediction performance.The experimental results show that the CNN-BiLSTM-ATT model has high accuracy and generalization ability in predicting the admission scores of Colleges and Universities.Compared with other models,it more effectively captures the changing trend of admission scores and achieves better evaluation indices,providing a valuable reference for college entrance examination volunteer application.关键词
CNN-BiLSTM-ATT/高考录取分数/预测/神经网络Key words
CNN-BiLSTM-ATT/college entrance examination admission score/prediction/Neural Network分类
信息技术与安全科学引用本文复制引用
马沅号,王红梅,刘浩强,陈建辉,刘星宇..基于CNN-BiLSTM-ATT混合模型的高校高考录取分数预测研究[J].现代信息科技,2026,10(4):24-31,8.基金项目
河南省高等教育教学改革研究与实践项目(研究生教育类)(2023SJGLX325Y,2023SJGLX019Y) （研究生教育类）
河南省高等教育教学改革研究与实践重点项目(2024SJGLX0149) （2024SJGLX0149）
郑州航空工业管理学院研究生教育创新计划基金项目(2024CX77) （2024CX77）
