| Education is the foundation of a nation.With the rapid development of the artificial intelligence industry,in order to promote education reform and innovation,the deep integration of artificial intelligence and education is imperative.In contemporary education,the problem of "excellent student's assignments are not enough,but poor students cannot finish" has been criticized by people.This problem can only be effectively solved by selective and precise exercises in a large number of exercises.However,most of the course knowledge contained in the exercises is unclear,which makes it impossible for people to quickly and effectively screen the exercises.In the field of natural language processing,text classification,as a commonly used information processing technology,is often used to solve the problem of text clutter.With the rise of deep learning,text classification model based on neural network has made significant progress in many fields of text classification.This brings new opportunities for classifying exercises by course knowledge points.In order to solve the problem of automatic classification of large-scale exercises according to course knowledge points,this paper researches and completes data processing processes such as data preprocessing,text data enhancement,and text representation of exercise.At the same time,this paper makes in-depth research on deep learning text classification models.Based on this,this paper combined the bidirectional LSTM and attention mechanism to design and implement an exercise classification model.A comparative experiment was performed to verify its effective modeling ability for exercise classification tasks.Finally,this paper applies the exercise classification model to design and implement an automatic exercise classification system.To a certain extent,it solves the problem of classifying exercises by course knowledge points for users,and also provides data support for personalized exercise recommendation. |
PDF Full Text Request