Diversified exercise recommendation method, system and equipment combining difficulty and weak knowledge points
The invention discloses a diversified exercise recommendation method, system and device combining difficulty and weak knowledge points, and the method comprises the steps: firstly coding a historical answer sequence of a student, and obtaining model input data in a standard format; modeling exercise...
Saved in:
Main Authors | , , , , , |
---|---|
Format | Patent |
Language | Chinese English |
Published |
30.05.2023
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | The invention discloses a diversified exercise recommendation method, system and device combining difficulty and weak knowledge points, and the method comprises the steps: firstly coding a historical answer sequence of a student, and obtaining model input data in a standard format; modeling exercises by using an e-DKT model to obtain the correct answer probability of the exercises of the students, and modeling knowledge points by using a k-DKT model to obtain the mastering degree of the knowledge points of the students; on the basis of the correct answer probability and the knowledge point mastering degree, the exercise individual difficulty and student weak knowledge points are calculated, and a filtering module is utilized to select exercises which are moderate in difficulty and contain the weak knowledge points from the exercise set to construct candidate subsets; selecting exercises with the maximum spatial distance from the candidate subsets by using a simulated annealing algorithm, and generating a reco |
---|---|
Bibliography: | Application Number: CN202310207879 |