Automatic Generation Of Learning Questions

Intelligent Tutoring System（ITS）is a kind of intelligent and user-friendly computer software,which is widely used in the education industry.It makes teaching not limited to traditional teaching mode.ITS has been proven to be of great help to students’ learning.The automatic generation of questions has always been a research hotspot of ITS.Currently,multiple-choice questions,including multiple-choice questions,continue to be the predominant question format in exam test questions.Therefore,the automatic generation of multiple-choice questions is a key point in the field of automatic generation of learning questions.The automatic generation of multiple-choice questions is based on text,ontology or semantic network,and the research goal is to generate distractors with high similarity to the correct options.Therefore,calculating the semantic similarity between the options has become the core problem in the generation of multiple-choice questions,and it is also a difficult point.This thesis is a study of the problem generation method based on semantic network.Firstly,the semantic link is expanded,and the semantic network of some junior high school mathematics and physics courses is constructed on the basis of optimizing the semantic link;then,Dennai＆Benslimane algorithm,the semantic similarity algorithm,is improved,and its penalty factor is weakened,and added a new influence factor link topology weight,and used the improved algorithm for the semantic similarity calculation in multiple-choice questions;finally,the multiple-choice difficulty control module was programmed and implemented automatically for multiple-choice questions under difficulty control.Generated code tests.The author supplements semantic links,enriches the semantic network in Krlab,and makes knowledge representation more reasonable.The semantic similarity algorithm is improved to make the confusion value of the distractor and the difficulty of the question more accurate and reasonable.It realizes the automatic generation of multiple-choice questions under the control of difficulty,and provides support for learners to choose the matching difficulty level.Finally,the automatic question generation module is developed,which contributes to the automatic question generation of ITS.