An estimating algorithm for English levels of student and difficulties of test question

In view of the problem that currently the students choose online test questions blindly, the establishment of the item recommendation system is necessary. According to the student´s level, an estimating algorithm is used to sort items and recommend the question which is in the front of the sort to the student. According to the study of student over a certain period of time and all the answers to the questions of a certain passage that students have given, a multiple attribute decision model based on positive and negative ideal point is established. The model can adjust the level of the student and the difficulty of test question adaptively and ensure that the system will recommend the most suitable test questions.