Method for achieving user ability prediction through man-machine interaction

Abstract

The invention relates to the technical field of machine learning, and discloses a method for achieving user ability prediction through man-machine interaction. According to the method, participation of any human teacher is not needed in other steps except data preparation and parameter setting in the first step, and therefore occupation to the manpower resource and particularly the education resource is reduced maximally; meanwhile, due to the fact that the method is achieved on the basis of the question step structure and the single step-level doing right accuracy rate, the performance abilities when a user answers various parts of a question can be more precisely estimated, and the purpose of improving the prediction precision by introducing step-level information of the question is achieved; and in addition, in the man-machine interaction process, the step-level right and wrong information of the user can be obtained on the condition that the teaching process is not influenced by means of the ingenious interaction design.

Description

technical field [0001] The invention belongs to the technical field of machine learning, and in particular relates to a method that can be applied to the education industry and complete user ability prediction through human-computer interaction. Background technique [0002] In the education industry, accurate prediction of students' performance on the topic has great reference significance for topic recommendation, explanation and practice, and even the formulation of the entire learning plan. [0003] In traditional teaching, this process is often carried out by teachers. However, first of all, teachers' cognition of students is very perceptual and cannot be quantified. They can only simply identify "easy to make mistakes" or "not easy to make mistakes", and cannot give an accurate quantitative value. Naturally, it is impossible to compare steps and questions A comparison between them provides enough information. At the same time, human memory also determines that teache...

AI Technical Summary

Problems solved by technology

However, since only the final options of the students can be obtained, this method can only treat all the abilities associated with a topic as a whole, ignoring the fact that students often have uneven abilities

When students are doing a question, the problem often appears at a certain point of the question, rather than all the abilities involved in the question, so for students who can complete some steps, the system often gives a lower evaluation for all the abilities involved , thus losing its precision

Images