Preposed knowledge point labeling method and device, question pushing method and device, equipment and medium

Abstract

The invention provides a preposed knowledge point labeling method and device, a question pushing method and device, equipment and a medium. The preposed-knowledge point labeling method comprises the following steps: inputting acquired answer data of a plurality of student users to each knowledge point into a Bayesian network model; performing structure learning on the Bayesian network model basedon the answer data to obtain a second knowledge point system structure, wherein the second knowledge point system structure is marked with a relationship between at least one knowledge point and the corresponding front knowledge point. In this way, marking of the front knowledge points of all the knowledge points can be automatically completed, on one hand, the time cost and the auditing cost of manual marking of teachers can be reduced, and on the other hand, compared with a method for obtaining the preposed knowledge points by performing similarity calculation based on text information of the knowledge points in related technologies, the technical scheme has the advantages that possible preposed relationships among the knowledge points are reversely deduced by utilizing the answer data of the students, so that the wrong labeling conditions of similar texts and no actual direct relationship are avoided, and the labeling accuracy is improved.

Description

technical field [0001] The present application relates to the technical field of data processing, and in particular to a method, device, equipment, and medium for labeling pre-knowledge points and pushing topics. Background technique [0002] Knowledge points are the basic units in subject learning, such as the judgment of congruent triangles, the properties of congruent triangles, etc., are all knowledge points. When students study knowledge points, they often study in the order of chapters in the book. However, there are actually intricate relationships between knowledge points. Specifically, some knowledge points are the basis for learning other knowledge points. These knowledge points that are the basis for learning other knowledge points are called pre-knowledge points. However, many knowledge points and their corresponding pre-knowledge points often have different textbooks, different grades, and even different school stages. If we can build the internal relationship...

AI Technical Summary

Problems solved by technology

[0003] In related technologies, teachers with relevant and rich teaching backgrounds are invited to manually mark the relationship between knowledge points. However, there may be hundreds or even more knowledge points in a certain subject, and there may be a gap between every two knowledge points. There is a relationship, which means that the teacher needs to spend a lot of time to judge whether there is a relationship between these knowledge points. Due to the large number, it is more likely to have wrong labeling and missing labeling while consuming time; another In a related technology, automatic labeling is realized by using the similarity of the text to identify knowledge points, but it only considers the similarity between texts, which does not mean that they must be related

For example, "Historical Significance of the Revolution of 1911" and "Historical Significance of the French Revolution" in the subject of history, these two knowledge points have a common text "Historical Significance of Revolution", so no matter what method is used to calculate the similarity, they will get similar High degree of conclusion, but there is no direct connection between these two knowledge points

Therefore, the error caused by using this method to automatically label the relationship between knowledge points is relatively large.

Images