A personalized topic recommendation method and system using a machine learning model

Abstract

The invention discloses a personalized topic recommendation method and system using a machine learning model. The method includes: collecting relevant topics of students, establishing a question bank, and using a machine learning model to perform multi-objective prediction on each topic in the question bank, Score the questions in the question bank according to the estimated goal and the contribution value of each question to the student's total score, select the first number of questions according to the scoring, establish the first set of questions, and evaluate all the questions under each estimated goal The topics in the first topic set are sorted, the corresponding rankings are converted into scores for fusion sorting, and the second number of topics are selected and recommended to students according to the set topic selection factors in the fusion sorting results; the present invention can provide students with Personalized questions can maximize the efficiency of doing questions and provide strong support for improving performance.

Description

Technical field[0001]The present invention relates to the field of Internet education, especially in the recommendation method and system of personalized topics using machine learning models.Background technique[0002]In order to improve the results of the test, it is often used "Questional Sea Tactics"; students can improve their level of knowledge and topics in the knowledge point through multi-top-top. However, there are quite a few topics, the students are actually unnecessary. For example, students have very skilled the topic of the triangle function, then save the number of questions of this type of topic, saving time to do other more unfamiliar topics.[0003]In the prior art, optimization of the topic of students is often based on the current total knowledge level of students, or based on knowledge points to the students' recommendations, such practices require teachers to be recommended. Topic There is definition, and it is only possible to distinguish several types of student...

AI Technical Summary

Problems solved by technology

[0003] In the existing technology, the optimization of the questions for students is often based on the current general knowledge level of the students, or recommending questions to students based on knowledge points. There are definitions, and only several types of students can be artificially distinguished, and it is impossible to screen questions according to each student's own situation, that is, it is impossible to truly personalize the most suitable question bank for students

Images