山西大学学报（自然科学版）2025，Vol.48Issue(1)：1-19,19.DOI:10.13451/j.sxu.ns.2024151
自适应个性化巩固学习模型
Adaptive Personalized Consolidated Learning Model
摘要
Abstract
"Consolidated Learning Recommendations"refers to the process of recommending learning content that students need to review and consolidate.This paper investigates the issue of personalized consolidated learning recommendations based on compe-tence-based knowledge space theory,and proposes an effective recommendation method to address the challenge of the lack of one-to-one correspondence between competence states and knowledge states.The method firstly calculates the inner master fringe based on the student's knowledge state.It uses this to deduce the knowledge states before and after potential regression.Based on these knowledge states,it identifies the tops or bottoms of the corresponding competence states.Finally,according to these tops or bot-toms,it recommends the skill set that needs to be consolidated.This approach helps prevent knowledge regression due to factors like forgetting.This paper presents two characterization theorems:the first uses a skill function to characterize the tops or bottoms of competence states,and the second uses a problem function to characterize the inner master fringe of knowledge states.By applying these theorems,the tops or bottoms of competence states and the inner master fringe of knowledge states can be directly obtained without constructing a knowledge structure.Finally,this paper presents algorithms to obtain the inner master fringe based on defini-tions and characterization theorems,and demonstrates through comparative practice that the latter reduces time consumption by an average of 77％ and memory usage by an average of 67％.关键词
知识状态/能力状态/内掌握边缘/技能函数/问题函数/个性化学习Key words
knowledge state/competence state/inner master fringe/skill function/problem function/personalized learning分类
信息技术与安全科学引用本文复制引用
王功勋,李进金..自适应个性化巩固学习模型[J].山西大学学报（自然科学版）,2025,48(1):1-19,19.基金项目
国家自然科学基金(12271191) （12271191）
