期刊论文

Study of Recommending Forum Respondents Based on Learner's Help-seeking Behavior

中文

小型微型计算机系统

1000-1220

2019

40

3

493-498

10.3969/j.issn.1000-1220.2019.03.006

国家社会科学基金一般项目（17BTQ061）资助；

本校为第一机构

在在线学习环境中,依据学习者的行为为其推荐合适的问题回答者,可有效提高其学习效果.目前许多的问题回答者推荐研究考虑到了问答数据,但没有考虑行为因素对推荐问题回答者的影响.本文结合问答数据与学生行为数据,提出了一种论坛问题回答者的推荐方法.该方法的基本思路是:获取学习者求助行为类别;根据学习者求助行为类别为其推荐论坛问题回答者.为此,采用聚类算法处理学习者的求助行为数据并得到该学习者求助行为类别标签;采用此学习者求助行为数据和求助行为类别标签作为训练数据,训练朴素贝叶斯模型,从而自动识别新的学习者求助行为的类别;在此基础上,提出使用卷积神经网络方法构建论坛回答者...

In the online learning environment,recommending appropriate question answerers based on the learner's behavior can effectively improve their learning effectiveness. Many current recommendation studies which aimed at choosing question answerers are based on question and answer data without consider the influence of behavioral factors on answering questions. This article combines question and answer data with student behavior data and proposes a recommendation method for forum question respondents. The basic idea of this method is to obtain the learner's help-seeking behavior category and recomm...