期刊论文

ARPDF:Achievement Rank Prediction of Learner Based on Dialogue Flow

中文

小型微型计算机系统

1000-1220

2019

40

2

267-274

10.3969/j.issn.1000-1220.2019.02.006

国家社会科学基金一般项目（17BTQ061）资助；

本校为第一机构

对话流所隐含的信息包括了学习者对所学课程内容的掌握程度和关注点,分析这些对话流对预测学习者的成绩,以支持教师提前对潜在成绩不良的学生进行及时干预有着重要意义.提出了一种基于对话流的学习者成绩等级预测算法ARPDF(Achievement Rank Prediction based on Dialogue Flow),首先采集对话流,通过对话流划分、对话状态矩阵生成实现了对该对话流的分析以获取到学习小组的对话状态矩阵;在此基础上,通过基于LSTM的预测模型获得学习小组学习者的成绩等级.在本文所提方法的基础上进行了实验,其结果表明了该算法是有效的.

The information contained in the dialogue flow includes the learners' mastery and focus of the content of the courses they have learned. It is of great significance to analyze these dialogue flows to predict the learner's performance in order to support teachers' timely intervention of potential bad grade students. This paper proposes a learners' achievement rank prediction algorithm based on dialogue flow ARPDF (Achievement Rank Prediction based on Dialogue Flow ). Firstly,the dialog flow is collected and through dialog flow partitioning and dialog state matrix generation,ARPDF finishes the a...