基于表情和行为特征的学习风格预测

发布时间：2018-03-18 06:52

本文选题：学习风格　切入点：行为特征　出处：《辽宁师范大学》2014年硕士论文　论文类型：学位论文

【摘要】：学习风格是表现学习者个性差异的主要因素之一，学习者的学习风格对提高数字化学习的效率是重要的。学习风格的研究起源于西方心理学领域，随后学者们从认知、人格、互动等多个方面探讨这一课题，并为其建立了多种理论模型。随着研究的深入，研究者开始整合已形成的纷繁复杂的学习风格模型，并探求统一于模型背后的东西。其中学习风格的测量工具是链接教育学习实践与学习风格理论的桥梁，服务于各个模型理论，最终服务于广大教学者和学习者。在对已有的研究成果进行考察时发现常见测量学习风格的方法包括问卷法、测验法等。这些传统的预测学习风格的方法只能通过学习者填写的信息进行预测，主观判断强烈，获得的变量常常受到共同方法误差的污染，导致预测结果不准确。学习风格量表的测试题目形式单一内容枯燥，会使学习者厌倦而随意填写，甚至会产生与实际完全相反的结果。针对显性预测方法中被测者主观因素严重影响预测结果的问题，本文首先详细阐述和辨析了学习风格的定义，再分析表情行为特征与学习风格之间的关系，介绍了各种学习风格模型，并分析各种学习风格类型的学习者的特征表现，最后提出了基于表情和行为特征的学习风格显性预测方法，采用多媒体形式展示测试材料，通过获取学习者处理测试材料时的表情和行为特征对其学习风格进行预测。生动的测试材料展现，，实时的表情获取，以及动态的答案选择和反应时间分析，可以更多地获取与学习风格有关的客观特征信息，从而使结果更准确。
[Abstract]:Learning style is one of the main factors to express learners' personality difference, and learners' learning style is important to improve the efficiency of digital learning. The study of learning style originated in the field of western psychology. Interaction and other aspects of this subject have been explored, and various theoretical models have been established for it. With the deepening of the research, researchers have begun to integrate the various and complex learning style models that have been formed. The measurement tool of learning style is the bridge between learning practice and learning style theory, which serves each model theory, and finally serves the majority of teachers and learners. In the investigation of the existing research results, we found that the common methods of measuring learning style include questionnaire method, test method and so on. These traditional methods of predicting learning style can only be predicted by the information filled in by the learner, and the subjective judgment is strong. The variables obtained are often contaminated by common method errors, resulting in inaccurate prediction results. It can even produce the exact opposite of the actual result. In order to solve the problem that the subjective factors of the subjects in the dominant prediction method seriously affect the prediction results, this paper firstly elaborates and discriminates the definition of learning style, and then analyzes the relationship between expression behavior and learning style. This paper introduces various learning style models, and analyzes the characteristics of learners with different learning styles. Finally, an explicit prediction method of learning style based on facial expression and behavioral features is proposed, and the test materials are displayed in multimedia form. The learning style is predicted by acquiring the expression and behavior characteristics of the learners when they deal with the test materials. Vivid test material presentation, real-time expression acquisition, dynamic answer selection and response time analysis. More objective feature information about learning style can be obtained to make the result more accurate.
【学位授予单位】：辽宁师范大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：B842.3

【参考文献】