初中数学错误模型研究及其在自动判卷系统中的应用

发布时间：2018-07-05 20:37

本文选题：错误模型 + 知识库　；参考：《电子科技大学》2016年硕士论文

【摘要】：信息技术发展至今,已经涉及到我们工作生活的各方各面,在教育领域,传统的教育手段和模式很难满足人们对教育的需求,于是智慧教育、互联网教育等概念逐渐被提出。得益于人工智能技术的发展,很多专家学者开始探索将人工智能与教育相结合,以人工智能技术与互联网的思维相结合重塑教育模式、内容、方法、体系。教育资源的缺乏与分配不均,无疑是教育方式变革面临的首要问题。本文的主要研究方向是在无教师参与、无参考答案的情况下,由系统自动对学生的答题内容进行评判,包括每个解题步骤的正误判断以及在错误的情况下评判出具体的错误类型。旨在减轻教育人力成本的同时,找出学生学习中的短板,有的放矢,有效促进学习,为互联网教育的一个新探索。本文主要研究内容如下:1、初中数学常见错误类型的研究及认知建模对大量初中数学解题中常犯的错误进行收集和分类,并对产生这些错误的现象及原因等进行深入分析和研究,构建一套较为完整的错误类型体系,涵盖初中数学几乎全部概念的错误类型。分析和归纳每种错误类型的共性与特性,基于人的认知过程,自动构建错误认知模型。2、错误模型知识库的研究与构建运用自然语言处理(NLP,Natural Language Processing)技术,在对题意和学生答题内容进行语义理解和表示的基础上,以初中数学公式、性质、定理、推论以及数学常识为基准,建立推理的规则库。并基于Drools正向规则引擎与符号计算工具,以题干的已知条件做为推理的起点,构建错误模型知识库,该知识库尽可能多的覆盖学生所有正确的答题内容以及与知识推理路径关联的错误模型。3、错误模型库在自动判卷中的应用基于错误模型知识库,攻克因主观题一题多解、答题形式多样化等而造成机器自动阅卷困难这一技术障碍,构建一个智能判卷系统。系统首先利用知识库结合符号计算工具对学生的答题内容的每一步骤进行正误判断,然后根据错误模型对学生错误的解题步骤进行具体的错误类型判断,并指出学生错误的具体原因,反馈给学生,让学生知其然知其所以然,实现学生错误的精准定位和判卷。
[Abstract]:The development of information technology has been related to all aspects of our work and life. In the field of education, the traditional means and models of education are difficult to meet the needs of people, so the concepts of wisdom education and Internet education have been put forward gradually. Thanks to the development of artificial intelligence, many experts and scholars began to explore the combination of artificial intelligence and education, and reshape the educational model, content, method and system with the combination of artificial intelligence technology and Internet thinking. The lack and uneven distribution of educational resources is undoubtedly the most important problem facing the reform of educational mode. The main research direction of this paper is to judge the students' answer content automatically by the system without the participation of teachers and the reference answers. This includes the correct and wrong judgment of each problem solving step and the judgment of the specific error type in the case of error. In order to reduce the cost of educational manpower, to find out the short board of students' learning, to effectively promote learning, and to provide a new exploration for Internet education. The main contents of this paper are as follows: 1. The research on common error types and cognitive modeling in junior high school mathematics collects and classifies the common mistakes made in a large number of junior high school mathematics problems, and makes a deep analysis and research on the phenomena and causes of these errors. Build a complete system of error types, covering almost all the concepts of junior high school mathematics error types. The commonness and characteristics of each error type are analyzed and summarized. Based on human cognitive process, the error cognitive model. 2. The research and construction of error model knowledge base using NLP Natural language processing (NLP) technology. Based on the semantic understanding and representation of the meaning of the questions and the students' answers, the rule base of reasoning is established on the basis of mathematical formulas, properties, theorems, corollaries and mathematical common sense in junior high school. Based on the drools forward rule engine and symbolic computing tool, the knowledge base of error model is constructed based on the known condition of problem as the starting point of reasoning. The knowledge base covers as much as possible all the correct answers of the students and the error model. 3. The application of the error model base in automatic examination paper identification is based on the error model knowledge base. Because of the technical obstacle of machine automatic marking due to the variety of answer forms, an intelligent marking system is constructed. The system first uses the knowledge base combined with symbolic calculation tools to judge each step of the students' answer questions correctly and wrongly, and then, according to the error model, makes a specific error type judgment on the students' wrong problem-solving steps. It also points out the specific reasons of students' mistakes, gives feedback to students, lets students know what they are doing, and realizes the accurate positioning and marking of students' mistakes.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：G633.6;TP391.1

【相似文献】