基于NET自适应考试系统的研究与设计-计算机技术专业论文.docxVIP

摘要 考试是教学活动中最重要的环节之一，随着计算机和网络技术日益发展成熟， 各种类型的网络考试系统不断涌现出来。目前大部分的考试系统都是基于传统的 经典测量理论（CTT），其存在的问题是不能客观、准确地测试出考生的知识和能 力水平。 本文介绍了经典测量理论基础及其局限性，详细阐述了项目反应理论基本概 念、IRT 三种模型及其特征曲线、信息函数、能力参数估计方法。在此基础上结合 计算机网络技术的发展提出了计算机自适应考试(CAT)模式。本系统在 Windows XP 下，选择 ASP.NET(C#)、SQL Server2005 开发在线考试系统，采用 (B/S)体系 结构有更好的安全性，用户端不需要安装任何应用程序。使用 UML 分析了系统的 体系结构，介绍了系统各功能模块的设计，并改进了按区分度的选题策略，有效 地控制试题曝光率。同时，根据系统运行中取得的测试数据进行了统计分析。 本系统能够根据不同能力被试者随机选择试题项目，减少了测试长度，与传 统在线考试系统相比提高了考试效率。最后对自适应考试系统进行了总结，并且 进一步展望了在线自适应考试系统今后的发展研究方向。 关键词：自适应考试 项目反应理论 选题策略 Abstract Exam is one of the most important links in teaching activity, now, with the increasing maturity of the development of computer technology and network technology, many kinds of network examination system are emerging. Most of the test systems are based on the classical test theory (CTT) currently. Therefore, some problems exist such as testers knowledge and ability levels cannot be tested objectively and accurately and so on. This thesis introduces us the basic knowledge and limitations of classical test theory, and also it describes the basic concept of item response theory, three models of IRT and their characteristic curves, the information function and the ability of parameter estimation in detail. On this basis, the Computerized Adaptive Testing (CAT) model is put forward with the combination of the rapid development of computer network technology. The system is in Windows XP, choosing ASP.NET(C#)、SQL Server2005 for developing online examination system. It is safer because of the use of B/S structure, so clients don’t have to install any user applications. The UML is used to analyze the system structure and introduce the design of each function module in it. The selection strategy by differentiation is improved and the rate of item exposure is effectively controlled. Meanwhile, it carries out a statistical analysis of test data obtained from the system operation. The system can randomly select the item according to the different ability