基于相关算法的分布式自适应习题库设计与实现

Most existing online testing systems only have little ways to generate a test paper, so they can not satisfy the requirements of the complex purpose of taking the test. In this paper, after the design objective and functional model are given, the design and implementation of the testing system are introduced in details, including the design of database, the design and implementation of the main functional models,etc. The basic requirement of the online testing system is to generate reasonable and scientific test paper, whose quality is decided by the algorithm of generating test paper. Through the analysis of exist algorithm of generating test paper, an algorithms based on improved Relevance Feedback algorithm of Vector Space model is designed, which is feasible and effective in the test results. The online testing system that this thesis designed have many advantages of generality, such as good scalability. It also can satisfy the requirements of the restrictive condition of generating the test paper. The system will contribute the school to held the separation of teaching and examination in the examination reform.