【作者】 李军；

【导师】 葛卫民；

【作者基本信息】 天津大学 ， 计算机应用技术， 2008， 硕士

【摘要】 随着Internet的快速发展和广泛应用,开发基于WEB技术的智能组卷系统已成为计算机辅助教学的一种重要手段。智能组卷算法是设计和实现一个智能组卷系统的关键。传统的组卷算法如随机法、回溯法等,在组卷成功率和效率方面不能满足实际需要,而遗传算法具有自适应全局寻优、智能搜索并且收敛性好等特性,能够更好地适应智能组卷的需求。本文重点研究了遗传算法在智能组卷中的应用问题。本文首先对智能组卷问题及其数学模型进行了描述,对智能组卷的主要方法进行了分析与评价,进而给出了利用遗传算法进行智能组卷的原理和方法。其次,针对传统的遗传算法应用于组卷系统时存在着容易过早收敛,在进化后期搜索效率较低等问题,本文在简单遗传算法的基础上,通过对遗传算法组卷的关键技术如组卷流程、遗传算子、种群规模和迭代中止条件等进行优化设计,提出了一种实用的智能组卷方法,最后,利用ASP.NET开发了一个基于上述组卷算法的智能组卷系统。本文对该系统的总体结构、数据库设计、各个功能模块及主要算法的设计和实现过程做了详细描述,并给出了组卷应用实例。本文研究和实现的智能组卷系统,能够对不同课程、知识点、题型的试题进行管理,可以按照不同的组卷要求利用遗传算法进行自动组卷。在组卷成功率和效率上能满足实际需要。更多还原

【Abstract】 With the rapid development and wide application of the Internet, the intellectual test paper composition system base on the WEB technology has become an important method in assisting instruction. Intellectual test paper composition algorithm is the key in the design and realization of intellectual test paper composition system. Traditional test paper composition algorithm like random method and backtracking, etc. can’t satisfy the real need in the aspects of success rate and efficiency of the test paper. However, the genetic algorithm has the features of adaptive global optimization, intellectual search and nice convergence, etc, which can meet the requirement of intellectual paper test composition better. The thesis mainly made a research in the application of generic algorithm in intellectual paper test composition.First off all, the thesis makes a description of the intellectual paper test composition and the mathematic model in the first place, analyzes and evaluates the major methods of intellectual paper test composition, then put forward the theory and method of intellectual paper test composition through the means of generic algorithm. Secondly, targets at the problems of application of traditional generic algorithm in test paper composition system like early convergence and low searching efficiency in the evolution anaphase, the thesis puts forward a practical intellectual test paper composition through optimizing design of the key technology like test paper composition process, genetic operators, population size and iterative suspension condition, etc. Lastly, the design and realization process of the general structure, database design, each functional module and the major algorithms of the system is described in great detail, and some concrete examples of paper test composition are also given in the thesis.Intellectual paper test composition researched and realized in the thesis can administrate test questions of different classes, knowledge points and question types; and can automatically make paper test composition by generic algorithm according to different requirements of test paper composition. The method can satisfy the realistic needs in the successful rate and efficiency of test paper composition.更多还原