Based On The Attribute On The Method Of Automatic Generation Of Research

With the development of the computer technology and artificial intelligence, the research on the system of autogenerating test paper has gradually aroused great concern of more and more experts. Different from just freeing teachers from trouble of manually generating test paper, now focus has been changed to the research of generating algorithm and how to improve the quality of the test item and the efficiency of test paper generating system.On the basis of analyzing a large quantities of references home and abroad, the thesis has made a survey on the auto generating test paper algorithm. After studying the foundational methods and theories in Attribute Theory, we used Qualitative Mapping(QM) and Conversion Degree Function(CDF) as auto generating test paper algorithm, and developed an auto generating test paper system based on Attribute Theory.On the basis of the system constructed in this article, the process of how to generate a series of qualified test papers was subscribed in detail from three steps of acquisition of user demands, design of test paper model and realization of test paper model. First, according to mathematic relationships of some index between test items and demand of test paper, user demands were quantified to a group of discrete indexes' distributive columns, namely the model of present test paper. Secondly, based on present test paper model, a mathematics model was established. At last, we conversed the process of test paper generating into multiobjective combinatorial optimization and the selection of test items using qualitative mapping in the test database, which have been prescribed in the result of optimization, to generate test paper two separeted process. In the process of generating test paper, every test item was expressed by a n-dimensional vector, we could find the ideal test item or all test items in its neighborhood according to the qualitative mapping, then separately counted the satisfactory degree of test item. By considering different attribute had different effect on satisfactory degree, weight was introduced too. we can count the satisfactory degree of every test item by using conversional degree function, then pick out all the best test items to combinea test paper.In order to verify the validity of auto generating test paper algorithm based on Attribute Theory, we made a lot of experiments on auto generating test paper, the result of experiments shows that the method is reasonable and effective. Because of the importance of algorithm on intelligence teaching system theory research and fact applying, our research is necessary, it is efficient to overcome the existing problems and has considerable theoretical and practical meaning, and it is also the basis of building a perfect system about Intelligence Computer Assisted Instruction.Shao Hongqing (Computer Science) Directed by Prof. Feng Jiali...