Research On The Design Of The General Bank Of Choice Question Based On QTI

Along with the development of information technology, the traditional education field has also undergone a significant change. Computer Aided Instruction (CAI) is becoming one of the hot spot of social public concern. Item bank system is an important part of CAI. In the designing and development process of item bank system, developers are always faced with two key problems to be solved:How to share resources among different applications? How to batch import test resources quickly and efficiently for establishing a based database of item bank system.In order to solve the problem of resource circulation among different systems, details of widely used international IMS QTI (Question Test and Interoperability) protocol. There are aimed at providing a unified format model for different systems and users and make it possible with conten integration. Through the operations, such as classification, sort the resource file into XML file which bases on the QTI and turns into a vehicle for circulation.Aiming at the problem of batch importing, there are two traditional ways:import items one by one and batch import. The first way requires substantial human and material resources, but the efficiency is not high. Another way needs to sort the resource file into a particular format document and then reads the document by batch process. This way shortrns the time for importing, but requires much time for sorting.To improve the practicability of item bank system, use a new method with words segmentation machine and naive Bayesian classifier: uses the segmentation algorithm to separate the text into phrases, reads pending words to extract the attributes, classify the pending words according to the sequence of attributes, tags the keywords which are identified, finally transforms the text into XML file based on QTI and then import into database after the parse. The design introduces the machine learning procedure, changes the traditional way and specifies the basic direction for future research.