Research And Application On The Adaptive Online Learning Testing

With the rapid development of the Internet technology,more and more learners choose to learn through the Internet.A variety of online learning models and methods based on intelligent and automated are in the ascendant.Online learning adaptive testing is one of the important aspects.The thesis focus on the theory and technology of adaptive online learning testing,and apply the latest achievements such as pedagogy and psychology to the study of online learning testing,and propose an adaptive strategy for learners.Based on the realization of the online learning adaptive testing system it improves the efficiency of learners' testing,and provides a new way to the personalized online learning ability testing.This thesis mainly carries out three aspects of work:Firstly,studying and designing item selection strategy of the adaptive online learning testing system.Analyzing features and limitations of the Maximum Fisher Information strategy,a-strategy and the improved algorithms.Putting a new reliable and feasible improved item selection strategy through a research on the classical adaptive testing item selection strategy.Comparing the improved item selection with the traditional item selection strategy from exposure rate of the item,average exposure rate,the accuracy and efficiency of testing and Chi-square test.Secondly,researching the testing methods of adaptive item selection strategy based on Monte Carlo simulation,simulating the algorithm proposed in this thesis.Designing structure of testing-method,coding testing-method program,using testing-method to simulate item selection strategy processing.Comparing the traditional item selection strategy and the improved strategy which is proposed in this thesis.Thirdly,researching and designing the adaptive online learning testing system based on the consideration of usability and reliability.Designing and implementing the system architecture.Realizing the function of each module of the adaptive testing system.Establishing the question bank of the adaptive testing system.Providing an effective way for the online learners.The new adaptive testing algorithm and model proposed in this thesis effectively reduce the item exposure rate of traditional strategy and improve the accuracy and the efficiency of the testing.The completed of the adaptive online learning testing system provides a reliable means for the differentiation of learners' individual learning ability,and it has a good application prospects and value.