In this paper, we present an efficient algorithm for integrating a collection of heterogeneous itembanks, called Heterogeneous Itembanks Integrator (HIBI). This algorithm is motivated by the desire to integrate itembanks provided by both the publishers and instructors, which more often than not use different content structures. By integrating all the itembanks, HIBI provides for the users of an e-Learning system a central repository of all the items with a consistent content structure that would significantly increase the learning contents available to the users of an e-Learning system. The proposed algorithm starts off with one of the itembanks as a reference itembank. Then, all the items on all the other itembanks are integrated into the reference itembank to create the so-called meta-itembank. Moreover, by treating the meta-itembank thus created as the reference itembank, it can be easily extended by using exactly the same algorithm. Our experimental results showed that the proposed algorithm could provide an extremely high quality result in terms of both the relevance of items and the computation time in Chinese item sets. In addition, the experimental results also showed that the accuracy rate of the meta-itembank described herein ranges from 96.3% to 97.6%.
|頁（從 - 到）||4319-4334|
|期刊||International Journal of Innovative Computing, Information and Control|
|出版狀態||Published - 2010 十月 1|
All Science Journal Classification (ASJC) codes