A novel online test-sheet composition approach for web-based testing

In e-learning environment, Web-based testing system can help to evaluate learners learning status precisely. To meet the needs of multiple assessment criteria, a new test-sheet construction model was proposed. Based on the proposed model, a differential evolution algorithm with effective coding strategy was designed to generate high quality test-sheets for Web- based testing. In simulation experiments, the proposed algorithm was applied to a series of item banks with different scales. Superiority of the proposed algorithm is demonstrated by comparing it with the simple genetic algorithm and differential evolution algorithm with binary coding strategy.