支援群組查詢之適應性資訊重組

Abstract

本論文提出群組查詢的新穎概念，並發展結構化資訊檢索方法，進而根據使用者對查詢結果的回饋，適應性調整查詢以符合個人的需求。群組查詢兼顧整體資料組合的查詢相似度與個別資料間的多樣差異性；結構化資訊檢索方法針對大量資料的重組與計算提供了快速的存取方式；而適應性調整查詢為群組查詢提高了應用上的彈性。『適應性資訊重組』模型整合了上述各項方法與特色，提供了一個支援群組查詢的一般化模型。在本研究為驗證模型可行性所設計的實驗中，整合了教師挑選題目以及學生接受測驗的兩個應用，並分析學生答題的反應來估計出該生對於各單元的精熟程度，亦是本論文在電腦輔助測驗領域中創新的想法與貢獻。

In this thesis we propose a group-based query concept, develop a structured information retrieval method, and adjust queries by user's feedback. The concept of group-based queries takes both query similarity and data diversity into account. The structured information retrieval method provides a fast solution for the access of huge amount of data. The adaptive adjustment based on previous queries expands the applied flexibility of group-based queries. Adaptive Information Reorganization, a general group-based query model we propose in this thesis, combines all above concepts. In this study we design a set of experiments to evaluate the proposed model by integrating the selection of problems by teachers and the problem solving process of students. Furthermore, we enable the model to measure individual student's degree of understanding on different course unit, which we believe is an innovative concept in the area of computer-assisted testing.