近年來，隨著網路學習的蓬勃發展，學習者參與教學活動所產生的學習歷程檔案，除可使學習者檢視自身學習歷程外，也提供教師於評量學生學習成效時一個重要的參考依據。在網路學習的環境中，經由資訊技術的導入，已可實現系統自動記錄學習者學習歷程的目標，包含學習時間、登入次數、文章點閱數及小組議題報告等資料皆可完整保留。故能累積為數眾多的歷程檔案，可供教師瞭解學生的學習狀況，並進一步作為評量學習成效之用途。本研究旨在發展一網路學習小組議題報告的電腦化評量機制，以解決傳統教師人工評量可能耗時費力或評量標準未能一致的問題。研究者參考網路學習小組議題報告的評量規準，並透過中文潛在語意分析與決策樹演算法建置電腦化評量系統。待系統發展完成，即經由比較電腦化評量與人工評量在評量時間耗費及評量得分結果的差異，以及邀請領域專家進行系統評估，作為驗證電腦化評量系統之方法。研究結果顯示：一、電腦化評量的評量時間顯著低於人工評量；二、電腦化評量與人工評量的評量結果無顯著差異；三、電腦化評量系統的「系統介面親和度」、「系統功能有用性」、「系統整體滿意度」皆獲領域專家一致的肯定。

The purpose of this study is to develop a Computerized Assessment System for a Report of Web-based Learning. The researchers reviewed the evaluation criterion of a Web-based Learning report, used the Latent Semantic Analysis (LSA) and Decision Trees to build the Computerized Assessment System. As the system was developed, the researchers compared the rating results in the time spent and the grades between the Computerized Assessment System and human assessment. Then they proceeded with the questionnaire assessment to get relevant data. The research results are summarized as follows : 1.The Computerized Assessment saves more time than the human assessment, 2.There is no significant difference in rating results between the Computerized Assessment and the human assessment, 3. "User friendliness", "System Usability" and "Use Satisfaction" for the Computerized Assessment System are highly approved by the experts.