在學術研究過程中，查詢其他學者、專家的研究成果與心得是一項不可或缺的研究活動。但是仍然有許多不同語言的論文因為翻譯上的差異，而無法在論文資料庫上作精確的查詢。在本篇論文中，我們提出以包含中英文關鍵字的雙語同義關鍵字集合為基礎之論文分群研究。本系統會先產生中英文同義關鍵字集合，再以此同義關鍵字集合資料庫對論文作自動分群處理。此方法不僅解決使用現有數位化辭典無法收錄新的關鍵字的問題，亦可解決大多數知識管理與資料探勘系統中同義關鍵字的問題。此外，我們也使用叢集系統來解決大部份文件分群系統的效能問題。透過多部電腦的計算處理，此系統不僅可以節省大量的時間，亦可達到高可用性與負載平衡。經由我們的實作結果，此方法可獲得更有效的分群結果。 Searching other published paper is a required activity for a researching progress. But there are still many various languages presentation articles, that make the precisely query hard to achievement. In this thesis, we propose an automatic theses clustering method based on the bilingual synonymous keyword sets which include Chinese and English keywords. First, the system generates synonymous Chinese and English keyword sets, and then bases on synonymous Chinese and English keyword sets clustering the theses automatically. The method not only solves the weakness of using digital dictionaries to do new keywords clustering problem, but also solves the synonymous keywords problem in most knowledge management or data mining method. And we also used cluster technology to solve traditional documents clustering performance problem. Through many computers’ process, the system not only can save a lot of time, but also attain to high availability and loading balance. The primary experiments prove that the system makes the theses clustering work effectively.
