An Information Retrieval and Analysis System for Historical and Literary Documents in Chinese

數位人文是近年來數位科技應用的一種重要趨勢，是一個結合數位技術與人文研究的新興領域，透過資訊科學的技術並結合大量數位化的文史資料，輔助文史學者進行較深入的人文研究，彌補並獲得在數位化之前無法用人工逐一比對所觀察到的現象或新的詮釋。 本研究以「中國近現代思想及文學史專業數據庫」的大量文史資料作為基礎，希望建構出一個能協助文史學者對中國近現代史的歷史觀念或現象較有效率的進行分析之檢索系統。本系統讓研究者可以透過具代表性的檢索詞彙與檢索條件搜尋出與欲研究的議題相關的文獻集合，再利用PAT Tree(Patricia tree)技術與自身的專業知識，擷取出研究議題底下的專業關鍵詞彙，並針對有興趣的關鍵詞彙或關鍵詞彙組合進行相關的時序分析，提供研究者以不同的面向觀察並詮釋出新的趨勢和結論。 我們找了一些具有文史工作相關背景經驗的使用者來實際使用本系統，並進行系統滿意度的評估，從評估的結果得知大部份的使用者都給了不錯的滿意度回饋，顯示本系統對於研究者在進行相關的文史研究上，可以達到一定程度的助益。In recent years, digital humanities have become an important trend of the application of information technologies. It’s an emerging field by combining information technologies with humanities studies. Historians can make intensive humanities studies through the combination of information science technologies and numerous digitized historical data, finding the new historical phenomenon that cannot be obtained by manual handling the historical data. In this paper, we construct an information retrieval and analysis system based on the text material available at the Database for the Study of Modern Chinese Thought and Literature that can help historians to make historical studies more efficiently. The system can let historians find the key documents which are related to the historical studies they make by representative search word. Besides, the system can use the PAT Tree (Patricia tree) method to extract useful keywords, with the help of historians to finalize the keyword selection. The system can also do the time analysis of one keyword or two keywords and help historians make new conclusions of historical studies. We let some people who have historical work experience use the system and let them evaluate the system. Most of the evaluation results are good. It means that the system can help historians to make historical studies more efficiently