Cross-language information retrieval

資訊檢索研究的目的在解決人類對於資訊的需求，發展至今不斷地消除一道道的資訊藩籬。隨著電腦網路的普及，網際網路快速地深入世界的每一個角落，普羅大眾對於「世界村」觀念感同身受的同時，語言的藩籬變得具體而殘酷，使用者很難檢索「不同文」的文獻資料。本文說明資訊檢索的語言藩籬，討論目前語言技術用於處理語言藩籬的可能方案，並且比較現有超越語言藩籬的資訊檢索系統與傳統資訊檢索系統之間的系統績效。

The purpose of research for information retrieval (IR) is to fulfill information need. IR has eradicated many information barriers from the very beginning. With the fast development of computer networks, the Internet has turned the world into a global village. But when the Internet users try to locate documents in different language, the once implicit language barrier becomes real and crucial. This paper attempts to explore the language barrier underlying the process of information retrieval, the possible solutions of this barrier, and a comparison of the performance of cross-language information retrieval and monolingual information retrieval systems.