臺灣地處地震帶，部分地區每年均要歷經多次的地震以及周期性的大地震，往往帶來嚴重的山崩以及土石鬆動的現象，若再歷經豪雨後，將帶來嚴重的土石流，造成生命財產的重大損失。1999年9月21日於臺灣中部地區發生芮氏規模7.3的地震，此次強震造成2000多人死亡，難以估計的財產損失，以及數量龐大的坡地災害。而在此次強震中亦獲得大量的坡地破壞資料及地震記錄，可供專家學者進行地震對山崩影響的研究，使得對地震引致山崩的行為有進一步的認識。本研究嘗試以九二一大地震所累積的大批地理資料，經地理資訊系統處理後建立資料倉儲，做為探勘的基礎，再結合地理資訊系統空間資料分析與處理的功能與資料探勘技術中的決策樹 (Decision Tree) 演算法建立預測模式，應用在空間資料庫的知識探索，獲得良好的準確性，期望透過此研究了解引發地震山崩的機制，並且建立山崩災害的潛感圖，提供防災決策支援上的參考。

An earthquake of the magnitude 7.3 on the Richter scale occurred in the middle region of Taiwan on September 21, 1999. This earthquake caused the death of more than two thousand people , severe property loss, and a large number of landslides. A large amount of landslide data and strong motion records were produced by the Chi-Chi earthquake . This research establishes a landslide database and data warehouse collected from the Chi-Chi earthquake in a Geographic Information System (GIS). Using the powerful data-processing function and spatial analysis ability of GIS and Decision Tree Data Mining modeling, , we have identified the possible mechanisms that induce landslides and produced a map of landslide susceptibility. Verification proved that the research result is good and the methods used by this research suitable for the forecast of risk assessment of landslide hazard triggered by earthquakes. The findings can be used as a tool for disaster mitigation decision support.