資料探勘技術可運用於許多領域，例如：行銷分析、決策支援、詐欺偵測、企業管理等等。資料探勘研究領域發展了許多技術從大量資料中分析出有用的資訊，而序列樣式探勘是其中重要的技術之一。過去的序列樣式探勘技術多為針對點式事件所設計，也就是說，這些技術所探勘的序列資料中，所有的事件皆發生於某個時間點。然而，在許多應用中，事件並非必然只發生在一個時間點，而可能持續發生於一段時間，這樣的事件稱之為區間式事件。在區間式事件所組成的序列中尋找頻繁樣式，稱之為區間序列樣式探勘。而在其它一些應用中，序列中的事件也許不必然為點式或區間式，而是兩種事件皆可能發生的情況。此類序列稱之為混合事件序列。而在這類序列中尋找頻繁序列即稱之為混合序列樣式探勘。由於傳統的序列樣式探勘方法無法用來探勘區間事件序列或混合序列樣式，因此本文提出兩個方法分別用以探勘區間序列樣式及混合序列樣式。經由一連串實驗過程 (包含人工資料及真實資料)，說明此二探勘方法皆為有效率及有效。 Data mining is useful in various domains, such as market analysis, decision support, fraud detection and business management, among others. Many approaches have been proposed to extract information and sequential pattern mining is one of the mostimportant methods. Previous studies of sequential pattern mining have discovered patterns from point-based event sequences. However, in some applications, event sequences may contain interval-based events or hybrid events (both point-based and interval-based events). Frequent patterns discovered from interval-based event sequences are called temporal patterns, and those discovered from hybrid event sequences are called hybrid temporal patterns. But because the existing methods for discovering sequential patterns are not applicable to mine temporal pattern or hybrid patterns, this study is dedicated to develop new methods to discover temporal patterns and hybrid temporal patterns. Both proposed methods have been verified for efficiency and effectiveness by using synthetic and real datasets.
