DSCP：在資料流環境下探勘連續性路徑型樣

DSCP: An Efficient Algorithm for Mining Consecutive Traversal Paths from Data Stream

隨著資訊科技的進步，交易、文件、日常處理的資料被電子化大量的累積，資料探勘的技術變得日益重要。其中，資料流探勘在資料探勘的領域中也扮演相當重要的角色，隨著各種新興應用的崛起，例如網路流量分析、網頁點選串流探勘、全球衛星定位系統路徑分析等，所處理的不再是靜態的資料，而是一連串即時且連續的動態資料流 （dynamic data stream）。本研究主要針對在資料串流中的滑動視窗模式（sliding window model）下探勘連續性路徑而提出了一個新的演算法DSCP （an efficient algorithms for mining consecutive traversal paths from data stream）。DSCP演算法的樹狀結構在探勘過程中容易維護並且不需對其樹狀結構的節點做大量的調整並配合本研究所提出的遮罩路徑過濾機制及連續路徑結合機制，而有效加快連續性路徑的探勘速度。

Due to the science and technology make a great progress, transactions, documents and data are transformed into electronic types, the large number of data has been accumulated. Therefore, data mining technology becomes more important than before in recent years. In data mining territory, data stream mining plays an important role. The data of new applications is not static any more. Net flow analysis, web-click mining and Global Position System are examples of applications. Instead of generating static data, they generate dynamic and real time data stream. In this paper we propose an algorithm DSCP (An Efficient Algorithms for Mining Consecutive Traversal Paths from Data Stream). It uses the concept of CTP, DSCP-tree structure, and sliding window model for mining consecutive traversal paths from data stream.