應用資料探勘技術於戰術航跡資料分析研究

Abstract

資料探勘的意義即是從大量的資料中萃取出潛藏的預測資訊，使本來只是靜態的資料活化，變成有用的資訊，做為決策支援的依據。一般而言，資料探勘是針對某些特定的領域，產生特有的關連規則，從巨量的資料庫中發掘出所需的資訊與知識，供決策支援之用。與過去的決策支援系統(Decision Support System)所不同處，在於資料探勘系統根據關連規則提供自動化的預測分析能力。 因此，現代組織紛紛採用資料探勘技術來挖掘環境中隱含的知識，以提昇其競爭之優勢然而縱觀國內的相關研究，發現將資料探勘技術應用於軍事上的研究並不多見，本研究將從現有之資料探勘技術切入，預計提出適合軍事任務資料倉儲及資料探勘架構，並將針對特定軍事任務主題，如戰術航跡特性歸類分析，進行資料模擬並進而建立分類法則與模式，以支持後續決策支援任務的技術開發，以驗證所提出的架構在分類問題上的可行性。 我們獲得了下列幾項的初步成果，第一是整理並分析國內外關於資料探勘的相關文獻，其次是定義出一個戰略航跡上可供資料探勘的問題並利用資料探勘軟體中群集分析演算法進行資料探勘，進而瞭解歷史資料所潛在的模式。

Data Mining is the process of automated extraction of hidden predictive information from databases and finding new and potentially useful knowledge from it. Data mining, the extraction of hidden predictive information from large databases, is a powerful new technology with great potential to help us to focus on the most important information in data warehouse since decision support requires historical, summarized, and consolidated data from one or more operational databases over a long period. Data which is used in decision support is typically subject-oriented and integrated, and its relative importance can change across time. This research aimed at applying data warehousing and data mining technology to the decision support system of military data. We pilot test this data mining for decision support methodology in the tactic air force line data analysis. The main results of this research including designing a data mining model that can be built on the data warehouse system to meet the required analysis needs of tactics air force line. Study the relative literatures of Data warehouse and data mining. Design architecture for Data warehouse and data mining system based on the need of defense of Air Force. Analysis architecture for Data warehouse and data mining system based on the need of defense of Air Force using classification method to support decision-makers much easier in relevant decision-making. Implement a prototype model for Data warehouse and data mining system based on the need of defense of Air Force.