分散式之智慧型網路釣魚偵測機制

An Intelligent Phishing Detection Mechanism on Distributed Environment

網路釣魚是一種使用社交工程與技術性的哄騙以竊取他人身份以及財務帳號資料的犯罪手段。對個人、企業組織以及社會，都有負面的影響。隨著網路釣魚的攻擊手法與途徑的快速變化，現有的反網路釣魚機制在有限的可擴展性、偵測方式係由固定法則處理以及欠缺時效性等缺點下，難以有效偵測與判斷網路釣魚。本研究提出一個分散式智慧型網路釣魚偵測架構與機制，透過行動代理人蒐集情報，讓各個功能主機能取得更新並分享資料，減少網路釣魚攻擊的擴散並提升偵測的有效性以及有效整合分散的資源。另外，使用分類與序列型樣等智慧型技術探索出網路釣魚特徵及其關聯性與順序性，以預測此特徵將伴隨何種特徵一起出現或後續會再發生何種攻擊特徵，進而達到智慧型網路釣魚自動偵測機制。本研究屬於概念性的研究，未來除了把這概念落實在真正的分散式網路釣魚偵測系統並進行實證。

Phishing is a significant security threat to the Internet, which is a social engineering technique used to fraudulently acquire sensitive information from users by masquerading as a legitimate entity. In order to promote accuracy of phishing detection and effectiveness of resource integration, this research proposed a conceptual framework that we adopt the distributed intelligent mechanism on phishing detection. This distributed intelligent phishing detection mechanism is established through classification and sequential patterns technologies to explore the primary phishing features and the associated and sequence of them. In addition to accuracy of the detection, the mechanism provides an emergency alert of phishing attack. In the future, this conceptual mechanism will be practiced and proved this effectiveness in the future.