| dc.creator (作者) | 蘇民揚 | zh_TW |
| dc.creator (作者) | 戴宏偉 | zh_TW |
| dc.creator (作者) | 龍京佑 | zh_TW |
| dc.date (日期) | 2006 | |
| dc.date.accessioned | 18-Dec-2017 17:40:19 (UTC+8) | - |
| dc.date.available | 18-Dec-2017 17:40:19 (UTC+8) | - |
| dc.date.issued (上傳時間) | 18-Dec-2017 17:40:19 (UTC+8) | - |
| dc.identifier.uri (URI) | http://nccur.lib.nccu.edu.tw/handle/140.119/115210 | - |
| dc.description.abstract (摘要) | 網路安全領域中，網路入侵偵測系統扮演重要的角色；而幾乎所有的網路入侵偵測工具都必須藉由分析網路流量方能成事。網路流量資訊以極快速的速度在持續改變中，如何能有效率的動態分析網路流量，便成為網路入侵偵測系統能否成功的重要關鍵。本文提出了一個快速的模糊關聯規則產生演算法，適用於即時分析快速變化的網路流量。細心安排資料結構及妥善的運用大量記憶體空間，我們每2 秒鐘統計一筆網路流量的資訊，針對6 種特徵、每種特徵分低、中、高，3 種程度做挖掘，處理一筆新進資料(動態探勘一次)平均只需要0.0067秒；而且時間不會隨著累積資料變多而增加，可以有效的符合線上即時流量分析的需求。 | |
| dc.description.abstract (摘要) | In network security field, network intrusion detection system (NIDS) plays an important role. Almost all NIDSs have to analyze traffic in first to complete their jobs. However, traffic information is changed so fast and consistently. The method to dynamically and efficiently analyze traffic information is a prerequisite for the success of a NIDS. The paper presents an algorithm for mining fuzzy association rules in changed very frequently incremental database, like traffic information. We repeatedly collected traffic information in period of 2 seconds to form a record, using elaborate data structure and mass memory our algorithm can mine a set of fuzzy association rule in 0.0067 seconds for an incremental record, while six features and three degrees per feature being considered. Moreover, the time for once mining, due to an incremental record, would not increase while historical dataset augmented. | |
| dc.format.extent | 201058 bytes | - |
| dc.format.mimetype | application/pdf | - |
| dc.relation (關聯) | TANET 2006 台灣網際網路研討會論文集 | zh_TW |
| dc.relation (關聯) | 資通安全、不當資訊防治 | zh_TW |
| dc.subject (關鍵詞) | 網路安全 ; 網路入侵偵測 ; 關聯規則 ; 模糊關聯規則 ; 網路流量分析 | zh_TW |
| dc.title (題名) | 適用於即時網路流量分析的快速模糊關聯規則產生方法 | zh_TW |
| dc.type (資料類型) | conference | |
