TOC
Kenjiro Cho
生駒 : 奈良先端科学技術大学院大学, 2018.12
Lecture Archive
| No. | Printing year | Location | Call Number | Material ID | Circulation class | Status | Waiting |
|---|---|---|---|---|---|---|---|
|
1 |
|
|
M016013 |
|
|
|
Finding frequent items (Heavy Hitters) in a large dataset has diverse applications and has been extensively studied. Items with hierarchical attributes such as IP addresses or time can be aggregated into groups such as subnets or coarser time. Identifying these frequent aggregates, known as Hierarchical Heavy Hitters, provides powerful means for traffic monitoring or anomaly detection. In this talk, I will review the existing algorithms for Heavy Hitter and Hierarchical Heavy Hitter problems, and then, introduce an efficient algorithm based on recursive space partitioning. I'll also introduce a traffic monitoring tool based on the proposed method that has been used for monitoring the WIDE backbone traffic.
2018
電子化映像資料（ 1時間28分10秒）
情報科学領域・コロキアム ; 平成30年度
講演日: 平成30年12月7日
講演場所: 情報科学棟大講義室(L1)
Japan
English (eng)
English (eng)
