A study on ambient DNS sensor(Fostering Joint International Research)

2017 Fiscal Year Final Research Report

• Project/Area Number: 15KK0019
• Research Category: Fund for the Promotion of Joint International Research (Fostering Joint International Research)
• Allocation Type: Multi-year Fund
• Research Field: Information network
• Research Institution: National Institute of Informatics
• Principal Investigator: Fukuda Kensuke 国立情報学研究所, アーキテクチャ科学研究系, 准教授 (90435503)
• Research Collaborator: Heidemann John 南カリフォルニア大学, 情報科学研究所, 教授 ; Kato Akira 慶應義塾大学, 大学院・メディアデザイン研究科, 教授 ; Mazel Johan 国立情報システムセキュリティ庁, 研究部門, 研究員
• Project Period (FY): 2016 – 2017
• Keywords: インターネット / DNS / セキュリティ

Outline of Final Research Achievements

We evaluate DNS backscatter for detecting internet-wide network scannings. We first construct a darknet taxonomy that is used for labeled data of network scans, and evaluate the effectiveness of the taxonomy with over 10 years darknet traffic data. We also evaluate the dependency of the labeled data source on the classification accuracy, and demonstrate that two class classification with SMOTE achieves acceptable performance on the scan detection.

Free Research Field

インターネット工学