Sharing and Automation for Privacy Preserving Attack Neutralization

Informazioni relative al progetto

SAPPAN

ID dell’accordo di sovvenzione: 833418

DOI 10.3030/833418

Progetto chiuso

Data della firma CE 25 Aprile 2019

Data di avvio 1 Maggio 2019

Data di completamento 30 Aprile 2022

Finanziato da

• INDUSTRIAL LEADERSHIP - Leadership in enabling and industrial technologies - Information and Communication Technologies (ICT)

Costo totale € 4 700 053,39

Contributo UE € 4 175 070,00

Coordinato da

FRAUNHOFER GESELLSCHAFT ZUR FORDERUNG DER ANGEWANDTEN FORSCHUNG EV
Germany

Risultati finali

Documents, reports (22)

Visualisation requirements

Report of practitioners’ requirements with respect to visualisation.

Dissemination Plan and Reports at M12

A plan for the dissemination of project results with a detailed set of strategic actions for the whole duration of the project and beyond.

Approach for capturing incident response and recovery steps

Approach to involve humans and capture response and recovery steps in new environment and/or from existing tools.

Global model based on shared local models, first version

Techniques for federated learning of a global anomaly detection model based on sharing locally trained models

Exploitation plan and reports at M12

Plan for the commercial exploitation and technology transfer for the project, which will define the strategy followed by the consortium to take exploitable outcomes of the project to market.

Global model based on shared anonymised data, final version

Techniques for distributed learning of a global anomaly detection model based on shared anonymized data

Dissemination Plan and Reports at M24

A plan for the dissemination of project results with a detailed set of strategic actions for the whole duration of the project and beyond

Sharing response handling information, first version

Methods for sharing response handling information in a privacypreserving fashion

Vocabulary for incident data and response and recovery actions

Interoperable and standardised vocabulary to express knowledge captured from existing tools outside of SAPPAN and from within the project

Global model without sharing local models, final version

Techniques for federated learning of a global anomaly detection model from locally trained models in a privacy preserving fashion

High-impact use cases analysis

Define SAPPAN use-cases.

Global model based on shared local models, final version

Techniques for federated learning of a global anomaly detection model based on sharing locally trained models

Dissemination Plan and Reports at M3

A plan for the dissemination of project results with a detailed set of strategic actions for the whole duration of the project and beyond.

Report on Information and Presentation Materials at M12

Materials that will help presenting the project’s results to all interested stakeholders.

Report on Information and Presentation Materials at M24

Materials that will help presenting the projects results to all interested stakeholders

Global model without sharing local models, first version

Techniques for federated learning of a global anomaly detection model from locally trained models in a privacy preserving fashion

Report on Information and Presentation Materials at M36

Materials that will help presenting the projects results to all interested stakeholders

Global model based on shared anonymised data, first version

Techniques for distributed learning of a global anomaly detection model based on shared anonymized data

IP plan, reports and IPR issues at M12

Update of the plan for IP management, including the outcome of negotiation meetings to determine the percentage of effective contribution of each individual partner to each IPR object.

Dissemination Plan and Reports at M36

A plan for the dissemination of project results with a detailed set of strategic actions for the whole duration of the project and beyond

Sharing response handling information, final version

Methods for sharing response handling information in a privacypreserving fashion

Privacy requirements

Report of practitioners’ requirements with respect to privacy.

Demonstrators, pilots, prototypes (9)

Demonstrator of visual support for designing detection models, initial version

Initial version of demonstrator for interactive detection model analysis building and visualisation of networkrelated data

SAPPAN dashboard, initial version

Prototypical development of dashboard integrating all visualisations that serves as enduser frontend

Demonstrator for visualisation support for distributed and federated learning

Demonstrator extending the Demonstrator of visual support for designing detection models with support for distributed and federated learning

SAPPAN demonstrator

SAPPAN system integrated from individual components

Demonstrator for tracking provenance in visual analyses, final version

Demonstrator that extends the Demonstrator of visual support for designing detection models with support for tracking the users activities and presenting them in an appropriate way to follow the analysis process

Demonstrator for tracking provenance in visual analyses, first version

Demonstrator that extends the Demonstrator of visual support for designing detection models with support for tracking the users activities and presenting them in an appropriate way to follow the analysis process

SAPPAN dashboard, final version

Prototypical development of dashboard integrating all visualisations that serves as enduser frontend

Demonstrator for uncertainty visualisation

Demonstrator for visualisation of uncertainty inherent to machine learning techniques and induced by anonymisation

Demonstrator of visual support for designing detection models, final version

Final version of demonstrator for interactive detection model analysis building and visualisation of networkrelated data

Other (4)

Algorithm to automate recommended response and recovery actions without human operators, final version

Algorithm to use captured knowledge to perform actions automatically

Algorithm to automate recommended response and recovery actions without human operators, first version

Algorithm to use captured knowledge to perform actions automatically.

Algorithm to recommend response and recovery actions to human operators, final version

Algorithm to use captured knowledge to recommend actions to human

Algorithm to recommend response and recovery actions to human operators, first version

Algorithm to use captured knowledge to recommend actions to human.

Websites, patent fillings, videos etc. (1)

Project Website

A website for public dissemination of the progress and achievements.

Pubblicazioni

Thesis and dissertations (26)

Laying Ground Truth in Network-based Software Fingerprinting

Autori: Olav Lamberts
Pubblicato in: 2020
Editore: RWTH Aachen University

Local Differential Privacy Preserving Intrusion Detection Systems

Autori: Melanie Martini
Pubblicato in: 2020
Editore: RWTH Aachen University

Graph-Based Analysis of IP Flows

Autori: Aneta Jablunkova
Pubblicato in: 2021
Editore: Masaryk University

A Study of Model Inversion Defenses in Deep Learning

Autori: Benedikt Holmes
Pubblicato in: 2020
Editore: RWTH Aachen University

Detecting Obfuscated Scripts With Machine Learning Techniques

Autori: Mariam Pogosova
Pubblicato in: 2020
Editore: Aalto University

Visual Correlation of Network Flows and Host-Based Data

Autori: Carlos Moreno Sanchez
Pubblicato in: 2020
Editore: University of Stuttgart

Applying Privacy Preserving Data Mining to Intrusion Detection Systems

Autori: Clemens Frank
Pubblicato in: 2019
Editore: RWTH Aachen University

Multiclass Classification of DGAs using Classical Machine Learning Approaches

Autori: Nils Faerber
Pubblicato in: 2020
Editore: RWTH Aachen University

Graph-based Anomaly Detection in Network Traffic

Autori: Denisa Sramkova
Pubblicato in: 2021
Editore: Masaryk University

Hardening of Domain Generation Algorithm Classifiers

Autori: Mike Lorang
Pubblicato in: 2020
Editore: RWTH University

Distributed and Automated Network Traffic Generation of Applications with a Graphical User Interface

Autori: Paul Suetterlin
Pubblicato in: 2021
Editore: RWTH Aachen University

DGA Detection on Encrypted DNS Traffic using Machine Learning

Autori: Markus Baumgart
Pubblicato in: 2020
Editore: RWTH Aachen University

Application Fingerprinting using Deep Learning based on Network Traffic

Autori: Raoul Offizier
Pubblicato in: 2022
Editore: RWTH Aachen University

Adversarial Attacks and Defenses on DGA classifiers

Autori: Konstantin Kaulen
Pubblicato in: 2021
Editore: RWTH Aachen University

Anonymization and Sharing of application-labeled Network Traces

Autori: Alexander Loebel
Pubblicato in: 2022
Editore: RWTH Aachen University

Software Specific Network Traffic Generation by Automation of the Graphical User Interface

Autori: Frederik Basels
Pubblicato in: 2020
Editore: RWTH Aachen University

Visual Comparison of Classifications from Different Machine Learning Models

Autori: Komail Mohammadi
Pubblicato in: 2020
Editore: University of Stuttgart

Application Fingerprinting based on System Events using Process Mining

Autori: Nicolas Heinen
Pubblicato in: 2021
Editore: RWTH Aachen University

Machine Learning for Phishing URL Detection

Autori: Juraj Smeriga
Pubblicato in: 2020
Editore: Masaryk University

A Condensation-based Anonymization Approach for Intrusion Detection

Autori: Jonas Rülfing
Pubblicato in: 2020
Editore: RWTH Aachen University

A Privacy-Preserving Machine Learning Approach for DGA Detection

Autori: Tim Amelung
Pubblicato in: 2021
Editore: RWTH Aachen University

Utilizing Adverserial Attacks for Iterative Hardening of DGA Classifiers

Autori: Nils Eberhardt
Pubblicato in: 2020
Editore: RWTH Aachen University

Application of Process Mining in Software Fingerprinting

Autori: Christian van Sloun
Pubblicato in: 2020
Editore: RWTH Aachen University

Machine Learning based Handling of Cyber Security Incidents

Autori: Marc Burian
Pubblicato in: 2019
Editore: RWTH Aachen University

Detection of new DGAs in the Multiclass DGA classification setting

Autori: Justus von Brandt
Pubblicato in: 2022
Editore: RWTH Aachen University

Process for Automated Threat Response to Phishing

Autori: Michal Čech
Pubblicato in: 2021
Editore: Masaryk University

Peer reviewed articles (4)

Predictive methods in cyber defense: Current experience and research challenges

Autori: Martin Husák; Václav Bartoš; Pavol Sokol; Andrej Gajdoš
Pubblicato in: Future Generation Computer Systems, Numero 115, 2021, Pagina/e 517-530, ISSN 0167-739X
Editore: Elsevier BV
DOI: 10.1016/j.future.2020.10.006

From Collaboration to Automation: A Proof of Concept for Improved Incident Response

Autori: Lasse Nitz, Martin Zadnik, Mehdi Akbari Gurabi, Mischa Obrecht, Avikarsha Mandal
Pubblicato in: ERCIM News, Numero 129, 2022, Pagina/e 31-32, ISSN 0926-4981
Editore: ERCIM EEIG
DOI: 10.24406/publica-146

Towards Privacy-Preserving Sharing of Cyber Threat Intelligence for Effective Response and Recovery

Autori: Lasse Nitz, Mehdi Akbari Gurabi, Avikarsha Mandal, Benjamin Heitmann
Pubblicato in: ERCIM News, Numero 126, 2021, Pagina/e 33-34, ISSN 0926-4981
Editore: ERCIM EEIG

Host Behavior in Computer Network: One-Year Study

Autori: Tomas Jirsik; Petr Velan
Pubblicato in: IEEE Transactions on Network and Service Management, Numero 18, 2021, Pagina/e 822-838, ISSN 1932-4537
Editore: Institute of Electrical and Electronics Engineers
DOI: 10.1109/tnsm.2020.3036528

Conference proceedings (16)

Interpretable Visualizations of Deep Neural Networks for DomainGeneration Algorithm Detection

Autori: Becker, Franziska; Drichel, Arthur; Müller, Christoph; Ertl, Thomas
Pubblicato in: 2020 IEEE Symposium on Visualization for Cyber Security (VizSec), 2020
Editore: IEEE
DOI: 10.1109/vizsec51108.2020.00010

First Step Towards EXPLAINable DGA Multiclass Classification

Autori: Arthur Drichel; Nils Faerber; Ulrike Meyer
Pubblicato in: ARES 2021: The 16th International Conference on Availability, Reliability and Security, 2021
Editore: ACM
DOI: 10.1145/3465481.3465749

DoH Insight - detecting DNS over HTTPS by machine learning

Autori: Dmitrii Vekshin, Karel Hynek, Tomas Cejka
Pubblicato in: Proceedings of the 15th International Conference on Availability, Reliability and Security, 2020, Pagina/e 1-8, ISBN 9781450388337
Editore: ACM
DOI: 10.1145/3407023.3409192

Privacy Illusion: Beware of Unpadded DoH

Autori: Karel Hynek, Tomas Cejka
Pubblicato in: IEEE IEMCON 2020, 2020
Editore: IEEE

Finding Phish in a Haystack: A Pipeline for Phishing Classification on Certificate Transparency Logs

Autori: Arthur Drichel; Vincent Drury; Justus von Brandt; Ulrike Meyer
Pubblicato in: ARES 2021: The 16th International Conference on Availability, Reliability and Security, 2021
Editore: ACM
DOI: 10.1145/3465481.3470111

Sharing FANCI Features: A Privacy Analysis of Feature Extraction for DGA Detection

Autori: Benedikt Holmes; Arthur Drichel; Ulrike Meyer
Pubblicato in: The Sixth International Conference on Cyber-Technologies and Cyber-Systems CYBER 2021, 2021
Editore: IARIA

Towards Evaluating Quality of Datasets for Network Traffic Domain

Autori: D. Soukup, P. Tisovčík, K. Hynek and T. Čejka
Pubblicato in: 17th International Conference on Network and Service Management (CNSM), 2021
Editore: IEEE
DOI: 10.23919/cnsm52442.2021.9615601

A Study on Collaborative Machine Learning for DGA Detection

Autori: Arthur Drichel; Benedikt Holmes; Justus von Brandt; Ulrike Meyer
Pubblicato in: Proceedings of the 3rd Workshop on Cyber-Security Arms Race (CYSARM ’21), 2021
Editore: ACM
DOI: 10.1145/3474374.3486915

On the Integration of Course of Action Playbooks into Shareable Cyber Threat Intelligence

Autori: V. Mavroeidis, P. Eis, M. Zadnik, M. Caselli and B. Jordan
Pubblicato in: IEEE International Conference on Big Data (Big Data), 2021, Pagina/e 2104-2108
Editore: IEEE
DOI: 10.1109/bigdata52589.2021.9671893

Analyzing the real-world applicability of DGA classifiers

Autori: Arthur Drichel, Ulrike Meyer, Samuel Schüppen, Dominik Teubert
Pubblicato in: Proceedings of the 15th International Conference on Availability, Reliability and Security, 2020, Pagina/e 1-11, ISBN 9781450388337
Editore: ACM
DOI: 10.1145/3407023.3407030

Making use of NXt to nothing - the effect of class imbalances on DGA detection classifiers

Autori: Arthur Drichel, Ulrike Meyer, Samuel Schüppen, Dominik Teubert
Pubblicato in: Proceedings of the 15th International Conference on Availability, Reliability and Security, 2020, Pagina/e 1-9, ISBN 9781450388337
Editore: ACM
DOI: 10.1145/3407023.3409190

Towards Inference of DDoS Mitigation Rules

Autori: Martin Zadnik
Pubblicato in: IEEE/IFIP Network Operations and Management Symposium, 2022, Pagina/e 1-5
Editore: IEEE

Detection of https brute-force attacks with packet-level feature set

Autori: LUXEMBURK, Jan; HYNEK, Karel; ČEJKA, Tomáš
Pubblicato in: 2021 IEEE 11th Annual Computing and Communication Workshop and Conference (CCWC), Numero 11, 2021, Pagina/e 0114-0122
Editore: IEEE
DOI: 10.1109/ccwc51732.2021.9375998

GRANEF: Utilization of a Graph Database for Network Forensics

Autori: Milan Cermak, Denisa Sramkova
Pubblicato in: 18th International Conference on Security and Cryptography (SECRYPT 2021), Numero 6. - 8. 7. 2021, 2021, Pagina/e 785-790, ISBN 978-989-758-524-1
Editore: SCITEPRESS
DOI: 10.5220/0010581807850790

VITALflow: Visual Interactive Traffic Analysis with NetFlow

Autori: Tina Tremell, Jochen Kögell, Florian Jauernigl, Sebastian Meierl, Dennis Thom, Franziska Becker, Christoph Müller, Steffen Koch
Pubblicato in: 7th International Workshop on Analytics for Network and Service Management (ANNET 2022), 2022
Editore: IEEE

Towards Privacy-Preserving Classification-as-a-Service for DGA Detection

Autori: Arthur Drichel; Mehdi Akbari Gurabi; Tim Amelung; Ulrike Meyer
Pubblicato in: 18th International Conference on Privacy, Security and Trust (PST), 2021
Editore: IEEE
DOI: 10.1109/pst52912.2021.9647755

Other (2)

Large Scale Measurement on the Adoption of Encrypted DNS

Autori: García, Sebastián; Hynek, Karel; Vekshin, Dmtrii; Čejka, Tomáš; Wasicek, Armin
Pubblicato in: 2021
Editore: arXiv
DOI: 10.48550/arxiv.2107.04436

Leveraging Machine Learning for DGA Detection

Autori: Drichel, Arthur
Pubblicato in: International Workshop on Next Generation Security Operations Centers, NG-SOC 2020, , 2020-08-25 - 2020-08-25, 2020
Editore: RWTH Aachen University