NEWS: Logpoint Acquires Muninn
Muninn Innovation Lab (MIL) has a specific focus on developing and implementing new and innovative solutions to address cybersecurity challenges such as false positives, malicious encrypted network traffic, sophisticated and stealthy attacks, the overwhelming volume of network traffic data etc.
MIL is Muninn’s gateway to collaborate closely with industry partners and academic institutions to stay at the forefront of cybersecurity research and development. MIL has established a highly successful collaboration with leading universities for grants, research projects and internships.
Industrial Postdoc:
Higher Order Threat Intelligence (HOTI):
Applying Machine Learning to Network Security Metadata
Periode: 2022-2024
Development of advanced maritime cybersecurity:
Muninn Maritime (M2)
Periode: 2019-2020
This project is a collaboration among Muninn, the Technical University of Denmark (DTU) and Innovationsfonden Danmark (IFD) under an industrial postdoc project to add new AI-based features to Muninn AI Detect which is the main product from Muninn. The main objective of this project is to improve the accuracy of network security notifications by inspecting network security metadata extracted from security alerts. As the rate of false alarms impacts Network Intrusion Detection System (NIDS) solutions' accuracy, this project will focus on alerts data to provide alerts analysis techniques for reducing false security alerts. The output of this R&D project will improve the cyber security threat detection and reporting capabilities of Muninn, thus providing more accurate and actionable alarms to Muninn customers. From a research perspective, this project proposes novel techniques for inspecting NIDS alerts using recent advancements in machine learning and deep learning.
Please contact us for more information on our open projects.
Cybersecurity solutions such as network intrusion detection and response (NDR) systems raise security notifications (alerts) normally based on hosts behaviors (signature-based or anomaly-based). As hosts are being used by different users, different hosts may have different patterns of security notifications. Therefore, the distribution of data from different hosts could be different in the dataset containing security notifications. In this case, clustering similar hosts in the same clusters may help security analysts get good understanding of a host under investigation. The output of clustering will also help supervised machine learning task as we may want to train our classifier on the data of hosts with the same data distribution (located in the same cluster).This project will look at data in the form of time series/sequence of security notifications and applies different approaches suitable for clustering of hosts.
Please contact us for more information on our open projects.
