The Comparison of Cybersecurity Datasets

Alshaibi, Ahmed; Alani, Mustafa M.; Al-Azzawi, Abeer; Konev, Anton; Shelupanov, Alexander

doi:10.3390/data7020022

Cited by 20 publications

(8 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In any case, the datasets used in the state-of-the-art literature are appropriate for improving the wide range of algorithms arising from new research, but will not be able to generate new and tailored samples for explorations where more specific information is needed [ 18 ].…”

Section: Aim and Motivation—the Problem With Cyber Threat Datasetsmentioning

confidence: 99%

ReinforSec: An Automatic Generator of Synthetic Malware Samples and Denial-of-Service Attacks through Reinforcement Learning

Hernandez-Suarez

Sánchez-Pérez

Olivares-Mercado

et al. 2023

Sensors

View full text Add to dashboard Cite

In recent years, cybersecurity has been strengthened through the adoption of processes, mechanisms and rapid sources of indicators of compromise in critical areas. Among the most latent challenges are the detection, classification and eradication of malware and Denial of Service Cyber-Attacks (DoS). The literature has presented different ways to obtain and evaluate malware- and DoS-cyber-attack-related instances, either from a technical point of view or by offering ready-to-use datasets. However, acquiring fresh, up-to-date samples requires an arduous process of exploration, sandbox configuration and mass storage, which may ultimately result in an unbalanced or under-represented set. Synthetic sample generation has shown that the cost associated with setting up controlled environments and time spent on sample evaluation can be reduced. Nevertheless, the process is performed when the observations already belong to a characterized set, totally detached from a real environment. In order to solve the aforementioned, this work proposes a methodology for the generation of synthetic samples of malicious Portable Executable binaries and DoS cyber-attacks. The task is performed via a Reinforcement Learning engine, which learns from a baseline of different malware families and DoS cyber-attack network properties, resulting in new, mutated and highly functional samples. Experimental results demonstrate the high adaptability of the outputs as new input datasets for different Machine Learning algorithms.

show abstract

Section: Aim and Motivation—the Problem With Cyber Threat Datasetsmentioning

confidence: 99%

ReinforSec: An Automatic Generator of Synthetic Malware Samples and Denial-of-Service Attacks through Reinforcement Learning

Hernandez-Suarez

Sánchez-Pérez

Olivares-Mercado

et al. 2023

Sensors

View full text Add to dashboard Cite

show abstract

“…Standard risk analysis processes have been vulnerable for insurance companies due to the lack of recorded claims data [30]. [12] All industrial internet of things (IIoT) attacks happen at the information transmission layer, contrary to most references.…”

Section: Literature Reviewmentioning

confidence: 99%

A literature review of financial losses statistics for cyber security and future trend

Sharif¹,

Mohammed²

2022

World J. Adv. Res. Rev.

View full text Add to dashboard Cite

Cybercrime directs to any criminal activity taken out utilizing computers or the internet. Attackers have chosen strategies such as social engineering, phishing, and malware as part of their cyber-attacks. A cyber-attack can lead to various effects, ranging from stealing individual data to extortion money or losing helpful information. Society and systems depend on critical infrastructures like power plants, hospitals, and financial services companies. This paper analyzes financial losses statistics for cyber security and future trends. The cost of cybercrime prevention is increasing day by day. Financial losses refer to damages to the wealth of an organization. This includes organizational losses, compensation, and legal fees. By financial loss, we mean increased costs or reduced income caused by the threat. We collect data from various datasets and information from sources. After collecting data, we analyze the data and create a different chart to identify the growth of cyber-attacks, cyber security, and cybercrime costs. We analyze global and worldwide cybercrime status. We also investigate state-wise cybercrime and the cyber security status of the United States of America. Our main objective of the analysis is to find out the financial losses and future trends of cybercrime and cyber security. From our study, we noticed that the number of cybercrimes and their management and prevention costs are rapidly increasing in the USA and worldwide.

show abstract

“… Like much of the state-of-the-art, CYSAS-S3 provides a large amount of Metadata . Among the different collections surveyed, CYSAS-S3 is the only one that combines host, network and Mission (operation line) traces dependent on the above domains [ 72 ]. CYSAS-S3 is the only one in which the cyber Kill-Chains are clearly visible; …”

Section: Design Principlesmentioning

confidence: 99%

“…Among the different collections surveyed, CYSAS-S3 is the only one that combines host, network and Mission (operation line) traces dependent on the above domains [ 72 ].…”

Section: Design Principlesmentioning

confidence: 99%

Introducing the CYSAS-S3 Dataset for Operationalizing a Mission-Oriented Cyber Situational Awareness

Choumanof

Sánchez

Mayo

et al. 2022

Sensors

View full text Add to dashboard Cite

The digital transformation of the defence sector is not exempt from innovative requirements and challenges, with the lack of availability of reliable, unbiased and consistent data for training automatisms (machine learning algorithms, decision-making, what-if recreation of operational conditions, support the human understanding of the hybrid operational picture, personnel training/education, etc.) being one of the most relevant gaps. In the context of cyber defence, the state-of-the-art provides a plethora of data network collections that tend to lack presenting the information of all communication layers (physical to application). They are synthetically generated in scenarios far from the singularities of cyber defence operations. None of these data network collections took into consideration usage profiles and specific environments directly related to acquiring a cyber situational awareness, typically missing the relationship between incidents registered at the hardware/software level and their impact on the military mission assets and objectives, which consequently bypasses the entire chain of dependencies between strategic, operational, tactical and technical domains. In order to contribute to the mitigation of these gaps, this paper introduces CYSAS-S3, a novel dataset designed and created as a result of a joint research action that explores the principal needs for datasets by cyber defence centres, resulting in the generation of a collection of samples that correlate the impact of selected Advanced Persistent Threats (APT) with each phase of their cyber kill chain, regarding mission-level operations and goals.

show abstract

The Comparison of Cybersecurity Datasets

Cited by 20 publications

References 52 publications

ReinforSec: An Automatic Generator of Synthetic Malware Samples and Denial-of-Service Attacks through Reinforcement Learning

ReinforSec: An Automatic Generator of Synthetic Malware Samples and Denial-of-Service Attacks through Reinforcement Learning

A literature review of financial losses statistics for cyber security and future trend

Introducing the CYSAS-S3 Dataset for Operationalizing a Mission-Oriented Cyber Situational Awareness

Contact Info

Product

Resources

About