Journal of Cyber Security, Privacy Issues and Challenges (e-ISSN: 2583-7656)

A Review of Cyber Vulnerability Assessment Methods

Manas Kumar Yogi — Thu, 09 Nov 2023 00:00:00 +0530

In an era defined by the ubiquity of digital
technologies, the assessment of cyber
vulnerabilities has become a cornerstone of
modern cybersecurity. This paper offers a
comprehensive review of cyber vulnerability
assessment methods, emphasizing the
importance of method selection and
recognizing their significance in safeguarding
digital assets and critical infrastructure. The
paper begins by elucidating the fundamental
concepts of cyber vulnerability assessment,
tracing its historical development, and
addressing the key challenges faced by
practitioners. A structured framework is
employed to categorize assessment methods
into network-based, host-based, and
application-based approaches. These methods
are meticulously analyzed, highlighting their
respective strengths and weaknesses, and
their real-world applications are exemplified
through case studies. Moreover, this review
identifies emerging trends and approaches
within the realm of vulnerability assessment,
from the integration of artificial intelligence
and threat intelligence feeds to the need for
continuous monitoring. The discussion
culminates in a reflection on the practical
implications of the comparative analysis,
emphasizing the critical importance of
selecting the right assessment method in the
context of cybersecurity. The paper concludes
with a series of thoughtful recommendations
for both practitioners and researchers,
fostering a deeper understanding of the
dynamic field of cyber vulnerability
assessment and its ever-evolving role in the
preservation of digital integrity and security.

Machine Learning Approach Based on Synthetic Minority Over-Sampling Technique and Isolation Forest for Insider Threat Detection

Wed, 29 Nov 2023 00:00:00 +0530

Detecting insider threats is challenging due to
insiders' deep familiarity with networks and
security protocols, allowing them to bypass
traditional security measures. While various
methods combat insider threats, creating
effective detection systems remains difficult.
Research advocates using Machine Learning
(ML) techniques, but handling imbalanced
datasets reduces accuracy. To tackle this, this
paper presents "SMOTE-IForest," merging
SMOTE and IForest for insider threat
detection. Testing on the CERT r6.2 dataset
achieved 80.0% accuracy in detecting user
behaviour. Additionally, it reached a 63.4%
detection rate with a 67.0% false positive rate,
boasting a high AUC of 96.0%, 93.30%
precision, and 88.80% f-measure. This model
addresses accuracy, detection, and false
positive rate issues. SMOTE improves dataset
balance by creating synthetic samples from
the minority class, enhancing classification
accuracy. IForest isolates anomalies,
efficiently handling high-dimensional data
without complex tuning, ideal for insider
threat detection. The "SMOTE-IForest"
model significantly strengthens insider threat
detection systems by overcoming dataset
imbalance and enhancing accuracy. Its
precision and f-measure distinguish between
normal and anomalous behaviour, aiding in
addressing setbacks associated with existing
studies' accuracy, detection, and false positive
rates.

An Investigative Study of IoT Forensics Mechanisms

Yamuna Mundru, Manas Kumar Yogi — Sat, 02 Dec 2023 00:00:00 +0530

The rapid proliferation of Internet of Things
(IoT) devices has revolutionized the way we
interact with the digital world, introducing
unprecedented convenience and efficiency.
However, this transformation has also given
rise to novel cybersecurity threats and
challenges. In this context, IoT forensics
emerges as a critical field dedicated to the
investigation, analysis, and preservation of
digital evidence from IoT devices, networks,
and systems. This research study presents an
investigative exploration of IoT forensics
methods to address the unique challenges
posed by the ever-expanding IoT landscape.
This investigation delves into the diverse and
heterogeneous nature of IoT devices,
communication protocols, and data types. It
evaluates the mechanisms employed in
collecting, preserving, and analyzing digital
evidence from IoT ecosystems, emphasizing
the need for standardization and scalability.
Real-time forensics, edge computing, and the
integration of artificial intelligence are
examined in the context of their benefits and
limitations for IoT forensic investigations.
Furthermore, it highlights the evolving role of
open-source tools and cloud-based forensics
in this emerging field. As IoT continues to
expand its influence across various industries,
understanding and advancing IoT forensics
methods is essential to secure these
interconnected systems, protect data privacy,
and uphold the rule of law in the digital age

Proactive Measures to Mitigate Ransomware Attacks

Rupal Choudhary, Shivam Tiwari — Tue, 05 Dec 2023 00:00:00 +0530

Ransomware attacks are one of the key
contributors to the overall threat to the IT
industry, the focus of this research is to
outline key approaches hackers use via
ransomware and how companies and
individuals can protect themselves from it.
Ransomware attacks have developed as a
widespread and major cybersecurity threat,
causing significant financial losses and data
breaches across a wide range of sectors. This
research report offers a comprehensive
examination of ransomware attacks, their
evolution, and the strategies employed by
cybercriminals. The study also delves into
effective countermeasures and preventive
measures that organizations can adopt to
safeguard their systems and data.
Ransomware is a type of software that is
aimed to prevent a user or organisation from
accessing files on their computer. Cyber
attackers place organisations in a position
where paying the ransom is the easiest and
cheapest option to recover access to their files
by encrypting these files and demanding a
ransom payment for the decryption key.
Some variants have introduced more
functionality, such as data theft, to provide
additional motivation for ransomware victims
to pay the ransom. Ransomware has fast
become the most apparent and popular sort
of malware. Recent ransomware attacks have
hampered hospitals' capacity to offer critical
services, paralyzed city public systems, and
caused severe damage to a variety of
organisations

Detecting Spam in Email using Cyber Security

Abhijeetpandey, Vinay Kumar Singh — Wed, 20 Dec 2023 00:00:00 +0530

Due to its cost-effectiveness and efficiency,
email is being used more often for everyday
commercial transactions and general
communication, which leaves it open to
assaults like spamming. Junk emails, often
known as spam emails, are unwanted
communications that are almost similar and
are sent at random to several recipients.
Bayesian Logistic Regression, Hidden Naïve
Bayes, Radial Basis Function (RBF) Network,
Voted Perceptron, Lazy Bayesian Rule, Logit
Boost, Rotation Forest, NNge, Logistic Model
Tree, REP Tree, Naïve Bayes, Multilayer
Perceptron, Random Tree, and J48 are
among the classification algorithms whose
performance is examined in this study. Using
the WEKA data mining tool, the algorithms'
performance was evaluated in terms of
Accuracy, Precision, Recall, F-Measure, Root
Mean Squared Error, Receiver Operator
Characteristics Area, and Root Relative
Squared Error. There was no use of feature
selection or performance-boosting techniques
to provide a fair assessment of the
classification algorithms' performance.
According to the research, there are several
classification algorithms that, when
thoroughly investigated using feature
selection techniques, can produce email
classification results that are more accurate.
The classifier with the greatest accuracy,
Rotation Forest, is determined to be 94.2%.
While no algorithm was able to sort spam
emails with 100% accuracy. Rotation Forest
came closest to producing the most accurate
result