Scribd
Upload
2 views

DissertationProp

The research project proposal focuses on utilizing machine learning to detect anomalies in network traffic, highlighting the importance of cybersecurity in an increasingly connected world. The primary objective is to evaluate the performance of various machine learning algorithms for effective anomaly detection, with a structured timeline for research, testing, and dissertation writing. The study aims to contribute to both academic knowledge and practical applications in enhancing network security.

Uploaded by

Kazumi
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
2 views

DissertationProp

The research project proposal focuses on utilizing machine learning to detect anomalies in network traffic, highlighting the importance of cybersecurity in an increasingly connected world. The primary objective is to evaluate the performance of various machine learning algorithms for effective anomaly detection, with a structured timeline for research, testing, and dissertation writing. The study aims to contribute to both academic knowledge and practical applications in enhancing network security.

Uploaded by

Kazumi
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 5

Research Project Proposal

Title:
Machine Learning for Detecting
Anomalies in Network Traffic.

Program: BSc in Computing with Foundation (Cyber


Security)
Course: Undergraduate Research Project (CLD6001)
Supervisor: Dr. Vasilis Anastopoulos
Student name: Thomais Paraskevaidi
Student ID:
Word Count: 694

1
Research Project Proposal

Table of Contents

Background.............................................................................................................................3
Research Objectives..........................................................................................................3-4
Research and Development method...................................................................................4
Project Timeline.....................................................................................................................4
Resources...............................................................................................................................5
References..............................................................................................................................5

Background
The rising dependency on network systems and devices has made network
security more important in the digital age. As connectivity grows, so are

2
Research Project Proposal

dangers. Strong steps are needed to protect networks from hostile activities.
Cyber threats range from advanced attacks to minor network traffic
anomalies, emphasizing the need for effective anomaly detection. Current
studies on machine learning for network anomaly identification show a
growing focus on complex algorithms to improve cybersecurity. The research
emphasizes the need for immediate recognition, precise classification, and
flexibility in responding to new threats.
Traditional methods for detecting network irregularities struggle to keep up
with cyber attackers' ever-changing schemes. Traditional approaches struggle
to detect minute deviations that may suggest hazards due to the volume and
complexity of data. Advanced methods, including machine learning, are
needed to detect abnormalities and secure computer networks.

Research Objectives
The objectives of the current undergraduate project are:

Primary Objectives:

- Assess the performance of different machine learning algorithms in detecting


various types of network anomalies.

Research Questions
1. Can machine learning algorithms effectively identify and classify network
anomalies with high accuracy?
2. Which Algorithms fits best unsupervised ML for Anomaly Detection in
Network Traffic?

Significance of Study

Academic Significance:
This paper enhances academic research about machine learning in network
security. New methods are explored to improve the field and permit future
research.

3
Research Project Proposal

Practical Significance:
This study improves real-world cybersecurity research and experimentation.
Advanced anomaly detection can improve network security by detecting
threats quickly.

Research and Development method


This study analyses academic research and tests machine learning to detect
network abnormalities using algorithms. Critically analyzing reliable academic
books and journals creates a solid theoretical foundation.

The practical component will focus on building and using machine learning
models to apply research findings. Models will be chosen based on their
capacity to detect abnormalities in network activity. The models' performance
will be measured using industry-standard methods.

After selecting the dataset based on network anomaly detection criteria, it will
be preprocessed for algorithmic analysis. The research will follow ethical
guidelines for data utilization, notably when managing network traffic data.

The research will follow ethical data utilization norms, especially in network
traffic data management. This method tries to demonstrate machine learning's
network anomaly detection capabilities. It does this by combining academic
theory with experimentation to provide a succinct yet comprehensive method.

Project Timeline
1. Conducting research and generating ideas while collecting relevant
material and resources. Deadline: January
2. Begin testing the ML Algorithms using Python with a deadline up to
middle February.
3. Start categorizing the Algorithms based on speed, efficiency and
capability with a deadline up to the start of March.

4
Research Project Proposal

4. Begin writing the dissertation more structured and refine the data that
has been collected with a deadline up to the end of March.
5. Final corrections, reference checking and formatting the dissertation up
to the final deadline in April.

Resources
 Machine Learning sources, such as journals, the university’s library etc.
 Data Collection from publicly accessible datasets.
 Other academic books that help with Data Processing and Analysis in
Python environment.

References
Bishop, C. M., 2006. Pattern Recognition and. s.l.:Springer Science+Business Media, LLC.

Bonaccorso, G., 2018. Machine Learning Algorithms. 2nd ed. s.l.:Packt.

Boutaba, R., 2018. A comprehensive survey on machine learning for networking. Journal of
Internet Services, 9(16).

Géron, A., 2019. Hands-on Machine Learning with Scikit-Learn, Keras & TensorFlow.
s.l.:O'Reilly Media, Inc..

Omar, S., 2013. Machine Learning Techniques for Anomaly Detection:. International Journal
of Computer Applications, 79(2).

Patel, S., 2018. Network Traffic Classification Analysis Using Machine Learning Algorithms.
Greater Noida, India, IEEE.

Rentao Gu, 2020. Machine learning for intelligent optical networks: A comprehensive survey.
Journal of Network and Computer Applications, Volume 157.

Shalev-shwartz, S., 2014. Understanding Machine Learning: From theory to Algorithms.


s.l.:Cambridge University Press.

Tawsook, K. L. a. T., 2012. Network traffic anomaly detection using machine learning
approaches. Maui, HI, USA, IEEE.

You might also like