VISVESVARAYA TECHNOLOGICAL UNIVERSITY

BELAGAVI, KARNATAKA

A Seminar Report on
“AI FIREWALLS”
Submitted in the partial fulfillment for the requirements for the conferment of Degree of
BACHELOR OF ENGINEERING
in

INFORMATION SCIENCE AND ENGINEERING

By
Ms. Pallavi A USN:1BY21IS410

Under the guidance of

Dr. Srinivas B V
Assistant Professor, BMSIT&M

2023-2024
 VISVESVARAYA TECHNOLOGICAL UNIVERSITY
BELAGAVI, KARNATAKA
BMS INSTITUTE OF TECHNOLOGY & MANAGEMENT
YELAHANKA, BENGALURU-560064

DEPARTMENT OF INFORMATION SCIENCE & ENGINEERING

CERTIFICATE

This is to certify that the Seminar (18ISS86) entitled AI FIREWALLS is a bonafide work carried
out by Ms. Pallavi A (1BY21IS410) partial fulfillment for the award of Bachelor of Engineering
Degree in Information Science and Engineering of the Visvesvaraya Technological University,
Belagavi during the year 2023-2024. It is certified that all corrections/suggestions indicated for
Internal Assessment have been incorporated in this report. The seminar report has been approved as
it satisfies the academic requirements with respect to seminar work for the B.E Degree.

Signature of the Guide Signature of the HOD

Dr. Srinivas B V Dr. Pushpa S K

Signature of the Coordinator

Dr. Drakshaveni G/Dr. Swetha M S/Mrs. Bhavya G
 ACKNOWLEDGEMENT

I am happy to present this Technical Seminar after completing it successfully. This seminar
would not have been possible without the guidance, assistance and suggestions of many
individuals. I would like to express our deep sense of gratitude and indebtedness to each and
every one who has helped us make this project a success.

I heartily thank our Principal, Dr. Sanjay H A, BMS Institute of Technology &
Management for his constant encouragement and inspiration in taking up this seminar.

I heartily thank our Head of Department Dr. Pushpa S K, Dept. of Information Science
and Engineering, BMS Institute of Technology & Management for her constant
encouragement and inspiration in taking up this seminar.

I heartily thank our seminar coordinator Dr. Drakshaveni G/Dr. Swetha M S/Mrs. Bhavya
G, Professor, Dept. of Information science and Engineering, for her constant follow-up
and advice throughout the course of the Project work.

I gracefully thank our seminar guide, Dr. Srinivas B V, Assistance Professor, Dept. of
Information Science and Engineering, for her encouragement and advice throughout the
course of the seminar.

Special thanks to all the staff members of Information Science Department for their help and
kind co-operation.

Lastly, I thank our parents and friends for their encouragement and support given to us in
order to finish this precious work.

By,
Pallavi A

i
 BMS INSTITUTE OF TECHNOLOGY & MANAGEMENT YELAHANKA,
BANGALORE-64

DEPARTMENT OF INFORMATION SCIENCE AND ENGINEERING

Declaration

I, hereby declare that the Technical Seminar titled AI FIREWALLS Is a record of original
project work undertaken for the award of the degree Bachelor of Engineering in Information
Science and Engineering of the Visvesvaraya Technological University, Belagavi during the
year 2023-2024. I have completed this project under the guidance of Dr. Srinivas B V.

I also declare that this project report has not been submitted for the award of any degree,
diploma, associate ship, fellowship or other title anywhere else.

Student Photo

USN 1BY21IS410

Name Pallavi A

Signature

ii
 ABSTRACT

Attacks on web applications and web-based services were conducted using Hyper-Text
Transfer Protocol (HTTP), which is also used as the communication protocol of web- based
applications. Due to the dynamic structure of web applications and the fact that they have
many variables, detection and prevention of web-based attacks are made more difficult. In
this study, a hybrid learning-based web application firewall (WAF) model is proposed to
prevent web-based attacks, by using signature-based detection (SBD) and anomaly-based
detection (ABD). Detection of known web-based attacks is done by using SBD, while
detection of anomaly HTTP requests is done by using ABD. Learning-based ABD is
implemented by using Artificial Neural Networks (ANN). Thus, an adaptation of the model
against zero-day attacks is ensured by learning-based ABD by using ANN. The proposed
model is tested by using WAF 2015, CSIC 2010 and ECML-PKDD datasets which are
open-source datasets. According to the test results, a high mean achievement percentage
(96.59 %) was obtained. Detection results are also compared to previous studies. After
comparison, the proposed model promises higher performance than what the existing
studies until now have to offer.

iii
 INDEX

ACKNOWLEDGEMENT……………………………… i

DECLARATION………………………………………... ii

ABSTRACT……………………………………………... iii

LIST OF FIGURES……………………………………... v

Chapter Title
Chapter No. Page No.

1 Introduction [1-3]
1.2 Motivation 3
1.3 Objective 3
2 Literature Survey [4-8]
2.1 Existing system 7
2.2 Problem Statement 8
2.3 Proposed System 8
3 Requirement Specification [9]
3.1 Functional Requirements 9
3.2 Non-Functional Requirements 9
3.3 Software & hardware Requirements 9
4 Design and Analysis [10-13]
4.1 Design 10

5 Implementation [14-16]
6 Future Scope [17-17]
7 Results [18-19]
8 Application [20-21]
9 Conclusion [22-22]
References 23
 LIST OF FIGURES
Figure No. Figure name Page No.
1 Data Flow Diagram 9
2 Categories 10
3 Implementation 11-12
4 User dashboard 14

v
 AI Firewalls
CHAPTER I
INTRODUCTION
Today’s networks change and develop on a regular basis to adapt to new business situations,
such as reorganizations, acquisitions, outsourcing, mergers, joint ventures, and strategic
partnerships and the increasing degree to which internal networks are connected to the Internet.
The increased complexity and openness of the network thus caused makes the question of
security more complicated than hitherto and necessitates the development of sophisticated
security technologies at the interface between networks of different security domains, such as
between Intranet and Internet or Extranet. The best way of ensuring interface security is the use
of a firewall. A firewall is a network security system designed to prevent unauthorized access to
or from a private network. Firewalls can be implemented as both hardware and software, or a
combination of both. Network firewalls are frequentlyused to prevent unauthorized Internet users
from accessing private networks connected to the packets, based on information that is available
in packet headers, such as protocol numbers, source and destination IP addresses and port
numbers, connection flags, and eventually some other IP options.

Internet, especially intranets. All messages entering or leaving the intranet pass through the
firewall, which examines each message and blocks those that do not meet the specified security
criteria. A Firewall can also be called a computer, router or other communication device that
filters access to the protected network.

Types of Firewalls
There are three types of firewalls. They are-
(1) Stateless (packet-filtering) firewall
(2) Stateful firewall
(3) Application layer (Proxy) firewall

Stateless Firewall:
A stateless firewall monitors each packet individually and isolate. It allows or denies
packets without knowing the packets content and connection state. A packet-filtering

Dept. of ISE, BMSIT&M 2023-2024 1

AI Firewalls

firewall gives permission to packet for passing through by checking its source and
destination address, protocol and destination port number. If these are not match with the
firewall rules then the packet is dropped.

Stateful firewall:
This firewall is more effective than the stateless firewall. It consists a list of all trusted
connection that are already established. When firewall gets a new packet, it is checked with
the list. If it is matched then the packet is through without further checking. If it is not
matched, the packet is sent for checked with the initial rules for new connection.

Application layer firewall:

Application layer or proxy firewall examines the packet at the application layer acting as an
intermediary between the client and the server. This firewall examines the entire network
packet rather than just the network address and the port number. In case of outgoing, server
allows most of the packets, because the server is usually trustworthy to itself. Still the
outgoing rule set in a way that can be used to prevent the server from unwanted
communication or malicious executable attackers.

Pros and Cons of these above-mentioned firewall:

 Attacks that are prepared by the process itself, cannot be handled.

 If an unauthorized user already gained access, this system is ineffective to work on this.
 In stateless filtering, there is needed more time to establish or drop connection because it
checks each packet individually.
 As it checks some definite number of port number and destination number without
knowing the content of the packet, it is not suitable for all trusted packet if it does not
consist those numbers.
 In stateful filtering, there is no rule for new packets.
 As intermediary layer, proxy firewall is always slow and time consuming.
 Sometimes, these processes drop some trusted packet.

Dept. of ISE, BMSIT&M 2023-2024 2

AI Firewalls

1.2 MOTIVATION

The motivation behind AI firewalls is like having a super-smart guard protecting your digital
home. Imagine your computer network is a house, and the internet is like a bustling street outside.
Just like you'd want a vigilant guard to keep out intruders and keep your family safe, AI firewalls
use advanced technology to watch over your network. They can spot sneaky cyber bad guys trying
to break in, even if they're using new tricks. With AI, these firewalls learn and get better at their
job over time, helping to keep your digital world secure from online threats.

1.3 OBJECTIVE

The objective of AI firewalls is to make our digital world safer by using smart technology to
protect computer networks from cyber threats. These firewalls aim to be like digital security
guards, constantly watching over our networks to keep out malicious attackers and unwanted
visitors. They use artificial intelligence, which is like teaching computers to think and learn,
to detect and stop cyber threats in real-time.

These smart firewalls have a few key goals:

1. Advanced Threat Detection: They aim to detect and block all kinds of cyber threats,
from viruses and malware to hackers trying to steal sensitive information.

2. Fast Response Time: AI firewalls need to react quickly to threats, stopping them in
their tracks before they can cause harm.

3. Reducing False Alarms: They work to minimize false alarms, so they don't bother us
with unnecessary alerts while still catching the real dangers.

4. Continuous Learning: These firewalls keep learning and getting smarter over time,
adapting to new types of threats and staying one step ahead of cyber attackers.

. With the help of Cloud technology, we ensure smooth operability and reliability.

Dept. of ISE, BMSIT&M 2023-2024 3

AI Firewalls

CHAPTER II

LITERATURE SURVEY
After a thorough search and evaluation of the available literature in the given project it has
been selected and enhanced in the particular area. The literature review of the documents that
support this system has been represented below.

1. Title: "A Survey on Artificial Intelligence Techniques for Cybersecurity"

 Authors: Md. Nasim Adnan, Fahim Ahmed Touqir, et al.
 Publication: 2020 IEEE 4th International Conference on Intelligent Computing
and Control Systems (ICICCS)
This survey comprehensively reviews the application of artificial intelligence
techniques, including machine learning and deep learning, in cybersecurity, covering
areas such as intrusion detection, malware analysis, and network security. It discusses
the challenges and opportunities of applying AI in cybersecurity and provides insights
into future research directions to enhance cyber defence mechanisms.
This paper presents a comprehensive survey on the utilization of artificial intelligence
(AI) techniques, encompassing machine learning and deep learning, within the realm
of cybersecurity. It examines the application of AI in various domains such as
intrusion detection, malware analysis, and network security. Through an exploration
of challenges and opportunities, the paper offers valuable insights into the integration
of AI into cybersecurity frameworks. Additionally, it outlines potential avenues for
future research to bolster cyber defense mechanisms.

2. Title: "A Survey on Deep Learning in Network Security"

 Authors: Alireza Bahramali, Hamed Mohammadi, et al.
 Publication: 2019 IEEE 17th International Conference on Software Engineering
Research, Management and Applications (SERA)

Dept. of ISE, BMSIT&M 2023-2024 4

AI Firewalls

Focusing on deep learning applications in network security, this survey explores

various deep learning models, including convolutional neural networks (CNNs) and
recurrent neural networks (RNNs), for tasks such as intrusion detection and malware
classification. It evaluates the performance and effectiveness of deep learning
approaches in detecting cyber threats and discusses open challenges and future
research directions in this field.
This paper presents a comprehensive survey on the utilization of deep learning
techniques in the domain of network security. It investigates the application of various
deep learning models like convolutional neural networks (CNNs) and recurrent neural
networks (RNNs) for tasks such as intrusion detection and malware classification. The
survey assesses the efficacy and performance of deep learning methodologies in
identifying cyber threats and addresses the existing challenges and potential research
avenues in this evolving field.

3. Title: "Deep Learning for Cybersecurity: A Survey"

 Authors: Ching-Hao Mao, Chun-Wei Tsai, et al.
 Publication: 2019 IEEE Access
This survey provides an overview of deep learning techniques applied to various
cybersecurity tasks, including malware detection, phishing detection, and network
intrusion detection. It reviews different deep learning architectures and methodologies
used in cybersecurity applications and discusses the advantages and limitations of
deep learning approaches. The survey also identifies research gaps and future
directions for leveraging deep learning in cybersecurity.
This paper presents a comprehensive survey of deep learning methods employed in
cybersecurity domains. Focusing on tasks such as malware detection, phishing
detection, and network intrusion detection, the survey examines various deep learning
architectures and methodologies utilized in cybersecurity applications. It evaluates the
strengths and weaknesses of deep learning approaches, identifies research gaps, and
outlines future directions for enhancing cybersecurity through deep learning
techniques.

Dept. of ISE, BMSIT&M 2023-2024 5

AI Firewalls

4. Title: "Deep Learning for Intrusion Detection in IoT Networks: A Comprehensive Survey"
 Authors: Muhammad Usama, Shamaila Ijaz, et al.
 Publication: 2019 IEEE Access
Focusing on intrusion detection in IoT networks, this survey explores the application
of deep learning techniques for detecting cyber threats in IoT environments. It reviews
different deep learning models, including deep belief networks (DBNs) and long
short-term memory (LSTM) networks, and evaluates their performance in detecting
IoT-specific attacks. The survey highlights the challenges and opportunities of using
deep learning for IoT security and discusses future research directions in this area.
This paper presents a comprehensive survey on the utilization of deep learning
techniques for intrusion detection in Internet of Things (IoT) networks. With a specific
focus on IoT environments, the survey investigates the efficacy of various deep
learning models, such as deep belief networks (DBNs) and long short-term memory
(LSTM) networks, in identifying cyber threats. Through a critical analysis of their
performance in detecting IoT-specific attacks, the survey sheds light on the potential
of deep learning in bolstering IoT security. Moreover, it elucidates the existing
challenges and opportunities in deploying deep learning for intrusion detection in IoT
networks and delineates future research avenues in this domain.

5. Title: "A Survey on Deep Learning Techniques for IoT Security"

 Authors: Amjad Gawanmeh, Abdelrahman Elfaki, et al.
 Publication: 2019 IEEE Access
This survey investigates the application of deep learning techniques for enhancing the
security of IoT systems, covering areas such as anomaly detection, malware detection,
and intrusion detection. It reviews various deep learning architectures and
methodologies used in IoT security applications and discusses their effectiveness in
mitigating IoT-specific threats. The survey also addresses the challenges and future
research directions in leveraging deep learning for IoT security.
This paper presents a comprehensive survey on the utilization of deep learning

Dept. of ISE, BMSIT&M 2023-2024 6

AI Firewalls

techniques to bolster the security of Internet of Things (IoT) systems. Focusing on

pivotal areas including anomaly detection, malware detection, and intrusion detection,
it explores diverse deep learning architectures and methodologies applied in IoT
security contexts. Evaluating the efficacy of these techniques in countering IoT-
specific threats, the survey also delves into the existing challenges and outlines future
research directions for harnessing deep learning in IoT security.

6. Title: "A Survey of Machine Learning for Big Data Processing in Cloud Computing"
 Authors: Qiang He, Zhihui Du, et al.
 Publication: 2019 IEEE Access
Focusing on machine learning applications in cloud computing, this survey explores
the use of machine learning techniques for big data processing, resource management,
and security in cloud environments. It reviews different machine learning algorithms
and frameworks used in cloud computing applications and discusses their performance
and scalability. The survey also identifies challenges and future research directions in
leveraging machine learning for big data processing in the cloud.
This paper presents a comprehensive survey of machine learning applications within
the realm of cloud computing. It delves into the utilization of machine learning
techniques for big data processing, resource management, and security within cloud
environments. By reviewing various machine learning algorithms and frameworks
employed in cloud computing applications, the paper assesses their performance and
scalability. Moreover, it identifies existing challenges and outlines future research
directions for leveraging machine learning in big data processing within the cloud.

Dept. of ISE, BMSIT&M 2023-2024 7

AI Firewalls

2.1 EXISTING SYSTEM

• A lot of time is wasted by a doctor to experience the patient's whole clinical history,
which rather might have been used for better treatment of a patient
• In Existing System, the greatest difficulties are security and interoperability, which
makes keeping up and sharing of these records a difficult task.

2.2 PROBLEM STATEMENT

Despite advancements, traditional firewalls struggle to detect sophisticated cyber threats

effectively. Existing solutions often generate high false positives and lack adaptability to
evolving attack techniques. Manual intervention in threat response workflows leads to delays
and inefficiencies. Integration challenges hinder seamless collaboration with other security
technologies. Scalability issues arise in handling the growing volume and complexity of
network traffic.

2.3 PROPOSED SYSTEM

 Implement an AI-driven firewall solution integrating advanced machine learning algorithms for
dynamic threat detection and prevention.
 Develop a system that integrates adaptive response mechanisms, allowing the firewall to
dynamically adjust security policies and countermeasures based on the detected threats and
evolving network conditions.

Dept. of ISE, BMSIT&M 2023-2024 8

AI Firewalls

CHAPTER III
SOFTWARE REQUIREMENT SPECIFICATION
3.1 FUNCTIONAL REQUIREMENTS
 Utilize advanced machine learning algorithms to analyze network traffic patterns and identify
indicators of compromise
 Identify anomalous activities that may indicate potential security threats, such as unauthorized
access attempts or data exfiltration.
 Automatically respond to security incidents based on predefined policies and threat
intelligence.
 Automatically respond to security incidents based on predefined policies and threat
intelligence.
 Protect sensitive data and ensure compliance with privacy regulations.

3.2 NON - FUNCTIONAL REQUIREMENTS

 Performance: Efficiently process and analyse large volumes of network traffic in real-time.
 Scalability: Scale horizontally to accommodate growing network traffic and expanding
infrastructure.
 Reliability: Maintain high availability and uptime to ensure continuous protection against
cyber threats.
 Security: Ensure the confidentiality, integrity, and availability of sensitive data and security
configurations.
 Usability: Provide a user-friendly interface for configuration, management, and monitoring of
firewall policies and alerts.

3.3 SOFTWARE REQUIREMENTS & HARDWARE REQUIREMENTS

 Operating System
 Firewall Software
 Machine Learning Frameworks
 Logging and Monitoring Tools

Dept. of ISE, BMSIT&M 2023-2024 9

AI Firewalls

CHAPTER IV

SYSTEM ARCHITECTURE

1.1 DESIGN
Data Flow Diagram.
First, researchers design a Data flow diagram of Packet Filtering Prediction System (fig
10), which is a preliminary step to create an overview of the system without going into
great detail, which can be later elaborated. It generally consists of overall application data
flow and prediction process. Packet Filtering Prediction System shows functionalities and
details of data flow of the system. It also shows low level functionalities such as how a user
can get a specific packet from server. The below diagram (fig 10) is used to visualize of
flow of packet and check with those rules.

Fig. 1. Data Flow Diagram

Dept. of ISE, BMSIT&M 2023-2024 10

AI Firewalls

Work Flow

The process of the proposed system has the minimum possibility of packet drops and can
deeply identify a packet is really containing rejected contents or not. That can free this system
from risks. The workflow of the system is discussed in following subsections:
Make up list categories for incoming packets. Here, Firewall list up the connection of packets
into three categories (fig 4). Established list contains the connections of trusted packets. Deny
list contains the connections are blocked. Third list is additional list containing the connections
of packets are not sure about those are safe or not.

Figure 2. List Categories

Ready for checking.
A firewall normally set connection of a packet to the established list, if anyhow it entered
into the own system. If that packet has risk materials, then it became unable to detect for a
traditional firewall. To remove this risk condition this firewall always continue an enquiry to
check established connections (fig 5) are trusted or not. Here shown some rules are
produced for exceptional packets by the system itself according to a packet.

Dept. of ISE, BMSIT&M 2023-2024 11

AI Firewalls

CHAPTER V

IMPLEMENTATION
The implementation phase constructs, installs and operates the new system. The most
crucial stage in achieving a new successful system is that it will work efficiently and
effectively. The interfaces of the system are discussed in following subsections.

Home

This is the main UI of the application. All functionality can be accessed from here.

Fig. 3. Home Page

Rules
The list of all rules can be found here. Author can add or remove any rule

Fig. 4. Rules Section

Dept. of ISE, BMSIT&M 2023-2024 12

AI Firewalls

Load Existing
Rules Users can load rules from an external source, shown in fig 13

Fig. 5. Load Existing Rules

Start and Stop
After selecting manual rules, user can start applying that rules or stop the started process

(fig 14).
Fig. 6. Start and Stop part

Dept. of ISE, BMSIT&M 2023-2024 13

AI Firewalls

CHAPTER VI
FUTURE SCOPE
 AI firewalls will evolve to better detect and mitigate emerging cyber threats, leveraging
advanced machine learning algorithms.

 They will increasingly incorporate behavioral analysis and anomaly detection to proactively
identify abnormal patterns within network traffic.

 Future AI firewalls will feature enhanced automation and orchestration capabilities to

streamline threat response workflows.

 They will provide more dynamic and adaptive security policies based on real-time analysis
of network conditions and threat intelligence.

 AI firewalls will provide native support for cloud services and edge computing
architectures, ensuring consistent security across distributed environments.

 They will integrate privacy-preserving technologies to protect sensitive data while still
enabling effective threat detection and prevention.

 Future AI firewalls will further integrate with external threat intelligence feeds and
collaborative defense initiatives to enhance their threat detection capabilities.

Dept. of ISE, BMSIT&M 2023-2024 14

AI Firewalls

CHAPTER VI
RESULTS
After starting the firewall, user can check by windows command prompt that the packet
transferring is actually stopped. Result part is shown in fig.

Fig.6. Results

Dept. of ISE, BMSIT&M 2023-2024 15

AI Firewalls

CHAPTER VII

APPLICATIONS

 Advanced Threat Detection and Prevention:

AI firewalls can analyze network traffic patterns and identify anomalies indicative of
potential security threats, such as malware infections, intrusion attempts, and data exfiltration.
By leveraging machine learning algorithms, AI firewalls can detect and mitigate advanced
threats in real-time, including zero-day exploits and polymorphic malware, that may evade
traditional security measures.

 Insider Threat Monitoring:

Insider threats, whether intentional or unintentional, pose a significant risk to organizations'
cybersecurity. AI firewalls can monitor user behavior and identify anomalous activities that
may indicate insider threats, such as unauthorized access to sensitive data or abnormal data
transfer patterns.

 Cloud Security:
With the increasing adoption of cloud services and infrastructure, organizations require robust
security solutions to protect their cloud-based assets and data. AI firewalls can provide cloud-
native security capabilities, such as monitoring network traffic between cloud instances,
detecting and preventing unauthorized access to cloud resources, and enforcing security
policies across hybrid and multi-cloud environments.

 Zero Trust Network Access:

Zero Trust Network Access is an approach to network security that assumes no trust, even
within the internal network, and requires strict access controls and authentication mechanisms
for all users and devices. AI firewalls can play a key role in implementing by providing
dynamic access controls based on user identity, device posture, and contextual information.

Dept. of ISE, BMSIT&M 2023-2024 16

AI Firewalls

CHAPTER VIII
CONCLUSION

In this work, researchers have come up with a model to find the real knowledge of
upcoming packets based on AI rules. As per this model, it was concluded that one factor was
content checking which determined the success rating of safe packet acceptation. There are
several aspects that can be given serious consideration for future work in this research such
as removing rule redundancy and reducing the number of rules in firewall policies by
combining two or more of them. It will be a great area of further work.

Dept. of ISE, BMSIT&M 2023-2024 17

AI Firewalls

REFERENCES

[1] Zhijia Wang, Changsha Research on Feature and Architecture Design of AI Firewall
2021 5th Annual International Conference on Data Science and Business Analytics
(ICDSBA) IEEE 10.1109/ICDSBA53075.2021.00024

[2] Mahoney, M.—Chan, P.: Learning Models of Network Traffic for Detecting Novel
Attacks. Technical report, Florida Institute of Technology 2019.

[3] Mukkamala, S.—Janowski, G.—Sung, A. H.: Intrusion Detection Using Neural

Networks and Support Vector Machines. Heidelberg: Physical/Springer 2020, pp. 121– 138.

[4] Gagnon, : Using Artificial Intelligence for Intrusion Detection. In Proceeding of the
2007 Conference on Emerging Artificial Intelligence Applications in Computer
Engineering, Amsterdam, The Netherlands 2019, pp. 295–306.

[5] G. EL HAJAL, R. ABI ZEID DAOU and Y. DUCQ, "A novel approach to classify
vulnerabilities based on authenticated measurements", iCatse International Conference on IT
Convergence and Security, 2021.

[6] Georges El Hajal U niv. Bordeaux, IMS laboratory, UMR 5218 CNRS, Talence, France;
Roy Abi Zeid Daou; Yves Ducq.: A Comparative Study of Anomaly Detection Schemes in
Network Intrusion Detection. In Proceedings of the Third SIAM International Conference
on Data Mining 2021, pp. 25–36.

Dept. of ISE, BMSIT&M 2023-2024 18