Authors

  • Sarvar Norboboyevich Tashev
    Shakhrisabz Branch of Tashkent Institute of Chemical Technology Address: 20 Shakhrisabz st.,181310, Shakhrisabz city, Republic of Uzbekistan

DOI:

https://doi.org/10.37547/ajast/Volume04Issue10-09

Keywords:

Network protocols vulnerability signatures traffic filtering

Abstract

This article details the importance of assessing the vulnerabilities of network protocols and systems and the steps required to implement them. The article emphasizes that the main goal of vulnerability research is the timely identification of information security problems and their elimination. At the same time, the dependence of the main indicators of network efficiency on the protocols used in it is also scientifically   substantiated.  Information security issues were  analyzed by integrating metrics obtained from assessing the impact of attacks in various areas of network packets into network monitoring systems. The article also provides an in-depth analysis of the advantages and disadvantages of network protocols, and also examines each of them from the point of view of information security.


background image

Volume 04 Issue 10-2024

55


American Journal Of Applied Science And Technology
(ISSN

2771-2745)

VOLUME

04

ISSUE

10

Pages:

55-64

OCLC

1121105677
















































Publisher:

Oscar Publishing Services

Servi

ABSTRACT

This article details the importance of assessing the vulnerabilities of network protocols and systems and the steps
required to implement them. The article emphasizes that the main goal of vulnerability research is the timely
identification of information security problems and their elimination. At the same time, the dependence of the main
indicators of network efficiency on the protocols used in it is also scientifically substantiated. Information security
issues were analyzed by integrating metrics obtained from assessing the impact of attacks in various areas of network
packets into network monitoring systems. The article also provides an in-depth analysis of the advantages and
disadvantages of network protocols, and also examines each of them from the point of view of information security.

KEYWORDS

Network protocols, vulnerability signatures, traffic filtering, systems pose, serious security threat, unauthorized
access.

INTRODUCTION

Network protocols are the basis for data transmission
in computer networks. They define the rules and data
formats used for communication between computers.
There are different network protocols, each of which is
designed for a specific type of data transfer. For

example, TCP/IP is one of the most common protocols
used on the Internet.

However, although network protocols provide
efficient communication between devices, they can be
vulnerable to attacks.

Research Article

ANALYSIS OF WEAKNESSES IN NETWORK PROTOCOLS AND SYSTEMS

Submission Date:

October 12, 2024,

Accepted Date:

October 17, 2024,

Published Date:

October 22, 2024

Crossref doi:

https://doi.org/10.37547/ajast/Volume04Issue10-09

Sarvar Norboboyevich Tashev

Shakhrisabz Branch of Tashkent Institute of Chemical Technology
Address: 20 Shakhrisabz st.,181310, Shakhrisabz city, Republic of Uzbekistan




Journal

Website:

https://theusajournals.
com/index.php/ajast

Copyright:

Original

content from this work
may be used under the
terms of the creative
commons

attributes

4.0 licence.


background image

Volume 04 Issue 10-2024

56


American Journal Of Applied Science And Technology
(ISSN

2771-2745)

VOLUME

04

ISSUE

10

Pages:

55-64

OCLC

1121105677
















































Publisher:

Oscar Publishing Services

Servi

Hackers can use various methods such as buffer
overflow to infiltrate the system and gain
unauthorized access to sensitive information.

To protect against such attacks, comprehensive
security measures must be implemented. This includes
using strong passwords, regularly updating software
and operating systems, and using anti-virus software.
It's also important to stay up-to-date on cybersecurity
updates and stay up-to-date.

1.1 Vulnerabilities in network protocols and systems

Vulnerabilities in network protocols and systems pose
a serious security threat. They can be used by attackers
to gain unauthorized access to data, compromise
systems, or perform other attacks.

Vulnerability research is an important task that allows
you to identify and fix security problems in a timely
manner. There are many ways to test for
vulnerabilities, which can be divided into two main
categories: static and dynamic.

Static methods Vulnerability research is based on the
analysis of the program's source code and protocols.
They allow you to detect potential security issues
without having to execute programs or protocols in
real-time.

The main static methods of vulnerability research are:

Source code analysis is a technique that involves
manually or automatically analyzing the source code of
programs and protocols for potential security
problems.

Analysis of vulnerability signatures is a method based
on the use of vulnerability signature databases. A
vulnerability signature database contains information
about known security issues that can be found in
software code.

Analysis Static Analyzer is a tool that automatically
analyzes the source code of programs and protocols
for potential security issues [1].

Dynamic methods of vulnerability research are based
on the real-time execution of programs and protocols.
They allow you to identify security problems that may
arise as a result of the interaction of various application
components and protocols.

The main dynamic methods of vulnerability research
are:

Penetration testing (Penetration testing) is a method
that involves simulating the actions of an intruder to
identify vulnerabilities in a system. Permissible
attempts are made to bypass existing security
measures from the position of a potential attacker,
possible scenarios for entering the corporate network
and achieving test goals are identified (for example,
obtaining

access

rights,

stealing

confidential

information, information making changes to the
systems, breaking the system, the operation of
individual network components and security systems
or business processes). The main stages of the work
are shown in Figure 1.1.


background image

Volume 04 Issue 10-2024

57


American Journal Of Applied Science And Technology
(ISSN

2771-2745)

VOLUME

04

ISSUE

10

Pages:

55-64

OCLC

1121105677
















































Publisher:

Oscar Publishing Services

Servi

Figure 1.1. The main stages of work

A test direction based on sociotechnical methods is shown in Figure 1.2.

Figure 1.2. Test direction based on social technical methods


background image

Volume 04 Issue 10-2024

58


American Journal Of Applied Science And Technology
(ISSN

2771-2745)

VOLUME

04

ISSUE

10

Pages:

55-64

OCLC

1121105677
















































Publisher:

Oscar Publishing Services

Servi

Attempts are made to allow unauthorized access to
the target organization's corporate network and
protected assets using social engineering techniques
and the use of "human factors". The methods, as a rule,
are aimed at users of the end systems, and allow to
determine the reaction of employees in various normal
and emergency situations, the level of awareness and
knowledge of employees about security requirements
[3].

Description of some examples of socio-technical
methods:

Phishing. Creating a fake web page for a legitimate
service (such as a remote mail access page) and
encouraging users to enter sensitive information (such
as passwords) into it.

Trojan horse. Sending emails to users with malicious
attachments and encouraging them to open them
through targeted cover letters, file names, callbacks,
and other approaches

Pretexting. Modeling a specific scenario, which
involves gaining the user's trust, in order to encourage
the user to take a certain action (by gathering
information in advance about the organization and
individual employees and their areas of responsibility).
For example, calling users under the guise of potential
trusted individuals (external or internal) to obtain
confidential information.

Don't

go.

Leaving

infected

media

with

logos/tags/filenames that encourage you to run them
in public areas of the organization (elevator, kitchen,
parking lot).

"Quid pro quo." Calls support service users with
messages about problems with their PCs and helps
them solve them.

'Reverse' Social Engineering'. Sending emails to users
from trusted addresses with "support" contacts,
causing problems with computers and waiting for calls
/ emails to solve problems, where necessary
information can be obtained.

Fuzzing is a technique that involves passing random or
specially selected data to program input to detect
security problems. In network protocols, ambiguity is
an effective way to identify vulnerabilities due to errors
in the code, interactions between different protocol
components, or mishandling with incorrect data.
Fuzzing can be used to test software, network
protocols, systems, and devices.

Log analysis is a technique that involves analyzing
system event logs to identify security issues.

Selection of methods of vulnerability study. The choice
of vulnerability scanning methods depends on various
factors, such as the complexity of the system, the type
of vulnerabilities to be detected, and the available
resources.

In general, static vulnerability scanning techniques are
more effective at identifying security problems that
involve bugs in code. Dynamic vulnerability scanning
methods are more effective in identifying security
issues related to the interaction of different system
components.

II. Analysis of traffic filtering and application efficiency
approaches

The black and white list system is one of the most
promising and widespread methods of automatically
identifying the type of content in incoming traffic.
Blacklists and whitelists are a set of rules for classifying
data as "trusted" or "untrusted" within data packets,
by which incoming content is filtered. Due to the
development of inappropriate content, blacklists and


background image

Volume 04 Issue 10-2024

59


American Journal Of Applied Science And Technology
(ISSN

2771-2745)

VOLUME

04

ISSUE

10

Pages:

55-64

OCLC

1121105677
















































Publisher:

Oscar Publishing Services

Servi

whitelists must be updated, modified and/or deleted
on an ongoing basis. A content filter is defined as a
program that blocks access to websites with
prohibited content. Content filtering allows you to
comply with the requirements of applicable laws and
easily protect users from unwanted information such
as extremist sites and resources [2].

Traditionally, content filters can be divided into two
types: products installed on the protected devices
themselves (host devices) and products that filter
network gateways and traffic passing through them.

The first type of product is mainly used in home and
personal devices. The second type of product is used in
Internet gateways with content filtering functions.
They help reduce the risks associated with Internet
browsing by filtering out dangerous or unwanted
content, malicious files, social media and other web
threats, as well as monitor users' web pages and
optimize the use of communication channels.

III. Formation of lists

The standard that describes the technical aspects of
blocking and filtering Internet services is called
RFC7754. This paper examines several technical
approaches to Internet blocking and filtering as they
relate to the overall Internet architecture.

Filtering is done in two ways:

work according to the "black" list: ALLOW

everything else;

work according to the "white" list: PROHIBIT

everything else.

Whitelists block all but permitted content. Blacklists
allow access to all content except what is expressly
prohibited. Custom blacklists deny access and
whitelists allow access to websites. Maintaining a
blacklist requires constant updating. You must
manually update records and remove records when
removing unwanted content. When using a whitelist,
you must monitor the availability of allowed resources
and add new ones. Often, whitelist updates do not
keep up with the needs of network users for new
resources. Using whitelists and blacklists, you can
specify which resources are queried and which
resources affect the performance and status of
application tests. Whitelists and blacklists are only
available for web and script tests. The blacklist always
overrides the whitelist when determining allowed or
blocked positions. Table 1.1 describes the filtering and
blocking mode for all scenarios, including whitelists
and blacklists.

Table

Filtering and blocking mode for whitelists and blacklists

6

White list

Mode

Code

Empty

Empty

Allow access

Filtering rules are not specified

Empty

The URL does not
match an entry in the
list

Block access

The URL is not whitelisted

Empty

The URL matches
the list entry

Allow access

The URL is whitelisted.
There are no entries in the blacklist


background image

Volume 04 Issue 10-2024

60


American Journal Of Applied Science And Technology
(ISSN

2771-2745)

VOLUME

04

ISSUE

10

Pages:

55-64

OCLC

1121105677
















































Publisher:

Oscar Publishing Services

Servi

to block access.

The URL does not
match an entry in the
list

Empty

Allow access

The URL is not blacklisted.
There are no whitelist entries to
deny access to non-whitelisted
URLs.

The URL matches
the list entry

Empty

Block access

The URL is blacklisted

The URL does not
match an entry in the
list

The URL does not
match an entry in the
list

Block access

The URL is not whitelisted

The URL does not
match an entry in the
list

The URL matches
the list entry

Allow access

The URL is whitelisted. The URL is
not blacklisted

The URL matches
the list entry

The URL does not
match an entry in the
list

Block access

The URL is not whitelisted. The
URL is blacklisted

The URL matches
the list entry

The URL matches
the list entry

Block access

The URL is blacklisted. A blacklist
entry overrides a whitelist entry.

METHODS

A traffic filtering system is usually a set of rules - a set
of entries (URL addresses, IP addresses, TCP ports, etc.
When the next packet arrives from the external
network, the filtering system intercepts this packet,
checks its metadata and content for a match with one
of the entries in the rule set, and if a match is found,
forwards the packet to the local network or drops, the
local network does not even know about the existence
of such a packet. At the same time, so that the user
understands that the traffic is blocked, the filtering
system sends to the local network, for example, a web
page with information about the reason for the block
(if the requested website if blocked) can transmit [5].

Threats are prevented with content filters

Phishing attacks are becoming more sophisticated and
social engineering techniques are being used. Phishing
sites usually have a short lifespan (5 days on average).

Modern browsers use several types of phishing
protection:

blacklists: when each new connection is opened, the

address of the resource being opened is passed to
servers with blacklists, where it is checked to see if the
address being opened is on the blacklist. If the address
is blacklisted, the browser will display an appropriate
error message.

white lists: the addresses opened from the white lists

are considered reliable, the remaining addresses are
checked using the black list.


background image

Volume 04 Issue 10-2024

61


American Journal Of Applied Science And Technology
(ISSN

2771-2745)

VOLUME

04

ISSUE

10

Pages:

55-64

OCLC

1121105677
















































Publisher:

Oscar Publishing Services

Servi

Both blacklists and whitelists contain a large number of
entries, which primarily causes additional problems
with the performance of user computers. In fact, you
cannot download the entire blacklist to the user's
computer. First, it contains a lot of records and
downloading it will load the user's Internet channel a
lot. Second, lists are updated regularly - it's very
difficult to keep lists updated on remote computers.

Third, even if the user's computer has an up-to-date
blacklist, searching for an address in it takes an
unacceptably long time.

In addition to phishing, the concept of MITM is also
widespread. The term refers to a type of cyberattack in
which attackers intercept a conversation or data
transmission by eavesdropping or pretending to be a
legitimate participant. The goal of a MITM attack is to
obtain sensitive information such as bank account
details, bank card numbers or login details that can be
used to commit further crimes such as identity theft or
money laundering. A successful MITM attack involves
two distinct steps: capture and decryption [6].

Modern

browsers

solve

the

problems

of

completeness, relevance and responsiveness using
two approaches:

1. Whitelists, which are very small in size, are stored on
the user's computer as chunks of website address
hashes, which saves hard disk space and allows for
quick searches for matching addresses. Whitelists are
also easy to update due to their relatively small size.

2. Large blacklists are stored on specialized cloud
servers, which are designed for fast, uniform
processing of large amounts of data. Such systems are
called reputable.

Filter traffic by IP addresses

IP addresses of the fourth and sixth versions have a
hierarchical structure, that is, a node's address is
written in blocks of numbers that each specific
network administrator can use to group nodes
according to logical or physical criteria.

To filter traffic by IP address, the filtering system first
receives the packet and reads the information from the
IP packet header. This header always contains the
Source address and the destination address in an
unencrypted form, which always point to the original
source of the packet (unless a VPN or other tunnel is
used - in this case it is impossible to filter traffic using
standard means and the only way to limit. in this case
access to hosts is to restrict access to the VPN servers
themselves) [7].

A VPN connection is called a "tunnel" between the
user's computer and the server's computer. Each node
encrypts the data before entering the "tunnel". When
you connect to a VPN, the system detects the network
and initiates authentication.

Next, the server gives permission, that is, gives the
right to perform certain actions: read mail, surf the
Internet, etc. Once the connection is established, all
traffic is encrypted between the computer and the
server. The computer has an IP address provided by the
Internet service provider. This IP blocks access to some
sites. The VPN server changes the IP to itself. Already,
all data from the VPN server is transmitted to external
resources requested by the user, VPN opens access to
blocked resources, and as a result, many are willing to
put up with low Internet speed and possible
application logs. Although VPN uses very strong
encryption algorithms, running a VPN client on a PC
does not guarantee 100% security of confidential data,
so you should choose a VPN provider carefully.

Filter traffic by TCP ports


background image

Volume 04 Issue 10-2024

62


American Journal Of Applied Science And Technology
(ISSN

2771-2745)

VOLUME

04

ISSUE

10

Pages:

55-64

OCLC

1121105677
















































Publisher:

Oscar Publishing Services

Servi

The principle of filtering by TCP ports is similar to
filtering by IP addresses, the difference is that TCP
ports are used to identify traffic carried in the transport
layer of the network model - application identifiers.
Accordingly, for this, the traffic filtering system must
also read the Destination port from the header.

A TCP packet enclosed in an IP packet. TCP packets are
also transmitted over the network in an unencrypted
form, which ensures the reliability of this filtering
method.

Filter traffic by URL

To implement this component, the device on which the
traffic filter is located must access a DNS server - a
server for converting domain names to IP addresses -
or be installed directly on a DNS server. This is because
packets on the Internet are identified only by IP
addresses, and URLs have precise translations to such
addresses and are created solely for the convenience
of the user.

Accordingly, when receiving a packet, the network
filter must determine the domain name from the
source IP address of the packet. After receiving a
domain name, the system can proceed by comparing
the received domain against a list of rules.

However, in general, the working principle of URL
filtering is more complicated than its predecessors.
This is because IP addresses do not always correspond
to actual domain names and vice versa. When a request
is sent to a website, there are virtual servers that
redirect the user to another server with a different
domain name [8]. This is usually done for load
balancing, but it complicates the work of the transport
filter

USING MACHINE LEARNING TO CLASSIFY TRAFFIC

The changing trends of network traffic, the
widespread use of encryption, the increase in the
speed of data transfer and, accordingly, the required
speed of their processing, the constant appearance of
new classes of traffic - all this required the emergence
of new methods of classification . For this purpose, the
creation of a set of distinguishing characteristics of
classes, based on the analysis of many examples of
these classes, machine learning methods were
proposed, which can significantly simplify work with
the automation of this process. In addition, most of the
proposed methods work with general properties of
streams rather than packet payloads, which solves the
problems of encryption and protection of user data. It
also gives an advantage in classification speed and
reduces the amount of memory required for decision
making. Next, machine learning techniques for
network traffic classification are discussed in detail: the
types of classification, the models and features used,
and the datasets on which the models are trained and
tested [1].

TYPES OF NETWORK TRAFFIC CLASSIFICATION

Network traffic can be segmented into individual
packets, and port-based and DPI-based classification
techniques can do the job, but most work is currently
done based on flow-based classification. There are
approaches where sender requests and receiver
responses are interpreted as two different counter-
flows, but a common solution is to combine these
flows into a single bi-directional flow. Since an Internet
stream usually contains a single complete session of
interaction between a sender and a receiver (client and
server), problems with classifying the entire stream as
a whole into a single class usually arise. does not come.
Since streams differ in their duration and the amount
of data transferred, the smallest and largest streams
are sometimes distinguished. Because of the


background image

Volume 04 Issue 10-2024

63


American Journal Of Applied Science And Technology
(ISSN

2771-2745)

VOLUME

04

ISSUE

10

Pages:

55-64

OCLC

1121105677
















































Publisher:

Oscar Publishing Services

Servi

significant difference in the size of these streams, the
classification results are sometimes examined
separately for the percentage of correctly classified
streams and the percentage of bytes classified. The
length of the stream is also of great importance.
Streams or stream fragments consisting of a small
number of packets may not contain enough
information to determine the class and therefore
require special treatment or are ignored. Traffic
classification can be done online, i.e. in real time, or
offline after the fact. The order of classification is
determined by the problem being solved. For example,
collecting network usage statistics for global
redistribution of resources, receiving information
about user activity for billing for Internet services and
recalculating these prices - all this does not require an
urgent response and is free of charge in any amount
mode can be processed. available information and

resources. Other tasks, such as ensuring the quality of
service to the user, detecting attacks and threats, and
immediately reallocating resources, require the fastest
response. In such cases, the classifier does not have the
opportunity to wait for the end of the flow and work
with the complete information about it, but is forced
to limit itself to only a part of the data, for example,
only the first N packets. flow the current classifier
model is also limited in that it must use system
resources sparingly and make flow decisions as quickly
as possible [9]. Also, in these cases, the decision is
usually made in parallel for multiple/multiple data
streams at the same time, so there are also limits on
the amount of RAM allocated for stream processing.
The choice of a set of classes for traffic classification
depends on the problem being solved. Figure 1.3 shows
the types of network traffic classification.


Figure 1.3. Types of network traffic classification

Classification by application-level protocols is
preferred by application-level protocols because such
classification is the most practically valuable.

Categorizing

Internet

traffic

by

applications

determines user activity and allows profiling, limiting
the activity of certain applications, and solving
marketing problems.

Classification according to the type of user action is
determined by the type of user activity, not by the
specific application used. In a sense, this is a
generalization of the previous type of classification.

CONCLUSION

It should be noted in this paper that weaknesses in
network protocols and systems. Filtering and blocking

Types of network traffic classification

Classification by
application-level

protocols

Classification of Internet

traffic generating

applications

Classification by types of

user actions


background image

Volume 04 Issue 10-2024

64


American Journal Of Applied Science And Technology
(ISSN

2771-2745)

VOLUME

04

ISSUE

10

Pages:

55-64

OCLC

1121105677
















































Publisher:

Oscar Publishing Services

Servi

mode for whitelists and blacklists, using machine
learning to classify traffic and types of network traffic
classification, types of network traffic classification are
considered as well.

REFERENCES

1.

Gulomov Sh.R. Tarmoqdagi zararli trafik turlari va
ularni aniqlash. Multidisciplinary Scientific Journal.
December,

Issue

24

|

2023,

-B.424-432

https://doi.org/10.5281/zenodo.10445437

2.

SN Tashev, AG G

aniev The Role of “Imagination” in

the Process of “Creative Thinking” Developing
Students'“Imagination” and “Creative Thinking”

Skills in Teaching Physics. Annals of the Romanian
Society for Cell Biology, 2021/3/6, 633-642
DOI:10.17762/pae.v58i1.1309

3.

SN

Tashev THE ROLE OF “IMAGINATION” IN THE

PROCESS OF “CREATIVE THINKING”, DEVELOPING
STUDENTS’ “IMAGINATION” AND “CREATIVE
THINKING” SKILLS IN TEACHING PHYSICS

PSYCHOLOGY

AND

EDUCATION,

3569-3575

DOI:10.17762/pae.v58i1.1309 License CC BY 4.0

4.

Y.B. Karamatovich, T.S. Norboboevich, N.I.
Ibrohimovich. Verification of the pocket filtering
based on method of verification on the model. 2019
International Conference on Information Science
and Communications Technologies (ICISCT)
DOI:10.1109/ICISCT47635.2019.9011901

5.

J. Ning et al., “Pine: Enabling privacy

-preserving

deep packet inspection on TLS with rule-hiding and

fast connection establishment,” in Proc. Eur. Symp.

Res.

Comput.

Secur.,

2020,

pp.

3

22

DOI:10.1007/978-3-030-58951-6_1

6.

J. Kim, S. Camtepe, J. Baek, W. Susilo, J. Pieprzyk,

and S. Nepal, “P2DPI: Practical and privacy

-

preserving deep packet inspection,” in Proc. ACM

Asia Conf. Comput. Commun. Secur., 2021, pp. 135

146. https://doi.org/10.1145/3433210.3437525

7.

Kim, S.; Yoon, S.; Narantuya, J.; Lim, H. Secure
collecting, optimizing, and deploying of firewall
rules in software-defined networks. IEEE Access
2020,

8,

15166

15177.

DOI:10.1109/ACCESS.2020.2967503

8.

Hakani, D. A Survey on Firewall for cloud security
with Anomaly detection in Firewall Policy. In
Proceedings of the International Conference on
Artificial Intelligence and Smart Communication,
Greater Noida, India, 27

29 January 2023; pp. 825

830. DOI:10.1109/AISC56616.2023.10085419

9.

Rajaboevich, G.; Dilbar, K.; Azatovna, A.;
Ismoilovna, Q. Comparative Analysis of Methods
Content Filtering Network Traffic. Int. J. Emerg.
Trends

Eng.

Res.

2020,

8,

1561

1569

DOI:10.30534/ijeter/2020/15852020

10.

Islam, M.; Uddin, M.; Hossain, D.; Dulal, M.; Ahmed,
D.; Shakil, M.; Moazzam, D.; Golam, M. Analysis and
Evaluation of Network and Application Security
Based on Next Generation Firewall. Int. J. Comput.
Digit.

Syst.

2023,

13,

193

202

DOI:10.3390/molecules26082297

References

Gulomov Sh.R. Tarmoqdagi zararli trafik turlari va ularni aniqlash. Multidisciplinary Scientific Journal. December, Issue 24 | 2023, -B.424-432 https://doi.org/10.5281/zenodo.10445437

SN Tashev, AG Ganiev The Role of “Imagination” in the Process of “Creative Thinking” Developing Students'“Imagination” and “Creative Thinking” Skills in Teaching Physics. Annals of the Romanian Society for Cell Biology, 2021/3/6, 633-642 DOI:10.17762/pae.v58i1.1309

SN Tashev THE ROLE OF “IMAGINATION” IN THE PROCESS OF “CREATIVE THINKING”, DEVELOPING STUDENTS’ “IMAGINATION” AND “CREATIVE THINKING” SKILLS IN TEACHING PHYSICS PSYCHOLOGY AND EDUCATION, 3569-3575 DOI:10.17762/pae.v58i1.1309 License CC BY 4.0

Y.B. Karamatovich, T.S. Norboboevich, N.I. Ibrohimovich. Verification of the pocket filtering based on method of verification on the model. 2019 International Conference on Information Science and Communications Technologies (ICISCT) DOI:10.1109/ICISCT47635.2019.9011901

J. Ning et al., “Pine: Enabling privacy-preserving deep packet inspection on TLS with rule-hiding and fast connection establishment,” in Proc. Eur. Symp. Res. Comput. Secur., 2020, pp. 3–22 DOI:10.1007/978-3-030-58951-6_1

J. Kim, S. Camtepe, J. Baek, W. Susilo, J. Pieprzyk, and S. Nepal, “P2DPI: Practical and privacy-preserving deep packet inspection,” in Proc. ACM Asia Conf. Comput. Commun. Secur., 2021, pp. 135–146. https://doi.org/10.1145/3433210.3437525

Kim, S.; Yoon, S.; Narantuya, J.; Lim, H. Secure collecting, optimizing, and deploying of firewall rules in software-defined networks. IEEE Access 2020, 8, 15166–15177. DOI:10.1109/ACCESS.2020.2967503

Hakani, D. A Survey on Firewall for cloud security with Anomaly detection in Firewall Policy. In Proceedings of the International Conference on Artificial Intelligence and Smart Communication, Greater Noida, India, 27–29 January 2023; pp. 825–830. DOI:10.1109/AISC56616.2023.10085419

Rajaboevich, G.; Dilbar, K.; Azatovna, A.; Ismoilovna, Q. Comparative Analysis of Methods Content Filtering Network Traffic. Int. J. Emerg. Trends Eng. Res. 2020, 8, 1561–1569 DOI:10.30534/ijeter/2020/15852020

Islam, M.; Uddin, M.; Hossain, D.; Dulal, M.; Ahmed, D.; Shakil, M.; Moazzam, D.; Golam, M. Analysis and Evaluation of Network and Application Security Based on Next Generation Firewall. Int. J. Comput. Digit. Syst. 2023, 13, 193–202 DOI:10.3390/molecules26082297

Rowsan Jahan Bhuiyan, Salma Akter, Aftab Uddin, Md Shujan Shak, Md Rasibul Islam, S M Shadul Islam Rishad, Farzana Sultana, & Md. Hasan-Or-Rashid. (2024). SENTIMENT ANALYSIS OF CUSTOMER FEEDBACK IN THE BANKING SECTOR: A COMPARATIVE STUDY OF MACHINE LEARNING MODELS. The American Journal of Engineering and Technology, 6(10), 54–66. https://doi.org/10.37547/tajet/Volume06Issue10-07

Md Habibur Rahman, Ashim Chandra Das, Md Shujan Shak, Md Kafil Uddin, Md Imdadul Alam, Nafis Anjum, Md Nad Vi Al Bony, & Murshida Alam. (2024). TRANSFORMING CUSTOMER RETENTION IN FINTECH INDUSTRY THROUGH PREDICTIVE ANALYTICS AND MACHINE LEARNING. The American Journal of Engineering and Technology, 6(10), 150–163. https://doi.org/10.37547/tajet/Volume06Issue10-17

Md Salim Chowdhury, Md Shujan Shak, Suniti Devi, Md Rashel Miah, Abdullah Al Mamun, Estak Ahmed, Sk Abu Sheleh Hera, Fuad Mahmud, & MD Shahin Alam Mozumder. (2024). Optimizing E-Commerce Pricing Strategies: A Comparative Analysis of Machine Learning Models for Predicting Customer Satisfaction. The American Journal of Engineering and Technology, 6(09), 6–17. https://doi.org/10.37547/tajet/Volume06Issue09-02

Md Shujan Shak, Md Shahin Alam Mozumder, Md Amit Hasan, Ashim Chandra Das, Md Rashel Miah, Salma Akter, & Md Nur Hossain. (2024). OPTIMIZING RETAIL DEMAND FORECASTING: A PERFORMANCE EVALUATION OF MACHINE LEARNING MODELS INCLUDING LSTM AND GRADIENT BOOSTING. The American Journal of Engineering and Technology, 6(09), 67–80. https://doi.org/10.37547/tajet/Volume06Issue09-09

Md Abu Sayed, Badruddowza, Md Shohail Uddin Sarker, Abdullah Al Mamun, Norun Nabi, Fuad Mahmud, Md Khorshed Alam, Md Tarek Hasan, Md Rashed Buiya, & Mashaeikh Zaman Md. Eftakhar Choudhury. (2024). COMPARATIVE ANALYSIS OF MACHINE LEARNING ALGORITHMS FOR PREDICTING CYBERSECURITY ATTACK SUCCESS: A PERFORMANCE EVALUATION. The American Journal of Engineering and Technology, 6(09), 81–91. https://doi.org/10.37547/tajet/Volume06Issue09-10