Professional Documents
Culture Documents
Synopsis
Synopsis
AKURDI , PUNE – 44
BE Information Technology
PROJECT SYNOPSIS
With the use of hardware and software, the TCP security system primarily acts as a
protective wall created between internal and external networks. Filtering and blocking
potentially dangerous and malicious network connections are its functions. The
concept of a network firewall first appeared in the 1980s, and as firewall technology
has continued to advance, new features like packet filtering, encryption, antivirus
software, and proxy servers have been added. Software specifically designed for
packet filtering can be used to see the header of packets passing through the firewall.
It will only allow packets that are deemed secure to pass through; otherwise, it will
trash them. Usually, the process requires a lot of calculation. However, using AI
technology, a smart firewall can more effectively recognize network behavior with
characteristic values. One of the developments in firewall technology is the use of
intelligent firewalls. In addition to enhancing the firewall's functionality, artificial
intelligence can provide it the ability to learn on its own and thwart the most recent
network attacks. Nowadays, there are an increasing number of cyberattacks, such as
the Distributed Denial of Service (DDoS) attack, which is one of the most challenging
security issues to resolve. DDoS assaults will become a significant concern with the
development of the internet of things. The development of firewall technology will
therefore be even more crucial in the 5G future.
LITERATURE SURVEY
PROJECT NAME AUTHOR NAME ABSTRACT KEYWORDS
Research on the Key Technology of Yong He The diversification of network network security, machine
Network Security Based on Machine services has generated massive learning, detection, anomaly
Learning amounts of Internet data, and detection, key technologies
traditional network security
technologies have been
difficult to meet the current
needs of network security in
terms of performance and self-
adaptability.
Machine Learning based Intrusion Sushant Sharma, Pavol Among three tested web-based attacks, detection,
Detection System for Web-Based Zavarsky, Sergey algorithms, the J48 decision machine learning, feature
Attacks Butakov tree algorithm provided the extraction .
highest True Positive rate,
Precision, and Recall.
Securing Machine Learning Engines in Agus Kurniawan, The experimental results show Machine Learning, CP-ABE,
IoT Applications with Attribute-Based Marcel Kyas that our proposed system can Pairing Encryption, Security
Encryption address security issues on System
machine learning computation
with low time consumption.
An Assessment of the Usability of Sean Oesch, Robert We also discovered that Machine Learning, Naval SOC,
Machine Learning Based Tools for the Bridges, Jared Smith, analysts lacked a clear mental cyber testbed
Security Operations Center Justin Beaver, John model of how these tools
Goodall, Kelly Huffer, generate scores, resulting in
Craig Miles, Dan mistrust and/or misuse of the
Scofield tools themselves.
HealthGuard: A Machine Learning- AKM Iqtidar Newaz, We trained HealthGuard with Healthcare, Smart Healthcare
Based Security Framework for Smart Amit Kumar Sikder, data collected for eight System, Anomaly Detection,
Healthcare Systems Mohammad Ashiqur different smart medical devices Smart Medical Devices,
Rahman for twelve benign events Implantable Medical Devices,
including seven normal user Security
activities and five disease
affected events.
PROBLEM STATEMENT:-
Cyber attack is one of the threat at the country level. The security of an entire
nation depends on it’s security. If any Cyber attack may occure then the security
of a nation is in danger. In order to improve the security as much as possible we
need to predict the TCP security action.
OBJECTIVES:-
The goal of the TCP security action prediction project is to make the machine
able to identify and solve the cyber attack on its own.
MOTIVATION:-
SYSTEM ARCHITECTUR:-
Software Requirements:-
• IDE : Spyder
• Database : SQLite
Hardware Requirements:-
• System : Intel I5 Processor.
• Monitor : 15.
• Ram : 8 GB
CONCLUSION:-
Our study investigates the use of machine learning classification models for
foretelling TCP security activities. By preprocessing and fine-tuning each
model, the prediction accuracy of each may be increased to more than 98
percent. The outcome of the AdaBoost model illustrates a strategy to address
the classification problem of class imbalance. The prediction accuracy of TCP
security action achieves over 98 percent by adopting ensemble learning
technique. More crucially, the ensembled model can categorise all classes
regardless of how unevenly distributed the classes are in the dataset.
Additionally, our research contributes positively to the study of intelligent
firewalls and is in the field of Internet security technology.
REFERENCES:-
(Signature)