International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 04 | Apr 2024
p-ISSN: 2395-0072
www.irjet.net
Cyber security threats prevention, detection and mitigation using machine learning techniques Rishu Nitin Verma 1, Kirti Rajeev Tiwari 2, Mohd Idrisi Akram 3, Prof. Simran Patil4 1,2,3B.E. Student, Dept. of Information Technology, Theem College of Engineering, Maharashtra, India 4Professor, Dept. of Information Technology, Theem College of Engineering, Maharashtra, India
---------------------------------------------------------------------***--------------------------------------------------------------------cybersecurity measures to counter these evolving challenges. In the realm of cyber threat management, encompassing prevention, detection, and mitigation, through a synthesis of Machine Learning (ML) techniques and Software-Defined Networking (SDN) infrastructure. By melding the predictive prowess of ML algorithms with the agile control mechanisms provided by SDN, this approach aims to fortify cybersecurity defenses and mitigate the impact of malicious activities. Central to this investigation are DDoS attacks, notorious for their ability to incapacitate networks by flooding them with an overwhelming influx of traffic. Leveraging SDN capabilities, the study pioneers a methodology for the real-time identification and neutralization of DDoS attacks. ML algorithms including Random Forest, Support Vector Machine (SVM), Decision Tree, k-Nearest Neighbors (KNN), Naive Bayes, and Logistic Regression are harnessed to scrutinize network traffic patterns and discern aberrant behavior indicative of potential DDoS assaults. Moreover, the study confronts the persistent menace of email spam, a favored vector for phishing exploits and malware dissemination. Harnessing the adeptness of Naive Bayes, a venerable ML algorithm for text categorization, the research explores strategies to accurately discern and filter out spam emails.
Abstract - Communication is essential for individuals with disabilities to fully participate in society. However, those who are blind, deaf, and mute face significant challenges in traditional communication methods. The paper proposes a multimodal approach to facilitate communication for individuals with these multiple disabilities. The approach combines sign language, translator, and assistive technologies to create a comprehensive communication system. Through voice assistant, individuals who are blind can receive information through voice or touch, while translator devices provide real-time feedback. For sign language gestures. Additionally, assistive technologies such as text-tospeech and speech-to-text software enable individuals who are mute to communicate verbally, which can be translated into tactile or visual formats for those who are blind or deaf. By integrating these modalities, individuals with multiple disabilities can overcome communication barriers and engage more effectively with their environment and peers. This multimodal approach offers a promising avenue for enhancing communication. In today’s advance world of science and technology, communication field has developed at such extent where we can connect to any part of the world within fraction of minutes and hour. We can send messages, make call or send documents, files to anyone, according to need. Communication has been important to express our thoughts, idea etc. but when it’s come about blind, deaf and mute people’s it become difficult for them and us to communicate with each other. So, here we have made such a software which shall help them to communicate with each other without depending upon any middle man. With the help of this we would be able to help mute, blind and deaf people to communicate with each other without depending upon each other.
Through the integration of ML algorithms within the SDN framework, this research strives to empower proactive threat mitigation, swift detection, and efficient response strategies against both DDoS attacks and email spam. Rigorous empirical evaluations will ascertain the effectiveness of this integrated ML-SDN solution, contributing to the advancement of cybersecurity practices and reinforcing resilience against the dynamic threat landscape.
Key Words: SDN; ML; SVM; KNN; DDOS; Email spam.
2. LITERATURE SURVEY
1. INTRODUCTION
Conducting a methodical literature review serves as a method to assess and interpret all existing research pertinent to a specific research query, subject, or phenomenon under examination. The study employed comprehensive scientific databases containing full-text papers, along with other relevant scholarly articles within the realm of social sciences. All academic papers and other relevant publications produced between 2009 and March 2020 were considered in the analysis.
The modern era of cybersecurity is fraught with an expanding spectrum of threats, ranging from the disruptive force of Distributed Denial of Service (DDoS) attacks to the insidious infiltration of email spam campaigns. These threats not only jeopardize critical systems but also undermine the integrity of data and compromise user privacy. Consequently, there is an urgent need to develop resilient and adaptive
© 2024, IRJET
|
Impact Factor value: 8.226
|
ISO 9001:2008 Certified Journal
|
Page 1161