• C Brundha
• Gowthami C
• Kundeti Sowmya Sree
• Udaykiran K
• Sneha Karamadi

The increasing reliance on the internet and social platforms has led to growing concerns over cyberbullying, intrusion, and Distributed Denial of Service (DDoS) attacks. These threats pose serious risks to personal privacy, organizational security, and public safety. To address these challenges, this work presents a comprehensive survey and review of recent developments in Machine Learning (ML) and Deep Learning (DL) techniques applied to cyber security. In the domain of cyberbullying detection, emotion-based analysis, hybrid models, and multi-modal deep learning approaches such as CNNs, RNNs, and YOLO have been employed for identifying harmful content across text and images. For intrusion detection, traditional ML algorithms like Support Vector Machines (SVM), k-Nearest Neighbors (k-NN), Decision Trees, and advanced DL techniques including LSTM and Autoencoders have shown promising results in classifying normal and malicious network traffic using benchmark datasets like NSL-KDD. Similarly, in Software Defined Networks (SDN), ML models such as Random Forest and Naïve Bayes have been explored to detect DDoS attacks by analyzing flow-based traffic data. These models are evaluated based on accuracy, detection rate, and computational efficiency. Despite significant progress, key challenges remain in the form of data imbalance, real-time detection, scalability, and handling contextual and multi-modal data. This study highlights the strengths and limitations of current approaches and identifies future directions to develop more robust and intelligent cybersecurity solutions.

Cybersecurity, Machine Learning, Intrusion Detection,Cyberbullying Detection, Distributed Denial of Service (DDoS).

""AI Against Cyber Harm: Detecting Threats in Social and Software-Defined Networks"", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 5, page no.b740-b744, May-2025, Available :http://www.jetir.org/papers/JETIR2505187.pdf

""AI Against Cyber Harm: Detecting Threats in Social and Software-Defined Networks"", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 5, page no. ppb740-b744, May-2025, Available at : http://www.jetir.org/papers/JETIR2505187.pdf