How machine learning and AI can help reduce the cyber- attacks
summary
Claims that ML and AI already cut alerts by automating anomaly detection, spam and phishing classification, and DDoS forecasting, citing a 22% increase in malware attacks and a 200% rise in DDoS in Q1 2019 and a typical 10 days to detect a breach. Uses time‑series forecasting, SVMs, logistic regression, decision trees and deep/deep‑learning end‑to‑end models; Gmail‑style classifiers reduce spam, time series spots seasonal spikes and can flag DDoS but struggles with adaptive traffic, false positives and overfitting, and requires heavy domain data and careful deployment.
AI-written summary · about
chapters (11)
- 0:00Introduction to AI and ML
- 2:18Defining machine learning
- 5:33The machine learning process
- 8:09Anomaly detection
- 10:13Addressing DDoS attacks
- 12:28Time series for security
- 18:35Phishing and classification
- 21:48Algorithms and decision trees
- 26:00Malware detection strategies
- 31:08AI implementation challenges
- 34:47Q&A and closing remarks
related talks
32mAI/Machine Learning: What is Actually Working in Cybersecurity?
22mApplying AI for Threat Detection in the World of Cybersecurity
11mAI-Powered Cybersecurity in the Age of Advanced Threats
44mAdvancing Cybersecurity with AI and Machine Learning: Opportunities and Challenges
27m"Cybertelligence": How AI is changing the face of cybersecurity- 44mAdvancing Cybersecurity with AI and Machine Learning: Opportunities and Challenges
44mAI Meets Cybersecurity: Boosting Speed Without Losing Control
41mLeveraging AI & Deep Learning in the Battle against Zero Day Cyber Attacks