Table of contents : 10 Machine Learning Blueprints You Should Know for Cybersecurity Contributors About the author About the reviewers Preface Who this book is for What this book covers To get the most out of this book Download the example code files Conventions used Get in touch Share Your Thoughts Download a free PDF copy of this book Chapter 1: On Cybersecurity and Machine Learning The basics of cybersecurity Traditional principles of cybersecurity Modern cybersecurity – a multi-faceted issue Privacy An overview of machine learning Machine learning workflow Supervised learning Unsupervised learning Semi-supervised learning Evaluation metrics Machine learning – cybersecurity versus other domains Summary Chapter 2: Detecting Suspicious Activity Technical requirements Basics of anomaly detection What is anomaly detection? Introducing the NSL-KDD dataset Statistical algorithms for intrusion detection Univariate outlier detection Elliptic envelope Local outlier factor Machine learning algorithms for intrusion detection Density-based scan (DBSCAN) One-class SVM Isolation forest Autoencoders Summary Chapter 3: Malware Detection Using Transformers and BERT Technical requirements Basics of malware What is malware? Types of malware Malware detection Malware detection methods Malware analysis Transformers and attention Understanding attention Understanding transformers Understanding BERT Detecting malware with BERT Malware as language The relevance of BERT Getting the data Preprocessing the data Building a classifier Summary Chapter 4: Detecting Fake Reviews Technical requirements Reviews and integrity Why fake reviews exist Evolution of fake reviews Statistical analysis Exploratory data analysis Feature extraction Statistical tests Modeling fake reviews with regression Ordinary Least Squares regression OLS assumptions Interpreting OLS regression Implementing OLS regression Summary Chapter 5: Detecting Deepfakes Technical requirements All about deepfakes A foray into GANs How are deepfakes created? The social impact of deepfakes Detecting fake images A naive model to detect fake images Detecting deepfake videos Building deepfake detectors Summary Chapter 6: Detecting Machine-Generated Text Technical requirements Text generation models Understanding GPT Naïve detection Creating the dataset Feature exploration Using machine learning models for detecting text Playing around with the model Automatic feature extraction Transformer methods for detecting automated text Compare and contrast Summary Chapter 7: Attributing Authorship and How to Evade It Technical requirements Authorship attribution and obfuscation What is authorship attribution? What is authorship obfuscation? Techniques for authorship attribution Dataset Feature extraction Training the attributor Improving authorship attribution Techniques for authorship obfuscation Improving obfuscation techniques Summary Chapter 8: Detecting Fake News with Graph Neural Networks Technical requirements An introduction to graphs What is a graph? Representing graphs Graphs in the real world Machine learning on graphs Traditional graph learning Graph embeddings GNNs Fake news detection with GNN Modeling a GNN The UPFD framework Dataset and setup Implementing GNN-based fake news detection Playing around with the model Summary Chapter 9: Attacking Models with Adversarial Machine Learning Technical requirements Introduction to AML The importance of ML Adversarial attacks Adversarial tactics Attacking image models FGSM PGD Attacking text models Manipulating text Further attacks Developing robustness against adversarial attacks Adversarial training Defensive distillation Gradient regularization Input preprocessing Ensemble methods Certified defenses Summary Chapter 10: Protecting User Privacy with Differential Privacy Technical requirements The basics of privacy Core elements of data privacy Privacy and the GDPR Privacy by design Privacy and machine learning Differential privacy What is differential privacy? Differential privacy – a real-world example Benefits of differential privacy Differentially private machine learning IBM Diffprivlib Credit card fraud detection with differential privacy Differentially private deep learning DP-SGD algorithm Implementation Differential privacy in practice Summary Chapter 11: Protecting User Privacy with Federated Machine Learning Technical requirements An introduction to federated machine learning Privacy challenges in machine learning How federated machine learning works The benefits of federated learning Challenges in federated learning Implementing federated averaging Importing libraries Dataset setup Client setup Model implementation Weight scaling Global model initialization Setting up the experiment Putting it all together Reviewing the privacy-utility trade-off in federated learning Global model (no privacy) Local model (full privacy) Understanding the trade-off Beyond the MNIST dataset Summary Chapter 12: Breaking into the Sec-ML Industry Study guide for machine learning and cybersecurity Machine learning theory Hands-on machine learning Cybersecurity Interview questions Theory-based questions Experience-based questions Conceptual questions Additional project blueprints Improved intrusion detection Adversarial attacks on intrusion detection Hate speech and toxicity detection Detecting fake news and misinformation Summary Index Why subscribe? Other Books You May Enjoy Packt is searching for authors like you Share Your Thoughts Download a free PDF copy of this book