Machine learning involves development and training of models used to predict future outcomes. This book is a practical guide to all the tips and tricks related to machine learning. It includes hands-on, easy to access techniques on topics like model selection, performance tuning, training neural networks, time series analysis and a lot more.

• Cover
• Title Page
• Copyright and Credits
• About Packt
• Contributors
• Table of Contents
• Preface
• Chapter 1: Quantifying Learning Algorithms
  • Statistical models
  • Learning curve
    • Machine learning
    • Wright's model
  • Curve fitting
    • Residual
  • Statistical modeling – the two cultures of Leo Breiman
  • Training data development data – test data
    • Size of the training, development, and test set
  • Bias-variance trade off
  • Regularization
    • Ridge regression (L2)
    • Least absolute shrinkage and selection operator 
  • Cross-validation and model selection
    • K-fold cross-validation
  • Model selection using cross-validation
  • 0.632 rule in bootstrapping
  • Model evaluation
    • Confusion matrix
  • Receiver operating characteristic curve
    • Area under ROC
  • H-measure
  • Dimensionality reduction
  • Summary
• Chapter 2: Evaluating Kernel Learning
  • Introduction to vectors
    • Magnitude of the vector
    • Dot product
  • Linear separability
  • Hyperplanes 
  • SVM
    • Support vector
  • Kernel trick
    • Kernel
    • Back to Kernel trick
  • Kernel types
    • Linear kernel
    • Polynomial kernel
    • Gaussian kernel
  • SVM example and parameter optimization through grid search
  • Summary
• Chapter 3: Performance in Ensemble Learning
  • What is ensemble learning?
    • Ensemble methods 
      • Bootstrapping
  • Bagging
  • Decision tree
    • Tree splitting
    • Parameters of tree splitting
  • Random forest algorithm
    • Case study
  • Boosting
    • Gradient boosting
      • Parameters of gradient boosting
  • Summary
• Chapter 4: Training Neural Networks
  • Neural networks
    • How a neural network works
    • Model initialization
    • Loss function
    • Optimization
    • Computation in neural networks
      • Calculation of activation for H1
    • Backward propagation
    • Activation function
      • Types of activation functions
  • Network initialization
    • Backpropagation
  • Overfitting
  • Prevention of overfitting in NNs
  • Vanishing gradient 
    • Overcoming vanishing gradient
  • Recurrent neural networks
    • Limitations of RNNs
    • Use case
  • Summary
• Chapter 5: Time Series Analysis
  • Introduction to time series analysis
  • White noise
    • Detection of white noise in a series
  • Random walk
  • Autoregression
  • Autocorrelation
  • Stationarity
    • Detection of stationarity
  • AR model
  • Moving average model
  • Autoregressive integrated moving average
  • Optimization of parameters
    • AR model
    • ARIMA model
  • Anomaly detection
  • Summary
• Chapter 6: Natural Language Processing
  • Text corpus
    • Sentences
    • Words
      • Bags of words
  • TF-IDF
    • Executing the count vectorizer
    • Executing TF-IDF in Python
  • Sentiment analysis
    • Sentiment classification
      • TF-IDF feature extraction
      • Count vectorizer bag of words feature extraction
        • Model building count vectorization
  • Topic modeling 
    • LDA architecture
    • Evaluating the model
    • Visualizing the LDA
    • The Naive Bayes technique in text classification
  • The Bayes theorem
    • How the Naive Bayes classifier works
  • Summary
• Chapter 7: Temporal and Sequential Pattern Discovery
  • Association rules
  • Apriori algorithm
    • Finding association rules
  • Frequent pattern growth
    • Frequent pattern tree growth
    • Validation 
      • Importing the library
  • Summary
• Chapter 8: Probabilistic Graphical Models
  • Key concepts
  • Bayes rule
  • Bayes network
    • Probabilities of nodes
    • CPT
    • Example of the training and test set
  • Summary
• Chapter 9: Selected Topics in Deep Learning
  • Deep neural networks
    • Why do we need a deep learning model?
    • Deep neural network notation
    • Forward propagation in a deep network
    • Parameters W and b
    • Forward and backward propagation
    • Error computation
  • Backward propagation
  • Forward propagation equation
  • Backward propagation equation
  • Parameters and hyperparameters
  • Bias initialization
    • Hyperparameters
    • Use case – digit recognizer
  • Generative adversarial networks
  • Hinton's Capsule network
    • The Capsule Network and convolutional neural networks
  • Summary
• Chapter 10: Causal Inference
  • Granger causality
  • F-test
    • Limitations
    • Use case
  • Graphical causal models
  • Summary
• Chapter 11: Advanced Methods
  • Introduction
  • Kernel PCA
  • Independent component analysis
    • Preprocessing for ICA
    • Approach
  • Compressed sensing
    • Our goal
  • Self-organizing maps
    • SOM
  • Bayesian multiple imputation
  • Summary
• Other Books You May Enjoy
• Index