Table of Contents Mastering Java Machine Learning Credits Foreword About the Authors About the Reviewers www.PacktPub.com eBooks, discount offers, and more Why subscribe? Customer Feedback Preface What this book covers What you need for this book Who this book is for Conventions Reader feedback Customer support Errata Piracy Questions Machine Learning Review Machine learning – history and definition What is not machine learning? Machine learning – concepts and terminology Machine learning – types and subtypes Datasets used in machine learning Machine learning applications Practical issues in machine learning Machine learning – roles and process Roles Process Machine learning – tools and datasets Datasets Summary Practical Approach to Real-World Supervised Learning Formal description and notation Data quality analysis Descriptive data analysis Basic label analysis Basic feature analysis Visualization analysis Univariate feature analysis Categorical features Continuous features Multivariate feature analysis Data transformation and preprocessing Feature construction Handling missing values Outliers Discretization Data sampling Is sampling needed? Undersampling and oversampling Stratified sampling Training, validation, and test set Feature relevance analysis and dimensionality reduction Feature search techniques Feature evaluation techniques Filter approach Univariate feature selection Information theoretic approach Statistical approach Multivariate feature selection Minimal redundancy maximal relevance (mRMR) Correlation-based feature selection (CFS) Wrapper approach Embedded approach Model building Linear models Linear Regression Algorithm input and output How does it work? Advantages and limitations Naïve Bayes Algorithm input and output How does it work? Advantages and limitations Logistic Regression Algorithm input and output How does it work? Advantages and limitations Non-linear models Decision Trees Algorithm inputs and outputs How does it work? Advantages and limitations K-Nearest Neighbors (KNN) Algorithm inputs and outputs How does it work? Advantages and limitations Support vector machines (SVM) Algorithm inputs and outputs How does it work? Advantages and limitations Ensemble learning and meta learners Bootstrap aggregating or bagging Algorithm inputs and outputs How does it work? Random Forest Advantages and limitations Boosting Algorithm inputs and outputs How does it work? Advantages and limitations Model assessment, evaluation, and comparisons Model assessment Model evaluation metrics Confusion matrix and related metrics ROC and PRC curves Gain charts and lift curves Model comparisons Comparing two algorithms McNemar's Test Paired-t test Wilcoxon signed-rank test Comparing multiple algorithms ANOVA test Friedman's test Case Study – Horse Colic Classification Business problem Machine learning mapping Data analysis Label analysis Features analysis Supervised learning experiments Weka experiments Sample end-to-end process in Java Weka experimenter and model selection RapidMiner experiments Visualization analysis Feature selection Model process flow Model evaluation metrics Evaluation on Confusion Metrics ROC Curves, Lift Curves, and Gain Charts Results, observations, and analysis Summary References Unsupervised Machine Learning Techniques Issues in common with supervised learning Issues specific to unsupervised learning Feature analysis and dimensionality reduction Notation Linear methods Principal component analysis (PCA) Inputs and outputs How does it work? Advantages and limitations Random projections (RP) Inputs and outputs How does it work? Advantages and limitations Multidimensional Scaling (MDS) Inputs and outputs How does it work? Advantages and limitations Nonlinear methods Kernel Principal Component Analysis (KPCA) Inputs and outputs How does it work? Advantages and limitations Manifold learning Inputs and outputs How does it work? Advantages and limitations Clustering Clustering algorithms k-Means Inputs and outputs How does it work? Advantages and limitations DBSCAN Inputs and outputs How does it work? Advantages and limitations Mean shift Inputs and outputs How does it work? Advantages and limitations Expectation maximization (EM) or Gaussian mixture modeling (GMM) Input and output How does it work? Advantages and limitations Hierarchical clustering Input and output How does it work? Advantages and limitations Self-organizing maps (SOM) Inputs and outputs How does it work? Advantages and limitations Spectral clustering Inputs and outputs How does it work? Advantages and limitations Affinity propagation Inputs and outputs How does it work? Advantages and limitations Clustering validation and evaluation Internal evaluation measures Notation R-Squared Dunn's Indices Davies-Bouldin index Silhouette's index External evaluation measures Rand index F-Measure Normalized mutual information index Outlier or anomaly detection Outlier algorithms Statistical-based Inputs and outputs How does it work? Advantages and limitations Distance-based methods Inputs and outputs How does it work? Advantages and limitations Density-based methods Inputs and outputs How does it work? Advantages and limitations Clustering-based methods Inputs and outputs How does it work? Advantages and limitations High-dimensional-based methods Inputs and outputs How does it work? Advantages and limitations One-class SVM Inputs and outputs How does it work? Advantages and limitations Outlier evaluation techniques Supervised evaluation Unsupervised evaluation Real-world case study Tools and software Business problem Machine learning mapping Data collection Data quality analysis Data sampling and transformation Feature analysis and dimensionality reduction PCA Random projections ISOMAP Observations on feature analysis and dimensionality reduction Clustering models, results, and evaluation Observations and clustering analysis Outlier models, results, and evaluation Observations and analysis Summary References Semi-Supervised and Active Learning Semi-supervised learning Representation, notation, and assumptions Semi-supervised learning techniques Self-training SSL Inputs and outputs How does it work? Advantages and limitations Co-training SSL or multi-view SSL Inputs and outputs How does it work? Advantages and limitations Cluster and label SSL Inputs and outputs How does it work? Advantages and limitations Transductive graph label propagation Inputs and outputs How does it work? Advantages and limitations Transductive SVM (TSVM) Inputs and outputs How does it work? Advantages and limitations Case study in semi-supervised learning Tools and software Business problem Machine learning mapping Data collection Data quality analysis Data sampling and transformation Datasets and analysis Feature analysis results Experiments and results Analysis of semi-supervised learning Active learning Representation and notation Active learning scenarios Active learning approaches Uncertainty sampling How does it work? Least confident sampling Smallest margin sampling Label entropy sampling Advantages and limitations Version space sampling Query by disagreement (QBD) How does it work? Query by Committee (QBC) How does it work? Advantages and limitations Data distribution sampling How does it work? Expected model change Expected error reduction Variance reduction Density weighted methods Advantages and limitations Case study in active learning Tools and software Business problem Machine learning mapping Data Collection Data sampling and transformation Feature analysis and dimensionality reduction Models, results, and evaluation Pool-based scenarios Stream-based scenarios Analysis of active learning results Summary References Real-Time Stream Machine Learning Assumptions and mathematical notations Basic stream processing and computational techniques Stream computations Sliding windows Sampling Concept drift and drift detection Data management Partial memory Full memory Detection methods Monitoring model evolution Widmer and Kubat Drift Detection Method or DDM Early Drift Detection Method or EDDM Monitoring distribution changes Welch's t test Kolmogorov-Smirnov's test CUSUM and Page-Hinckley test Adaptation methods Explicit adaptation Implicit adaptation Incremental supervised learning Modeling techniques Linear algorithms Online linear models with loss functions Inputs and outputs How does it work? Advantages and limitations Online Naïve Bayes Inputs and outputs How does it work? Advantages and limitations Non-linear algorithms Hoeffding trees or very fast decision trees (VFDT) Inputs and outputs How does it work? Advantages and limitations Ensemble algorithms Weighted majority algorithm Inputs and outputs How does it work? Advantages and limitations Online Bagging algorithm Inputs and outputs How does it work? Advantages and limitations Online Boosting algorithm Inputs and outputs How does it work? Advantages and limitations Validation, evaluation, and comparisons in online setting Model validation techniques Prequential evaluation Holdout evaluation Controlled permutations factor graph / Factor graph factor graph, messaging in / Messaging in factor graph input and output / Input and output working / How does it work? advantages and limitations / Advantages and limitations publish-subscribe frameworks / Publish-subscribe frameworks Q Query by Committee (QBC) about / Query by Committee (QBC) Query by disagreement (QBD) about / Query by disagreement (QBD) R R-Squared / R-Squared Radial Basis Function (RBF) / Inputs and outputs Rand index / Rand index Random Forest / Random Forest Random Forest (RF) / Feature relevance and analysis, Random Forest random projections (RP) about / Random projections (RP) inputs / Inputs and outputs outputs / Inputs and outputs working / How does it work? advantages / Advantages and limitations limitations / Advantages and limitations RapidMiner about / Machine learning – tools and datasets, Case Study – Horse Colic Classification URL / Machine learning – tools and datasets experiments / RapidMiner experiments visualization analysis / Visualization analysis feature selection / Feature selection model process flow / Model process flow model evaluation metrics / Model evaluation metrics real-time Big Data Machine Learning about / Real-time Big Data Machine Learning SAMOA / SAMOA as a real-time Big Data Machine Learning framework machine learning algorithms / Machine Learning algorithms tools / Tools and usage usage / Tools and usage experiments / Experiments, results, and analysis results / Experiments, results, and analysis analysis / Experiments, results, and analysis results, analysis / Analysis of results real-time stream processing / Real-time stream processing real-world case study about / Real-world case study tools / Tools and software software / Tools and software business problem / Business problem machine learning, mapping / Machine learning mapping data collection / Data collection data quality analysis / Data quality analysis data sampling / Data sampling and transformation data transformation / Data sampling and transformation feature analysis / Feature analysis and dimensionality reduction dimensionality reduction / Feature analysis and dimensionality reduction models, clustering / Clustering models, results, and evaluation results / Clustering models, results, and evaluation, Outlier models, results, and evaluation evaluation / Clustering models, results, and evaluation, Outlier models, results, and evaluation outlier models / Outlier models, results, and evaluation reasoning, Bayesian networks patterns / Reasoning patterns causal or predictive reasoning / Causal or predictive reasoning evidential or diagnostic reasoning / Evidential or diagnostic reasoning intercausal reasoning / Intercausal reasoning combined reasoning / Combined reasoning receiver operating characteristics (ROC) / Machine learning – concepts and terminology Recurrent neural networks (RNN) about / Recurrent Neural Networks structure / Structure of Recurrent Neural Networks learning / Learning and associated problems in RNNs issues / Learning and associated problems in RNNs Long short term memory (LSTM) / Long Short Term Memory Gated Recurrent Units (GRUs) / Gated Recurrent Units regression about / Formal description and notation regularization about / Regularization L2 regularization / L2 regularization L1 regularization / L1 regularization reinforcement learning / Machine learning – types and subtypes representation, Bayesian networks about / Representation definition / Definition resampling / Is sampling needed? Resilient Distributed Datasets (RDD) / Spark architecture Restricted Boltzmann Machines (RBM) about / Restricted Boltzmann Machines definition and mathematical notation / Definition and mathematical notation Conditional distribution / Conditional distribution free energy / Free energy in RBM training / Training the RBM sampling / Sampling in RBM contrastive divergence / Contrastive divergence , How does it work? persistent contrastive divergence / Persistent contrastive divergence ROC curve / ROC and PRC curves roles, machine learning about / Roles business domain expert / Roles data engineer / Roles project manager / Roles data scientist / Roles machine learning expert / Roles S SAMOA about / Machine learning – tools and datasets, SAMOA as a real-time Big Data Machine Learning framework URL / Machine learning – tools and datasets architecture / SAMOA architecture sampling about / Machine learning – concepts and terminology, Sampling uniform random sampling / Machine learning – concepts and terminology stratified random sampling / Machine learning – concepts and terminology cluster sampling / Machine learning – concepts and terminology systematic sampling / Machine learning – concepts and terminology sampling-based techniques, Bayesian networks about / Sampling-based techniques forward sampling with rejection / Forward sampling with rejection, How does it work? Samza / SAMOA as a real-time Big Data Machine Learning framework scalar product of vectors / Scalar product of vectors ScatterPlot Matrix / Multivariate feature analysis scatter plots / Multivariate feature analysis self-organizing maps (SOM) about / Self-organizing maps (SOM) inputs / Inputs and outputs output / Inputs and outputs working / How does it work? advantages / Advantages and limitations limitations / Advantages and limitations self-training SSL about / Self-training SSL inputs / Inputs and outputs outputs / Inputs and outputs working / How does it work? advantages / Advantages and limitations limitations / Advantages and limitations Query by Committee (QBC) / Query by Committee (QBC) semantic features / Semantic features semantic reasoning / Semantic reasoning and inferencing semi-supervised learning / Machine learning – types and subtypes Semi-Supervised Learning (SSL) about / Semi-supervised learning representation / Representation, notation, and assumptions notation / Representation, notation, and assumptions assumptions / Representation, notation, and assumptions assumptions, to be true / Representation, notation, and assumptions techniques / Semi-supervised learning techniques self-training SSL / Self-training SSL multi-view SSL / Co-training SSL or multi-view SSL co-training SSL / Co-training SSL or multi-view SSL label SSL / Cluster and label SSL cluster SSL / Cluster and label SSL transductive graph label propagation / Transductive graph label propagation transductive SVM (TSVM) / Transductive SVM (TSVM) advanatages / Advantages and limitations disadvanatages / Advantages and limitations data distribution sampling / Data distribution sampling Semi-Supervised Learning (SSL), case study about / Case study in semi-supervised learning tools / Tools and software software / Tools and software business problem / Business problem machine learning, mapping / Machine learning mapping data collection / Data collection data quality, analysis / Data quality analysis data sampling / Data sampling and transformation data transformation / Data sampling and transformation datasets / Datasets and analysis datasets, analysis / Datasets and analysis feature analysis, results / Feature analysis results experiments / Experiments and results results / Experiments and results analysis / Analysis of semi-supervised learning sentiment analysis / Sentiment analysis and opinion mining sequential data / Datasets used in machine learning shrinking methods embedded approach / Embedded approach Sigmoid function / Sigmoid function Sigmoid Kernel / How does it work? Silhouettes index / Silhouette's index similarity measures about / Similarity measures Euclidean distance / Euclidean distance Cosine distance / Cosine distance pairwise-adaptive similarity / Pairwise-adaptive similarity extended Jaccard Coefficient / Extended Jaccard coefficient Dice coefficient / Dice coefficient singular value decomposition (SVD) / Dimensionality reduction, Singular value decomposition (SVD) Singular Value Decomposition (SVD) / Advantages and limitations sliding windows about / Sliding windows SMILE reference link / Tools and software Smile URL / Machine learning – tools and datasets about / Machine learning – tools and datasets software / Tools and software source-sink frameworks / Source-sink frameworks Spark-MLlib about / Machine learning – tools and datasets URL / Machine learning – tools and datasets Spark core, components Resilient Distributed Datasets (RDD) / Spark architecture Lineage graph / Spark architecture Spark MLlib used, as Big Data Machine Learning / Spark MLlib as Big Data Machine Learning platform architecture / Spark architecture machine learning / Machine Learning in MLlib tools / Tools and usage usage / Tools and usage experiments / Experiments, results, and analysis results / Experiments, results, and analysis analysis / Experiments, results, and analysis reference link / Experiments, results, and analysis k-Means / k-Means k-Means, with PCA / k-Means with PCA k-Means with PCA, bisecting / Bisecting k-Means (with PCA) Gaussian Mixture Model (GMM) / Gaussian Mixture Model Random Forest / Random Forest results, analysis / Analysis of results Spark SQL / Spark SQL Spark Streaming about / Real-time Big Data Machine Learning sparse coding about / Sparse coding spectral clustering about / Spectral clustering input / Inputs and outputs output / Inputs and outputs working / How does it work? advantages / Advantages and limitations limitations / Advantages and limitations SQL frameworks / SQL frameworks standard deviation / Standard deviation standardization about / Document collection and standardization input / Inputs and outputs output / Inputs and outputs working / How does it work? Statistical-based about / Statistical-based input / Inputs and outputs outputs / Inputs and outputs working / How does it work? advantages / Advantages and limitations limitations / Advantages and limitations stemming / Stemming or lemmatization step execution mode / Amazon Elastic MapReduce Stochastic Gradient Descent (SGD) about / How does it work? / Supervised learning experiments stop words removal about / Stop words removal input / Inputs and outputs output / Inputs and outputs working / How does it work? stratified sampling / Stratified sampling stream / SAMOA architecture stream computational technique about / Basic stream processing and computational techniques, Stream computations frequency count / Stream computations point queries / Stream computations distinct count / Stream computations mean / Stream computations standard deviation / Stream computations correlation coefficient / Stream computations sliding windows / Sliding windows sampling / Sampling stream learning / Machine learning – types and subtypes stream learning, case study about / Case study in stream learning tools / Tools and software software / Tools and software business problem / Business problem machine learning, mapping / Machine learning mapping data collection / Data collection data sampling / Data sampling and transformation data transformation / Data sampling and transformation feature analysis / Feature analysis and dimensionality reduction dimensionality reduction / Feature analysis and dimensionality reduction models / Models, results, and evaluation results / Models, results, and evaluation evaluation / Models, results, and evaluation supervised learning experiments / Supervised learning experiments concept drift experiments / Concept drift experiments clustering experiments / Clustering experiments outlier detection experiments / Outlier detection experiments results, analysis / Analysis of stream learning results Stream Processing Engines (SPE) / Real-time stream processing stream processing technique about / Basic stream processing and computational techniques structured data sequential data / Datasets used in machine learning Structure Score Measure / Measures to evaluate structures subfields about / NLP, subfields, and tasks Subspace Outlier Detection (SOD) / How does it work? Sum of Squared Errors (SSE) / Clustering models, results, and evaluation, Experiments, results, and analysis supervised learning / Machine learning – types and subtypes experiments / Supervised learning experiments Weka, experiments / Weka experiments RapidMiner, experiments / RapidMiner experiments reference link / Results, observations, and analysis and unsupervised learning, common issues / Issues in common with supervised learning assumptions / Assumptions and mathematical notations mathematical notations / Assumptions and mathematical notations Support Vector Machines (SVM) / How does it work? support vector machines (SVM) about / Support vector machines (SVM) algorithm input / Algorithm inputs and outputs algorithm output / Algorithm inputs and outputs working / How does it work? advantages / Advantages and limitations limitations / Advantages and limitations Syntactic features about / Syntactic features Syntactic Language Models (SLM) about / Syntactic features Synthetic Minority Oversampling Technique (SMOTE) / Undersampling and oversampling T tasks about / NLP, subfields, and tasks Term Frequency (TF) / Term frequency (TF) term frequency (TF) / Frequency-based techniques term frequency-inverse document frequency (TF-IDF) / Term frequency-inverse document frequency (TF-IDF) text categorization about / Text categorization text clustering / Text clustering about / Text clustering feature transformation / Feature transformation, selection, and reduction selection / Feature transformation, selection, and reduction reduction / Feature transformation, selection, and reduction techniques / Clustering techniques evaluation / Evaluation of text clustering text mining topics / Topics in text mining categorization/classification / Text categorization/classification topic modeling / Topic modeling clustering / Text clustering named entity recognition (NER) / Named entity recognition Deep Learning / Deep learning and NLP NLP / Deep learning and NLP text processing components about / Text processing components and transformations document collection / Document collection and standardization standardization / Document collection and standardization tokenization / Tokenization stop words removal / Stop words removal lemmatization / Stemming or lemmatization local-global dictionary / Local/global dictionary or vocabulary? vocabulary / Local/global dictionary or vocabulary? feature extraction/generation / Feature extraction/generation feature representation / Feature representation and similarity similarity / Feature representation and similarity feature selection / Feature selection and dimensionality reduction dimensionality reduction / Feature selection and dimensionality reduction text summarization / Text summarization time-series forecasting / Machine learning – types and subtypes tokenization about / Tokenization input / Inputs and outputs output / Inputs and outputs working / How does it work? tools / Tools and software about / Tools and usage Mallet / Mallet KNIME / KNIME tools, machine learning RapidMiner / Machine learning – tools and datasets Weka / Machine learning – tools and datasets Knime / Machine learning – tools and datasets Mallet / Machine learning – tools and datasets Elki / Machine learning – tools and datasets JCLAL / Machine learning – tools and datasets KEEL / Machine learning – tools and datasets DeepLearning4J / Machine learning – tools and datasets Spark-MLlib / Machine learning – tools and datasets H2O / Machine learning – tools and datasets MOA/SAMOA / Machine learning – tools and datasets Neo4j / Machine learning – tools and datasets GraphX / Machine learning – tools and datasets OpenMarkov / Machine learning – tools and datasets Smile / Machine learning – tools and datasets topic modeling about / Topic modeling probabilistic latent semantic analysis (PLSA) / Probabilistic latent semantic analysis (PLSA) with mallet / Topic modeling with mallet business problem / Business problem machine learning, mapping / Machine Learning mapping data collection / Data collection data sampling / Data sampling and transformation transformation / Data sampling and transformation feature analysis / Feature analysis and dimensionality reduction dimensionality reduction / Feature analysis and dimensionality reduction models / Models, results, and evaluation results / Models, results, and evaluation evaluation / Models, results, and evaluation text processing results, analysis / Analysis of text processing results training phases competitive phase / How does it work? cooperation phase / How does it work? adaptive phase / How does it work? transaction data / Datasets used in machine learning transductive graph label propagation about / Transductive graph label propagation input / Inputs and outputs output / Inputs and outputs working / How does it work? advantages / Advantages and limitations limitations / Advantages and limitations transductive SVM (TSVM) about / Transductive SVM (TSVM) output / Inputs and outputs input / Inputs and outputs working / How does it work? advantages / Advantages and limitations limitations / Advantages and limitations transformations about / Text processing components and transformations Tree augmented network (TAN) about / Tree augmented network input and output / Input and output working / How does it work? advantages and limitations / Advantages and limitations Tunedit about / Datasets URL / Datasets U UCI repository reference link / Data Collection UC Irvine (UCI) database about / Datasets URL / Datasets uncertainty sampling about / Uncertainty sampling working / How does it work? least confident sampling / Least confident sampling smallest margin sampling / Smallest margin sampling label entropy sampling / Label entropy sampling advantages / Advantages and limitations limitations / Advantages and limitations undersampling / Undersampling and oversampling univariate feature analysis about / Univariate feature analysis categorical features / Categorical features continuous features / Continuous features univariate feature selection information theoretic approach / Information theoretic approach statistical approach / Statistical approach unnormalized measure / Factor types unstructured data / Datasets used in machine learning mining, issues / Issues with mining unstructured data unsupervised learning / Machine learning – types and subtypes specific issues / Issues specific to unsupervised learning assumptions / Assumptions and mathematical notations mathematical notations / Assumptions and mathematical notations outlier detection, used / Unsupervised learning using outlier detection usage / Tools and usage US Forest Service (USFS) / Data collection US Geological Survey (USGS) / Data collection V V-Measure about / V-Measure Homogeneity / V-Measure Completeness / V-Measure validation techniques / Training, validation, and test set Variable elimination (VE) algorithm / Variable elimination algorithm variance / Variance vector about / Vector scalar product / Scalar product of vectors vector space model (VSM) about / Vector space model binary / Binary Term Frequency (TF) / Term frequency (TF) inverse document frequency (IDF) / Inverse document frequency (IDF) term frequency-inverse document frequency (TF-IDF) / Term frequency-inverse document frequency (TF-IDF) version space sampling about / Version space sampling Query by disagreement (QBD) / Query by disagreement (QBD) very fast decision trees (VFDT) / Hoeffding trees or very fast decision trees (VFDT) output / Inputs and outputs advantages / Advantages and limitations limitations / Advantages and limitations Very Fast K-means Algorithm (VFKM) / Advantages and limitations visualization analysis about / Visualization analysis univariate feature analysis / Univariate feature analysis multivariate feature analysis / Multivariate feature analysis Vote Entropy disadvanatages / How does it work? W weighted linear sum (WLS) / How does it work? weighted linear sum of squares (WSS) / How does it work? weighted majority algorithm (WMA) about / Weighted majority algorithm input / Inputs and outputs output / Inputs and outputs working / Advantages and limitations advantages / Advantages and limitations limitations / Advantages and limitations Weka URL / Machine learning – tools and datasets about / Machine learning – tools and datasets, Case Study – Horse Colic Classification experiments / Weka experiments Sample end-to-end process, in Java / Sample end-to-end process in Java experimenter / Weka experimenter and model selection model selection / Weka experimenter and model selection Weka Bayesian Network GUI / Weka Bayesian Network GUI Welchs test / Welch's t test Widmer / Widmer and Kubat Wilcoxon signed-rank test / Wilcoxon signed-rank test Word sense disambiguation (WSD) / Word sense disambiguation wrapper approach / Wrapper approach Z Z-Score Normalization / Outliers ZeroMQ Message Transfer Protocol (ZMTP) / Message queueing frameworks ... Errata Piracy Questions Machine Learning Review Machine learning – history and definition What is not machine learning? Machine learning – concepts and terminology Machine learning – types and subtypes... subtypes Datasets used in machine learning Machine learning applications Practical issues in machine learning Machine learning – roles and process Roles Process Machine learning – tools and datasets... distribution Central limit theorem Error propagation Index Mastering Java Machine Learning Mastering Java Machine Learning Copyright © 20 17 Packt Publishing All rights reserved No part of this book