Index
A
- A/B testing / Bayesian Bernoulli bandits
- accuracy / Key quality metrics
- versus model fitness / Model fitness
- action-value iterative update / Action-value iterative update
- activation convolution neural network / Convolution layers
- activation function / Activation function
- Actors
- scalability / Scalability with Actors
- adaptive modeling / Model categorization
- aggregation effect / Tuning the number of clusters
- Akka / Scala as a scalable language
- Akka framework
- about / Akka
- URL, for downloading / Akka
- master-workers design / Master-workers
- futures / Futures
- AlgeBird
- reference link / Algebraic and numerical libraries
- Algebird / Other libraries and frameworks
- algebraic libraries
- alpha (forward pass) / Alpha (forward pass)
- Analysis of Variance (ANOVA) / Challenging model complexity
- anomaly / Anomaly detection with one-class SVC
- anomaly detection
- with one-class SVC / Anomaly detection with one-class SVC
- Apache Commons Math
- reference link / Leveraging Java libraries
- about / Apache Commons Math
- description / Description
- licensing / Licensing
- installation / Installation
- URL, for download / Installation
- Apache Commons Math library
- exceptions, handling / Training workflow
- Apache Mesos resource manager
- URL / Deploying Spark
- Apache Public License 2.0
- URL / Licensing
- Apache Spark / Scala as a scalable language
- about / Apache Spark core, Extending Spark
- need for / Why Spark?
- design principles / Design principles
- experimenting with / Experimenting with Spark
- deploying / Deploying Spark
- URL / Deploying Spark
- shell, using / Using Spark shell
- Kullback-Leibler divergence / Kullback-Leibler divergence
- Kullback-Leibler evaluator / Kullback-Leibler evaluator
- area under PRC / Area under PRC
- area under ROC / Area under ROC
- areaUnderROC (AuROC)
- about / Training the model
- attributes
- about / What is a model?
- Auto-Regressive Integrated Moving Average (ARIMA) / Alternative preprocessing techniques
- Auto-Regressive Moving Average (ARMA) / Alternative preprocessing techniques
- autoencoders
- sparse autoencoder / Sparse autoencoder, Categorization
- undercomplete autoencoder / Undercomplete autoencoder, Categorization, Feed-forward sparse, undercomplete autoencoder
- topology of hidden layers, characteristics / Undercomplete autoencoder
- deterministic autoencoder / Deterministic autoencoder
- complete autoencoder / Categorization
- overcomplete autoencoder / Categorization
- regularized autoencoder / Categorization
- stochastic autoencoder / Categorization
- feed-forward sparse autoencoder / Feed-forward sparse, undercomplete autoencoder
- implementing / Implementation
- autonomous systems
- about / Understanding the challenge
B
- batch EM / Online EM
- batch training
- versus online training / Online versus batch training
- about / Online versus batch training
- Baum-Welch estimator (EM) / Baum-Welch estimator (EM)
- Bayesian Bernoulli bandits / Bayesian Bernoulli bandits
- Bayesian networks
- about / Probabilistic graphical models
- Bellman optimality equations / Bellman optimality equations
- Bernoulli model / Multivariate Bernoulli classification
- best practices, Scala programming
- encapsulation / Encapsulation
- class constructor template / Class constructor template
- companion objects, versus case classes / Companion objects versus case classes
- enumeration, versus case classes / Enumerations versus case classes
- overloading / Overloading
- design template, for immutable classifiers / Design template for immutable classifiers
- beta (backward pass) / Beta (backward pass)
- bias-variance decomposition / Bias-variance decomposition
- BinaryClassificationEvaluator
- about / Validating the model
- BinaryLogisticRegressionSummary
- about / Training the model
- binary restricted Boltzmann machines
- about / Binary restricted Boltzmann machines
- conditional probabilities / Conditional probabilities
- sampling / Sampling
- log-likelihood gradient / Log-likelihood gradient
- contrastive divergence / Contrastive divergence
- configuration parameters / Configuration parameters
- unsupervised learning / Unsupervised learning
- binary SVC
- about / The binary SVC
- LIBSVM / LIBSVM
- design / Design
- configuration parameters / Configuration parameters
- interface, creating to LIBSVM / Interface to LIBSVM
- training / Training
- classification / Classification
- margin / C-penalty and margin
- C-penalty / C-penalty and margin
- C-penalty, optimizing / C-penalty and margin
- kernel, evaluation / Kernel evaluation
- application, to risk analysis / Application to risk analysis
- binomial logistic regression / Step 6 – evaluating the model
- bitwise swap / Fast Fisher-Yates shuffle
- Bloomberg / Naïve Bayes and text mining
- Boltzmann machine / Boltzmann machine
- bootstrapping
- with replacement / Bootstrapping with replacement
- overview / Overview
- resampling / Resampling
- implementation / Implementation
- pros and cons / Pros and cons of bootstrap
- Box-Muller transform
- about / Box-Muller transform
- Breeze / Other libraries and frameworks
- Bregman distance
- about / The divergences
- Broyden-Fletcher-Goldfarb-Shanno (BFGS) / BFGS
- Broyden-Fletcher-Goldfarb-Shannon method / Numerical optimization
C
- cake pattern
- canonical forms / The hidden Markov model (HMM)
- evaluation / The hidden Markov model (HMM)
- training / The hidden Markov model (HMM)
- decoding / The hidden Markov model (HMM)
- category M
- about / Abstraction
- centroid / K-means
- Cholesky factorization
- about / Cholesky factorization
- chromosomes
- about / Evolutionary computing
- class constructor template / Class constructor template
- classification / Step 4 – Classification, Classification
- classification, multilayer perceptron (MLP)
- about / Training and classification
- regularization / Regularization
- model generation / Model generation
- Fast Fisher-Yates shuffle / Fast Fisher-Yates shuffle
- prediction / Prediction
- model fitness / Model fitness
- cluster analysis
- about / K-mean clustering
- clustering
- about / K-mean clustering
- clustering algorithms
- K-means / K-mean clustering, K-means
- Expectation-Maximization / K-mean clustering
- clusters
- defining / Defining clusters
- initializing / Initializing clusters
- clusters assignment, K-means algorithm / Step 2 – Clusters assignment
- clusters configuration, K-means algorithm / Step 1 – Clusters configuration
- CNBC / Naïve Bayes and text mining
- Colt
- reference link / Algebraic and numerical libraries
- companion objects
- versus case classes / Companion objects versus case classes
- complete autoencoder / Categorization
- complex Fourier transform / Fourier analysis
- conditional independence
- about / Probabilistic graphical models
- conditional random field (CRF)
- about / Conditional random fields, Introduction to CRF
- linear chain CRF / Linear chain CRF
- potential functions (fi) / Linear chain CRF
- identity potential functions / Linear chain CRF
- transition feature functions / Linear chain CRF
- state feature functions / Linear chain CRF
- versus hidden Markov model (HMM) / Comparing CRF and HMM
- configuration parameters, binary SVC
- about / Configuration parameters
- SVM formulation / The SVM formulation
- SVM kernel function / The SVM kernel function
- SVM execution / The SVM execution
- confusion matrix
- conjugate gradient / Conjugate gradient
- connectionism / The biological background
- constructive tuning / Regularization
- consumer price index (CPI)
- context bounds / Context bounds
- context free Thompson sampling / Prior/posterior beta distribution
- continuation-passing style (CPS)
- continuous-time Kalman filter / Benefits and drawbacks
- contract
- about / Error handling
- contravariant vectors
- about / Manifolds
- control learning
- about / A solution – Q-learning
- convexity / The kernel trick
- convex minimization / Ln roughness penalty
- convolution layer / Local receptive fields
- convolution neural networks
- about / Convolution neural networks
- local receptive fields / Local receptive fields
- weight sharing / Weight sharing
- convolution layers / Convolution layers
- sub-sampling layers / Sub-sampling layers
- implementing / Putting it all together
- Cooley-Tukey algorithm / Discrete Fourier transform (DFT)
- cosine distance / Measuring similarity
- Counter class / Counter
- covariant functor
- about / Functors
- covariant functor
- about / Functors
- covariant vectors
- about / Manifolds
- cross-validation
- about / Cross-validation
- 1-fold cross-validation / One-fold cross-validation
- K-fold cross-validation / K-fold cross-validation
- crossover
- about / Crossover
- population / Population
- chromosomes / Chromosomes
- genes / Genes
- curse of dimensionality
- about / Curse of dimensionality
- curve-fitting algorithms / Alternative preprocessing techniques
D
- Darwinian process
- about / The origin
- data
- about / Modeling
- profiling / Profiling data
- data clustering / Clustering
- data partitioning / Clustering
- data scientist
- about / Defining a methodology
- data segmentation / Clustering
- DataSourceConfig class / Data extraction
- Davidon-Fletcher-Powell method / Numerical optimization
- DBpedia / Basics information retrieval
- decision boundary / Visualizing model features
- decoding, hidden Markov model (HMM)
- about / Decoding (CF-3)
- Viterbi algorithm / The Viterbi algorithm
- deep belief network (DBM) / Restricted Boltzmann Machines (RBMs)
- Deep belief networks (DBNs) / Boltzmann machine
- Deep Boltzmann machines (DBMs) / Boltzmann machine
- def
- overriding, with val / Understanding the problem
- density estimation / Unsupervised learning
- dependency injection
- deployment modes, Apache Spark
- standalone mode / Deploying Spark
- local mode / Deploying Spark
- Yarn clusters manager / Deploying Spark
- Apache Mesos resource manager / Deploying Spark
- descriptive models / Model categorization
- design
- versus model / Model versus design
- designing / Model versus design
- design principles, Spark
- about / Design principles
- in-memory persistency / In-memory persistency
- laziness / Laziness
- transforms and actions / Transforms and actions
- shared variables / Shared variables
- design template
- for immutable classifiers / Design template for immutable classifiers
- destructive tuning / Regularization
- deterministic autoencoder / Deterministic autoencoder
- DFT-based filtering / DFT-based filtering
- differential operator / Differential operator
- dimension reduction / Dimension reduction
- directed graphical models
- about / Probabilistic graphical models
- Dirichlet Latent Allocation / Retrieving textual information
- Discrete Fourier transform (DFT)
- about / Discrete Fourier transform (DFT)
- discrete Fourier transform (DFT) / Performance
- discrete Kalman filter
- about / The discrete Kalman filter
- state space estimation / The state space estimation
- recursive algorithm / The recursive algorithm
- discrete Markov chain
- about / The Markov property
- discretization / Value encoding
- discretized streams (DStreams)
- about / Discretized streams
- discriminative kernels
- about / Common discriminative kernels
- discriminative models / Discriminative models
- divergences
- about / The divergences
- Kullback-Leibler (KL) divergence / The divergences
- Jensen-Shannon metric / The divergences
- mutual information / The divergences
- Bregman distance / The divergences
- DMatrix class / DMatrix class
- DNA
- about / Evolutionary computing
- DocumentsSource class / Documents extraction
- domain
- about / Defining a methodology
- dynamic programming
- overview / Overview dynamic programming
E
- eclipse Scala IDE
- about / Eclipse Scala IDE
- reference link / Eclipse Scala IDE
- Eigen-decomposition
- about / Eigenvalue decomposition
- encapsulation
- about / Encapsulation
- encoding scheme
- flat encoding / Flat encoding
- hierarchical encoding / Hierarchical encoding
- enumerations
- versus case classes / Enumerations versus case classes
- epoch
- about / Training epoch
- epoch training, multilayer perceptron (MLP)
- about / Training epoch
- input forward propagation / Step 1 – input forward propagation
- error backpropagation / Step 2 – error backpropagation
- exit condition / Step 3 – exit condition
- implementing / Putting it all together
- Epsilon-greedy algorithm / Epsilon-greedy algorithm
- error backpropagation, multilayer perceptron (MLP)
- about / Step 2 – error backpropagation
- weights adjustment / Weights adjustment
- error propagation / Error propagation
- computational model / The computational model
- error handling
- about / Error handling
- error insensitive zone / Overview
- Euclidean distance / Measuring similarity
- European Central Bank
- URL / Financial data sources
- evaluation, hidden Markov model (HMM)
- about / Evaluation (CF-1)
- alpha (forward pass) / Alpha (forward pass)
- beta (backward pass) / Beta (backward pass)
- evaluation, multilayer perceptron (MLP)
- about / Evaluation
- execution profile / Execution profile
- learning rate, impact / Impact of learning rate
- momentum factor, impact / Impact of the momentum factor
- number of hidden layers, impact / Impact of the number of hidden layers
- test case / Test case
- exception handling / Implementation
- exchange-traded funds (ETFs) / Test case
- expectation-maximization (EM) / Training (CF-2)
- Expectation-Maximization algorithm
- about / Expectation-Maximization (EM)
- Gaussian mixture model / Gaussian mixture model
- overview / EM overview
- implementation / Implementation
- classification / Classification
- testing / Testing
- explicit model
- about / Monadic data transformation
- explicit models
- about / Explicit models
- exponential moving average
- about / Exponential moving average
- extended Kalman filter (EKF) / Benefits and drawbacks
- Extended Kalman Filters (EKF) / The discrete Kalman filter
- Extended Learning Classifiers (XCS)
- about / Learning classifier systems
F
- 1-fold cross-validation / One-fold cross-validation
- F-score
- for binomial classification / F-score for binomial classification
- for multinomial classification / F-score for multinomial classification
- F1-measure / Key quality metrics
- F1-score / Key quality metrics
- False Negatives (FNs) / Key quality metrics
- false positive rate (FPR) / Area under ROC
- about / Training the model
- False Positives (FPs) / Key quality metrics
- Fast Fisher-Yates shuffle / Fast Fisher-Yates shuffle
- Fast Fourier Transform (FFT) / Discrete Fourier transform (DFT)
- feature extraction
- about / Extracting features
- feature functions / Linear chain CRF
- features
- about / What is a model?
- features extraction
- automation / Test case 2 – features selection
- features selection
- about / Selecting features
- federal fund rate (FDR)
- feed-forward neural networks (FFNN)
- about / Feed-forward neural networks (FFNN)
- biological background / The biological background
- mathematical background / Mathematical background
- without hidden layers / The multilayer perceptron (MLP)
- feed-forward sparse autoencoder / Feed-forward sparse, undercomplete autoencoder
- filtering
- versus smoothing / The discrete Kalman filter
- filtering algorithms / Modularizing
- final val
- versus val / C-penalty and margin
- finances 101
- about / Finances 101
- fundamental analysis / Fundamental analysis
- technical analysis / Technical analysis
- options trading / Options trading
- financial data sources / Financial data sources
- financial data sources
- about / Financial data sources
- financial metrics
- earnings per share (EPS) / Fundamental analysis
- Price/Earnings Ratio (PE) / Fundamental analysis
- Price/Sales Ratio (PS) / Fundamental analysis
- Price/Book Value Ratio (PB) / Fundamental analysis
- Price to Earnings/Growth (PEG) / Fundamental analysis
- Operating Income / Fundamental analysis
- Net Sales / Fundamental analysis
- Operating Profit Margin / Fundamental analysis
- Net Profit Margin / Fundamental analysis
- Short Interest / Fundamental analysis
- Short Interest Ratio / Fundamental analysis
- Cash per Share / Fundamental analysis
- Pay-out Ratio / Fundamental analysis
- Annual Dividend yield / Fundamental analysis
- Dividend Coverage Ratio / Fundamental analysis
- Growth Domestic Product (GDP) / Fundamental analysis
- Consumer Price Index (CPI) / Fundamental analysis
- Federal Fund rate / Fundamental analysis
- first-order discrete Markov chain
- first order predicate logic / First order predicate logic
- fitness function
- about / Fitness score
- fixed fitness function / Fitness score
- evolutionary fitness function / Fitness score
- approximate fitness function / Fitness score
- fixed lag smoothing / Fixed lag smoothing
- complex strategies / Fixed lag smoothing
- Fn score / Key quality metrics
- forms, models
- parameteric / What is a model?
- differential / What is a model?
- probabilistic / What is a model?
- graphical / What is a model?
- directed graphs / What is a model?
- numerical methods / What is a model?
- chemistry / What is a model?
- taxonomy / What is a model?
- grammar / What is a model?
- lexicon / What is a model?
- inference logic / What is a model?
- Fourier analysis
- about / Fourier analysis
- Discrete Fourier transform (DFT) / Discrete Fourier transform (DFT)
- DFT-based filtering / DFT-based filtering
- market cycles, detecting / Detection of market cycles
- Fourier transform / Fourier analysis
- frameworks
- about / Tools and frameworks
- Java / Java
- Scala / Scala
- SBT / Simple build tool
- Apache Commons Math / Apache Commons Math
- JFreeChart / JFreeChart
- libraries / Other libraries and frameworks
- tools / Other libraries and frameworks
- frequency domain / Discrete Fourier transform (DFT)
- function language
- Scala / Scala as a functional language
- functor
- about / Scala as a functional language
- functors
- about / Functors
- fundamental analysis
- about / Fundamental analysis
- futures
- about / Futures
- blocking / Blocking on futures
- callbacks / Future callbacks
- implementing / Putting it all together
G
- Gauss-Newton method / Numerical optimization, Training workflow
- Gauss-Newton technique
- about / Gauss-Newton
- Gaussian distribution / Z-score and Gauss
- Gaussian mixture / Class likelihood
- Gaussian mixture model
- about / Gaussian mixture model
- Gaussian noise / The transition equation
- Gaussian sampling
- about / Gaussian sampling
- Box-Muller transform / Box-Muller transform
- generalized autoregressive conditional heteroscedasticity (GARCH) / Alternative preprocessing techniques
- generalized linear models (GLM) / Logistic regression
- generative models / Generative models
- Generic message handler
- about / Blocking on futures
- genetic algorithm, implementations
- about / Implementation
- software design / Software design
- key components / Key components
- selection process / Selection
- population growth, controlling / Controlling population growth
- configuration / GA configuration
- crossover / Crossover
- mutation / Mutation
- reproduction / Reproduction
- solver / Solver
- genetic algorithms (GA)
- about / Genetic algorithms and machine learning
- for trading strategies / GA for trading strategies
- advantages / Advantages and risks of genetic algorithms
- risks / Advantages and risks of genetic algorithms
- genetic algorithms components
- about / Genetic algorithm components
- genetic encoding / Encodings
- genetic operators / Genetic operators
- fitness function / Fitness score
- genetic encoding
- about / Encodings
- value encoding / Value encoding
- predicate encoding / Predicate encoding
- solution encoding / Solution encoding
- encoding scheme / The encoding scheme
- genetic operators
- about / Genetic operators
- selection / Genetic operators, Selection
- crossover / Genetic operators, Crossover
- mutation / Genetic operators, Mutation
- Gibbs sampling / Test
- G measure / Key quality metrics
- GNU Lesser General Public License (LGPL)
- about / Licensing
- goal state
- about / Putting it all together
- Google finances
- reference link / Financial data sources
- Googles Breeze
- reference link / Abstraction
- gradient descent methods
- steepest descent / Steepest descent
- conjugate gradient / Conjugate gradient
- stochastic gradient descent / Stochastic gradient descent
- graph-structured CRF
- about / Introduction to CRF
- graphical models
- about / Probabilistic graphical models
- GraphX
- about / Overview
- greedy approach
- about / Solver
- grid search
- about / Grid search
- growth domestic product (GDP)
H
- Hadoop Distributed File System (HDFS) / Step 2 – loading data
- Hadoop distributed file system (HDFS)
- about / Overview
- hard margin / The separable case (hard margin)
- heat kernel function / Kernel monadic composition
- Hessian matrix / Jacobian and Hessian matrices
- hidden layers / The multilayer perceptron (MLP)
- hidden Markov model (HMM)
- about / The hidden Markov model (HMM)
- notation / Notation
- lambda model / The lambda model
- design / Design
- evaluation (CF-1) / Evaluation (CF-1)
- training (CF-2) / Training (CF-2)
- decoding (CF-3) / Decoding (CF-3)
- implementing / Putting it all together
- ViterbiPath class / Putting it all together
- ViterbiPath singleton / Putting it all together
- test case / Test case 1 – Training, Test case 2 – Evaluation
- as filtering technique / HMM as filtering technique
- versus conditional random field (CRF) / Comparing CRF and HMM
- performance consideration / Performance consideration
- Hidden Naïve Bayes (HNB) / Training
- higher kinded types (HKTs)
- about / Higher kinded types
- hinge loss / The non-separable case (soft margin)
- HMM constructor
- config argument / Putting it all together
- xt argument / Putting it all together
- form argument / Putting it all together
- quantize argument / Putting it all together
I
- identically distributed population (i.i.d) / The purpose of sampling
- IITB CRF Java library
- evaluation / Training the CRF model
- immutable statistics / Immutable statistics
- immutable transformations
- about / Explicit models
- implementation, regularized CRF
- about / Implementation
- CRF classifier, configuring / Configuring the CRF classifier
- CRF model, training / Training the CRF model
- CRF model, applying / Applying the CRF model
- implicit model
- about / Monadic data transformation
- implicit models
- about / Implicit models
- incremental EM / Online EM
- initialization
- about / Genetic operators
- input forward propagation, multilayer perceptron (MLP)
- about / Step 1 – input forward propagation
- computational flow / Computational flow
- error functions / Error functions
- operating modes / Operating modes
- softmax / Softmax
- input value
- about / Error handling
- IntelliJ IDEA Scala plugin
- about / IntelliJ IDEA Scala plugin
- reference link / IntelliJ IDEA Scala plugin
- intercept / Step 5 – implementing the classifier
- iterators / Lazy views
J
- Jacobian J matrix / Numerical optimization
- Jacobian matrix / Jacobian and Hessian matrices
- Java
- Java libraries
- leveraging / Leveraging Java libraries
- Java Native Interface (JNI) / Algebraic and numerical libraries
- Java Virtual Machine (JVM)
- about / Overview
- JBlas
- reference link / Leveraging Java libraries
- jBlas
- reference link / Algebraic and numerical libraries
- Jensen-Shannon metric
- about / The divergences
- JFreeChart
- about / JFreeChart, Plotting data
- description / Description
- licensing / Licensing
- installation / Installation
- URL, for installation / Installation
K
- K-armed bandit
- about / K-armed bandit
- exploration-exploitation trade-offs / Exploration-exploitation trade-offs
- expected cumulative regret / Expected cumulative regret
- Bayesian Bernoulli bandits / Bayesian Bernoulli bandits
- Epsilon-greedy algorithm / Epsilon-greedy algorithm
- K-fold cross-validation / K-fold cross-validation
- K-means
- with MLlib / K-means using MLlib
- K-means algorithm
- defining / Defining the algorithm
- steps / Defining the algorithm
- exit condition / Step 2 – Clusters assignment
- K-means clustering
- about / K-means
- similarity measures / Measuring similarity
- evaluation, setting up / Evaluation
- results, evaluating / The results
- number of clusters, tuning / Tuning the number of clusters
- output, validating / Validation
- K-means components
- creating / Creating K-means components
- Kalman filter / The discrete Kalman filter
- recursive characteristic / The discrete Kalman filter
- optimal characteristic / The discrete Kalman filter
- non-linear systems / The discrete Kalman filter
- Kalman smoothing / Kalman smoothing
- Kernel functions
- about / Kernel functions
- overview / Overview
- discriminative kernels / Common discriminative kernels
- monadic composition / Kernel monadic composition
- monadic composition, interpretation / Kernel monadic composition
- in SVM / Kernel monadic composition
- kernel functions, types
- linear kernel (dot product) / Common discriminative kernels
- polynomial kernel / Common discriminative kernels
- radial basis function (RBF) / Common discriminative kernels
- sigmoid kernel / Common discriminative kernels
- Laplacian kernel / Common discriminative kernels
- log kernel / Common discriminative kernels
- Kernel PCA
- about / Kernel PCA
- kernels, types
- probabilistic kernels / Common discriminative kernels
- smoothing kernels / Common discriminative kernels
- reproducible kernel Hilbert spaces / Common discriminative kernels
- kernel trick
- about / The kernel trick
- key components
- population / Population
- chromosomes / Chromosomes
- genes / Genes
- Kohonen's self-organizing maps / Manifolds
- Kullback-Leibler (KL) divergence
- about / The divergences, The Kullback-Leibler divergence
- overview / Overview
- implementation / Implementation
- testing / Testing
- Kullback-Leibler divergence
- about / Kullback-Leibler divergence
- implementation / Implementation
- Kullback-Leibler evaluator
- about / Kullback-Leibler evaluator
L
- L1 regularization / Challenging model complexity
- Lagrange multipliers / Max-margin classification, Lagrange multipliers
- Laplacian Eigenmaps
- about / Manifolds
- Laplacian kernel / Common discriminative kernels
- Lasso regularization / Ln roughness penalty
- Latent Dirichlet Allocation (LDA)
- about / Probabilistic graphical models
- lazy direct acyclic graph (DAG)
- about / Use case – continuous parsing
- lazy value trigger / Step 3 – instantiation
- lazy views / Lazy views
- LDL decomposition
- about / LDL decomposition
- Learning Classifier Systems (LCS)
- about / Learning classifier systems, Introduction to LCS
- components / Introduction to LCS
- learning strategy, combining with evolutionary approach / Combining learning and evolution
- terminology / Terminology
- extended learning classifier systems / Extended learning classifier systems
- XCS components / XCS components
- portfolio management, application / Application to portfolio management
- XCS core data / XCS core data
- XCS rules / XCS rules
- covering phase / Covering
- implementation, example / Example of implementation
- benefits / Benefits and limitations of learning classifier systems
- limitations / Benefits and limitations of learning classifier systems
- learning vector quantization (LVQ)
- about / K-mean clustering
- least squares problem / Numerical optimization
- lemmatization / Basics information retrieval
- Levenberg-Marquardt algorithm
- about / Levenberg-Marquardt
- Levenberg-Marquardt method / Numerical optimization, Training workflow
- Levenberg-Marquardt optimizer / Alternative preprocessing techniques
- LIBSVM
- about / LIBSVM
- URL / LIBSVM
- reference / LIBSVM
- need for / LIBSVM
- Java code / LIBSVM
- svm_node / Interface to LIBSVM
- scaling / Application to risk analysis
- LIBSVM, Java class
- linear algebra
- about / Linear algebra
- QR decomposition / QR decomposition
- LU factorization / LU factorization
- LDL decomposition / LDL decomposition
- Cholesky factorization / Cholesky factorization
- SVD / Singular Value Decomposition (SVD)
- Eigen-decomposition / Eigenvalue decomposition
- algebraic libraries / Algebraic and numerical libraries
- numerical libraries / Algebraic and numerical libraries
- linear chain CRF
- about / Introduction to CRF, Linear chain CRF
- linear chain structured graph CRF
- about / Introduction to CRF
- linear kernel (dot product) / Common discriminative kernels
- linear regression
- about / Linear regression
- univariate linear regression / Univariate linear regression
- ordinary least squares regression (OLS) / Ordinary least squares (OLS) regression
- concept / Test case 2 – features selection
- versus support vector regression (SVR) / SVR versus linear regression
- linear SVM
- about / The linear SVM
- separable case (hard margin) / The separable case (hard margin)
- non-separable case (soft margin) / The non-separable case (soft margin)
- Local Linear Embedding
- about / Manifolds
- logistic regression
- about / Logistic regression
- logistic function / Logistic function
- design / Design
- training workflow / Training workflow
- classification / Classification
- logistic regression, test case
- about / Let's kick the tires
- workflow, writing / Writing a simple workflow
- issues, scoping / Step 1 – scoping the problem
- data, loading / Step 2 – loading data
- data, preprocessing / Step 3 – preprocessing data, Immutable normalization
- immutable normalization / Immutable normalization
- patterns, discovering / Step 4 – discovering patterns
- data, analyzing / Analyzing data
- data, plotting / Plotting data
- model features, visualizing / Visualizing model features
- label, visualizing / Visualizing label
- classifier, implementing / Step 5 – implementing the classifier
- optimizer, selecting / Selecting an optimizer
- model, training / Training the model
- observations, classifying / Classifying observations
- model, evaluating / Step 6 – evaluating the model
- log kernel / Common discriminative kernels
- loss function / Selecting an optimizer
- loss function approach
- about / Solver
- Lotka-Volterra equation
- about / Selection
- LU factorization
- about / LU factorization
M
- machine learning
- need for / Why machine learning?
- classification / Classification
- prediction / Prediction
- optimization / Optimization
- regression / Regression
- about / Genetic algorithms and machine learning
- machine learning algorithms
- taxanomy / Taxonomy of machine learning algorithms
- unsupervised learning / Unsupervised learning
- supervised learning / Supervised learning
- discriminative models / Discriminative models
- semi-supervised learning / Semi-supervised learning
- reinforcement learning / Reinforcement learning
- macro formulas
- for multinomial precision / F-score for multinomial classification
- for recall / F-score for multinomial classification
- Manhattan distance / Measuring similarity
- manifolds
- about / Manifolds
- Markov chain
- about / The hidden Markov model (HMM)
- Markov Chain Monte Carlo (MCMC)
- about / Markov Chain Monte Carlo (MCMC)
- overview / Overview
- Metropolis-Hastings (MH) / Metropolis-Hastings (MH)
- implementation / Implementation
- testing / Test
- Markov decision process (MDP) / K-armed bandit
- Markov decision processes
- about / Markov decision processes
- Markov property / The Markov property
- first-order discrete Markov chain / The first-order discrete Markov chain
- Markov property
- about / The Markov property
- master-workers design
- about / Master-workers
- messages exchange / Messages exchange
- worker Actors / Worker Actors
- workflow controller / The workflow controller
- master Actor / The master Actor
- with routing / Master with routing
- DFT / Distributed discrete Fourier transform
- limitations / Limitations
- mathematical abstractions
- supporting / Supporting mathematical abstractions
- variable declaration / Step 1 – variable declaration
- model definition / Step 2 – model definition
- instantiation / Step 3 – instantiation
- mathematical notations / Mathematical notations for the curious
- mathematics
- linear algebra / Linear algebra
- first order predicate logic / First order predicate logic
- Hessian matrix / Jacobian and Hessian matrices
- Jacobian matrix / Jacobian and Hessian matrices
- optimization techniques / Summary of optimization techniques
- max-margin classification
- about / Max-margin classification
- mean / Immutable statistics
- mean square error (MSE) / Bias-variance decomposition
- measurement equation / The state space estimation, The measurement equation
- measurement noise covariance / The measurement equation
- memory management
- about / Explicit models
- methodology
- defining / Defining a methodology
- Metropolis-Hastings (MH) / Metropolis-Hastings (MH)
- mixin composition
- for ITransform / Instantiating the workflow
- mixins
- composing, to build workflow / Composing mixins to build workflow
- about / Composing mixins to build workflow
- mixins linearization
- about / Understanding the problem
- MLlib library
- about / Overview, MLlib library
- components / Overview
- RDDs, creating / Creating RDDs
- using, for K-means / K-means using MLlib
- tests / Tests
- model
- about / What is a model?
- versus design / Model versus design
- model assessment
- about / Assessing a model
- validation / Validation
- area under curves / Area under the curves
- cross-validation / Cross-validation
- bias-variance decomposition / Bias-variance decomposition
- overfitting / Overfitting
- model categorization
- about / Model categorization
- predictive models / Model categorization
- descriptive models / Model categorization
- model complexity
- challenging / Challenging model complexity
- model fitness
- versus accuracy / Model fitness
- modeling
- about / Modeling
- models
- forms / What is a model?
- model validation
- about / Validation
- key quality metrics / Key quality metrics
- F-score, for binomial classification / F-score for binomial classification
- F-score, for multinomial classification / F-score for multinomial classification
- modules
- defining / Defining modules
- monad
- about / Scala as a functional language
- monadic composition
- about / Monads
- monadic data transformation
- about / Monadic data transformation
- monads
- about / Monads, Monads to the rescue
- Monitor class / Monitor
- Monte Carlo approximation
- about / Monte Carlo approximation
- overview / Overview
- implementation / Implementation
- Monte Carlo EM / Online EM
- Monte Carlo integration / Sampling
- morphism
- about / Error handling
- moving averages
- about / Moving averages
- simple moving average / Simple moving average
- weighted moving average / Weighted moving average
- exponential moving average / Exponential moving average
- on multi-dimensional time series / Exponential moving average
- multi-class scoring / F-score for binomial classification
- multilayer perceptron (MLP)
- about / The multilayer perceptron (MLP)
- activation function / Activation function
- network topology / Network topology
- design / Design
- configuration / Configuration
- network components / Network components
- model / Model
- problem types (modes) / Problem types (modes)
- online training, versus batch training / Online versus batch training
- epoch, training / Training epoch
- training / Training and classification
- classification / Training and classification
- evaluation / Evaluation
- limitations / Benefits and limitations
- benefits / Benefits and limitations
- multinomial Naïve Bayes
- about / Introducing the multinomial Naïve Bayes
- formalism / Formalism
- frequentist perspective / The frequentist perspective
- predictive model / The predictive model
- zero-Frequency problem / The zero-frequency problem
- multivariate Bernoulli classification
- about / Multivariate Bernoulli classification
- model / Model
- implementation / Implementation
- mutation
- about / Mutation
- population / Population
- chromosome / Chromosomes
- genes / Genes
- mutual information
- about / The divergences, The mutual information
N
- n-fold cross-validation / Application to risk analysis
- NASDAQ
- URL / Financial data sources
- natural language processing (NLP) / The feature functions model
- natural selection
- about / Selection
- Naïve Bayes
- pros and cons / Pros and cons
- Naïve Bayes classifier
- used, for text mining / Naïve Bayes and text mining
- Naïve Bayes classifiers
- about / Naïve Bayes classifiers
- multinomial Naïve Bayes / Introducing the multinomial Naïve Bayes
- implementation / Implementation
- design / Design
- training / Training
- classification / Classification
- F1 Validation / F1 Validation
- features extract / Features extraction
- testing / Testing
- Naïve Bayes models
- about / Probabilistic graphical models
- network components, multilayer perceptron (MLP)
- about / Network components
- network topology / Network topology
- hidden layers / Input and hidden layers
- input layers / Input and hidden layers
- output layers / Output layer
- synapses / Synapses
- connections / Connections
- weights initialization / Weights initialization
- news
- macro trends / Naïve Bayes and text mining
- micro updates / Naïve Bayes and text mining
- Nondeterministic Polynomial (NP) problems
- about / NP problems
- categories / NP problems
- nonlinear least squares minimization
- about / Nonlinear least squares minimization
- Gauss-Newton technique / Gauss-Newton
- Levenberg-Marquardt algorithm / Levenberg-Marquardt
- non linear models
- about / Nonlinear models
- Kernel PCA / Kernel PCA
- manifolds / Manifolds
- nonlinear SVM
- about / The nonlinear SVM
- max-margin classification / Max-margin classification
- kernel trick / The kernel trick
- normalization
- about / Immutable normalization
- normalized inner product / Measuring similarity
- null frequencies
- handling / Implementation
- numerical libraries
- numerical optimization
- about / Numerical optimization
- Newton (or second-order techniques) / Numerical optimization
- Quasi-newton (or first-order techniques) / Numerical optimization
- Nyquist / Discrete Fourier transform (DFT)
O
- object oriented language
- observation
- about / Extracting features
- one-class SVC
- anomaly detection / Anomaly detection with one-class SVC
- online EM / Online EM
- online training
- versus batch training / Online versus batch training
- about / Online versus batch training
- operations, time series
- transpose operator / Transpose operator
- differential operator / Differential operator
- optimization techniques
- gradient descent methods / Steepest descent
- Quasi Newton algorithms / Quasi-Newton algorithms
- nonlinear least squares minimization / Nonlinear least squares minimization
- Lagrange multipliers / Lagrange multipliers
- dynamic programming, overview / Overview dynamic programming
- options trading
- about / Options trading
- option trading
- Q-learning, used / Option trading using Q-learning
- option property / Option property
- option model / Option model
- quantization / Quantization
- ordinary least squares regression (OLS)
- about / Ordinary least squares (OLS) regression
- design / Design
- implementation / Implementation
- test case / Test case 1 – trending, Test case 2 – features selection
- output unit activation function / Activation function
- output value
- about / Error handling
- overcomplete autoencoder / Categorization
- overfitting / Overfitting
- emprical estimation / Bias-variance decomposition
- versus sparsity / Sparsity updating equations
- overloading / Overloading
P
- padding / Value encoding
- parallel collections
- about / Parallel collections
- processing / Processing a parallel collection
- benchmark framework / Benchmark framework
- performance evaluation / Performance evaluation
- Parallel Colt
- reference link / Leveraging Java libraries
- parent chromosomes
- preserving / Crossover
- partial functions
- reusability / Error handling
- about / Error handling
- runtime validation / Error handling
- versus partially applied functions / DFT-based filtering
- Partial Least Square Regression (PLSR) / Validation
- partially applied functions
- versus partial functions / DFT-based filtering
- partially connected networks / Network topology
- particle filter / Alternative preprocessing techniques
- penalized least squares regression / Ln roughness penalty
- performance evaluation, Spark
- about / Performance evaluation
- tuning parameters / Tuning parameters
- considerations / Performance considerations
- pros and cons / Pros and cons
- plate model
- about / Probabilistic graphical models
- polynomial kernel / Common discriminative kernels
- population growth
- controlling / Controlling population growth
- pre-processing techniques
- alternative techniques / Alternative preprocessing techniques
- precision / Key quality metrics
- precision-recall curve (PRC) / Area under PRC
- Predicted Residual Error Sum of Squares (PRESS) / Validation
- predictive models / Model categorization
- price pattern
- about / Price patterns
- principal components analysis (PCA)
- about / Principal components analysis (PCA)
- algorithm / Algorithm
- covariance matrix / Algorithm
- implementation / Implementation
- test case / Test case
- evaluation / Evaluation
- extending / Extending PCA
- validation / Validation
- categorical features / Categorical features
- performance / Performance
- probabilistic graphical models
- about / Probabilistic graphical models
- directed graphical models / Probabilistic graphical models
- Bayesian networks / Probabilistic graphical models
- Naïve Bayes models / Probabilistic graphical models
- probabilistic kernels / Common discriminative kernels
- projection
- about / Functors
- proposal distribution / Overview
- Proteins / Overview
- protein sequence annotation / Overview
- Python
- reference link / Overview
Q
- Q-learning, implementation
- about / Implementation
- software design / Software design
- states / The states and actions
- actions / The states and actions
- search space / The search space
- action-value / The policy and action-value
- policy / The policy and action-value
- components / The Q-learning components
- training / The Q-learning training
- tail recursion / Tail recursion to the rescue
- validation / Validation
- prediction / The prediction
- Q-learning algorithm
- about / A solution – Q-learning
- terminology / Terminology
- concept / Concept
- value of policy / Value of policy
- Bellman optimality equations / Bellman optimality equations
- temporal difference, for model free learning / Temporal difference for model-free learning
- action-value iterative update / Action-value iterative update
- used, for option trading / Option trading using Q-learning
- QR Decomposition
- about / QR decomposition
- QStar class / The Viterbi algorithm
- Quandl
- URL / Financial data sources
- quantization / Value encoding, Quantization
- Quasi Newton algorithms
R
- radial basis function (RBF) / Common discriminative kernels
- terminology / Common discriminative kernels
- recall / Key quality metrics
- receiver operating characteristic (ROC)
- about / Training the model
- Receiver Operating Characteristics (ROC) / Area under ROC
- recombination
- about / Evolutionary computing
- reconstruction error minimization
- about / Step 3 – Reconstruction error minimization
- tail recursive implementation / Tail recursive implementation
- iterative implementation / Iterative implementation
- recursive algorithm
- about / The recursive algorithm
- prediction / Prediction
- correction / Correction
- Kalman smoothing / Kalman smoothing
- fixed lag smoothing / Fixed lag smoothing
- experimentation / Experimentation
- benefits / Benefits and drawbacks
- drawbacks / Benefits and drawbacks
- regression
- about / Regression
- regression model / Design
- regularization
- about / Regularization, Ln roughness penalty
- Ln roughness penalty / Ln roughness penalty
- in machine learning / Ln roughness penalty
- model estimation / Ln roughness penalty
- feature selection / Ln roughness penalty
- overfitting / Ln roughness penalty
- computation / Ln roughness penalty
- ridge regression / Ridge regression
- regularized autoencoder / Categorization
- regularized CRF
- text analytics / Regularized CRF and text analytics
- feature functions model / The feature functions model
- design / Design
- implementation / Implementation
- testing / Tests
- Reinforcement learning
- Q-learning, implementation / Implementation
- Q-learning, used for option trading / Option trading using Q-learning
- reinforcement learning / Model categorization, Reinforcement learning, K-armed bandit
- about / Reinforcement learning
- Q-learning algorithm / A solution – Q-learning
- implementing / Putting it all together
- evaluation / Evaluation
- pros and cons / Pros and cons of reinforcement learning
- relative fitness degradation
- about / Selection
- relative strength index (RSI) / Terminology
- replicate / Resampling
- reproducible kernel Hilbert spaces / Common discriminative kernels
- resampling / Overview
- resilient distributed dataset (RDD)
- about / Overview, Apache Spark core, Using Spark shell
- creating / Creating RDDs
- Restricted Boltzmann machines (RBMs) / Restricted Boltzmann Machines (RBMs)
- reusable ML pipelines
- about / Reusable ML pipelines
- Apache Spark application, debugging with ScalaTest / Apache Spark and ScalaTest
- reusable ML transforms
- about / Reusable ML transforms
- features, encoding / Encoding features
- model, training / Training the model
- predictive model / Predictive model
- summary statistics, training / Training summary statistics
- model, validating / Validating the model
- grid search / Grid search
- ridge regression / Ln roughness penalty
- about / Ridge regression
- design / Design
- implementation / Implementation
- test case / Test case
S
- @specialized annotation / Discrete Fourier transform (DFT)
- sampling
- purpose / The purpose of sampling
- Scala
- about / Why Scala?, Scala, Scala
- used, as functional language / Scala as a functional language
- abstraction / Abstraction
- HKTs / Higher kinded types
- functors / Functors
- monads / Monads
- used, as object oriented language / Scala as an object oriented language
- used, as scalable language / Scala as a scalable language
- eclipse Scala IDE / Eclipse Scala IDE
- IntelliJ IDEA Scala plugin / IntelliJ IDEA Scala plugin
- time series / Time series in Scala
- object, creating / Object creation
- Streams / Streams
- parallel collections / Parallel collections
- reference link / Overview
- scalability
- with Actors / Scalability with Actors
- Actor model / The Actor model
- partitioning / Partitioning
- reactive programming / Beyond Actors – reactive programming
- scalable language
- Scala / Scala as a scalable language
- ScalaNLP / Other libraries and frameworks
- Scala programming
- about / Scala programming
- libraries / List of libraries and tools
- tools / List of libraries and tools
- code snippet fromat / Code snippets format
- Scala reactive library
- example, reference link / Beyond Actors – reactive programming
- Scala standard library
- reference link / Scala
- Scalaz
- reference link / Abstraction
- scientific model
- about / What is a model?
- selection process
- about / Selection
- self-reference
- semi-supervised learning / Semi-supervised learning
- Sequential Minimal Optimization (SMO) / The non-separable case (soft margin), LIBSVM
- service level agreement (SLA)
- need for / Why streaming?
- shared variables
- about / Shared variables
- broadcast values / Shared variables
- accumulator variables / Shared variables
- shrinkage
- about / Ln roughness penalty
- sigmoid activation
- versus tanh / Weight sharing
- sigmoid kernel / Common discriminative kernels
- similarity
- visualization / Overview
- similarity measures
- Manhattan distance / Measuring similarity
- Euclidean distance / Measuring similarity
- normalized inner product / Measuring similarity
- simple build tool (sbt)
- about / Deploying Spark
- Simple Build Tool (SBT)
- about / Simple build tool
- reference link / Simple build tool
- simple moving average
- about / Simple moving average
- singular value decomposition (SVD) / Performance
- Singular Value Decomposition (SVD)
- smoothing
- versus filtering / The discrete Kalman filter
- smoothing kernels / Common discriminative kernels
- soft margin / The non-separable case (soft margin)
- software developer
- about / Defining a methodology
- source code, Scala
- about / Source code
- conventions / Convention
- context bounds / Context bounds
- presentation / Presentation
- primitives / Primitives and implicits
- implicits / Primitives and implicits
- immutability / Immutability
- SparkSQL
- about / Overview
- Spark Streaming
- sparse autoencoder
- about / Sparse autoencoder
- sparsity
- versus overfitting / Sparsity updating equations
- sparsity updating equations / Sparsity updating equations
- spectral density estimation / Fourier analysis
- Spectral theory / Fourier analysis
- stackable trait injection
- state space estimation
- about / The state space estimation
- transition equation / The state space estimation, The transition equation
- measurement equation / The state space estimation, The measurement equation
- steepest descent / Steepest descent
- stemming / Basics information retrieval
- steps, K-means algorithms
- clusters configuration / Step 1 – Clusters configuration
- clusters assignment / Step 2 – Clusters assignment
- reconstruction error minimization / Step 3 – Reconstruction error minimization
- classification / Step 4 – Classification
- stepwise EM / Online EM
- stimuli / The biological background
- stochastic autoencoder / Categorization
- stochastic gradient descent / Stochastic gradient descent
- Stochastic Gradient Descent (SGD) / Selecting an optimizer
- streaming engine
- about / Streaming engine
- need for / Why streaming?
- batch processing / Batch and real-time processing
- real-time processing / Batch and real-time processing
- architecture / Architecture overview
- discretized streams (DStreams) / Discretized streams
- continuous parsing, use case / Use case – continuous parsing
- checkpointing / Checkpointing
- Streams
- about / Streams
- memory, allocating / Memory on demand
- memory, reusing designs / Design for reusing Streams memory
- streams / Lazy views
- subject-matter expert
- about / Defining a methodology
- sum of the squared error (SSE) / Online versus batch training
- supervised learning
- about / Supervised learning
- generative models / Generative models
- support vector classifier (SVC)
- about / Support vector classifier (SVC)
- binary SVC / The binary SVC
- support vector machine (SVM)
- about / The support vector machine (SVM)
- optional mathematical formulation / The support vector machine (SVM)
- linear SVM / The linear SVM
- nonlinear SVM / The nonlinear SVM
- support vector classifier (SVC) / Support vector classifier (SVC)
- anomaly detection, with one-class SVC / Anomaly detection with one-class SVC
- support vector regression (SVR) / Support vector regression (SVR)
- performance considerations / Performance considerations
- support vector regression (SVR)
- about / Support vector regression (SVR)
- overview / Overview
- versus linear regression / SVR versus linear regression
- L2 regularization / SVR versus linear regression
- SVM dual problem / Max-margin classification
- SVM model
- accuracy / Training
T
- tagging / The feature functions model
- tagging model / Basics information retrieval
- tail recursion
- about / Tail recursion to the rescue
- tanh
- versus sigmoid activation / Weight sharing
- technical analysis
- about / Technical analysis
- terminology / Terminology
- trading data / Trading data
- trading signal / Trading signal and strategy
- trading strategy / Trading signal and strategy
- price patterns / Price patterns
- temporal difference
- for model free learning / Temporal difference for model-free learning
- tensors
- about / Manifolds
- test case, multilayer perceptron (MLP)
- about / Test case
- implementation / Implementation
- models evaluation / Models evaluation
- hidden layers' architecture, impact / Impact of hidden layers' architecture
- testing, regularized CRF
- about / Tests
- convergence profile, training / The training convergence profile
- training set size impact, evaluating / Impact of the size of the training set
- L2 regularization factor, evaluating / Impact of L2 regularization factor
- text analytics
- text mining
- with Naïve Bayes classifier / Naïve Bayes and text mining
- information retrieval / Basics information retrieval
- implementation / Implementation
- documents, analyzing / Analyzing documents
- relative terms frequency, extracting / Extracting relative terms frequency
- features, generating / Generating the features
- testing / Testing
- textual information, retrieving / Retrieving textual information
- classifier, evaluating / Evaluating text mining classifier
- theory of evolution
- about / Evolution
- origin / The origin
- Nondeterministic Polynomial (NP) problems / NP problems
- evolutionary computing / Evolutionary computing
- Thompson sampling
- about / Thompson sampling
- Bandit context / Bandit context
- prior/posterior beta distribution / Prior/posterior beta distribution
- implementation / Implementation
- simulated exploration / Simulated exploration and exploitation
- simulated exploitation / Simulated exploration and exploitation
- versus UCB1 algorithm / Implementation
- time-domain function / Discrete Fourier transform (DFT)
- time dependency model / The measurement equation
- time series
- about / Time series in Scala
- context bound / Context bounds
- types / Types and operations
- operations / Types and operations
- lazy views / Lazy views
- trading data
- about / Trading data
- trading signal
- about / Trading signal and strategy
- trading strategies
- GA / GA for trading strategies
- definition / Definition of trading strategies
- operators / Trading operators
- cost function / The cost function
- market signals / Market signals
- about / Trading strategies
- signal encoding / Signal encoding
- test case / Test case – Fall 2008 market crash
- creating / Creating trading strategies
- optimizer, configuring / Configuring the optimizer
- finding / Finding the best trading strategy
- tests / Tests
- weighted score / The weighted score
- unweighted score / The unweighted score
- trading strategy
- about / Trading signal and strategy
- training, hidden Markov model (HMM)
- about / Training (CF-2)
- Baum-Welch estimator (EM) / Baum-Welch estimator (EM)
- training, Naïve Bayes classifiers
- about / Training
- Likelihood class / Class likelihood
- binomial model / Binomial model
- multinomial model / Multinomial model
- classifier components / Classifier components
- training files
- raw dataset / Training the CRF model
- tagged dataset / Training the CRF model
- training workflow, logistic regression
- about / Training workflow
- optimizer, configuring / Step 1 – configuring the optimizer
- Jacobian matrix, computing / Step 2 – computing the Jacobian matrix
- convergence of optimizer, managing / Step 3 – managing the convergence of optimizer
- least squares problem, defining / Step 4 – defining the least squares problem
- sum of square errors, minimizing / Step 5 – minimizing the sum of square errors
- testing / Test
- traits
- transition equation
- transpose operator / Transpose operator
- transposition operator
- about / Genetic operators
- TrueFx (Forex)
- URL / Financial data sources
- True Negatives (TNs) / Key quality metrics
- true positive rate (TPR)
- about / Training the model
- True Positives (TPs) / Key quality metrics
- tuning, genetic algorithm / Mutation
- tuning parameters
- about / Performance evaluation
- Twitters Algebird
- reference link / Abstraction
- two-step lag smoothing / Experimentation
U
- UCB1 algorithm
- versus Thompson sampling / Implementation
- unapply method
- about / Genes
- undercomplete autoencoder / Undercomplete autoencoder, Categorization, Feed-forward sparse, undercomplete autoencoder
- univariate linear regression
- about / Univariate linear regression
- implementation / Implementation
- test case / Test case
- unsupervised learning / Unsupervised learning
- data clustering / Clustering
- dimension reduction / Dimension reduction
- upper bound confidence
- about / Upper bound confidence
- confidence interval / Confidence interval
- for Bernoulli variables / Confidence interval
- implementation / Implementation
- utility classes, Scala programming
- about / Utility classes
- data extraction / Data extraction
- financial data sources / Financial data sources
- documents extraction / Documents extraction
- DMatrix class / DMatrix class
- Counter class / Counter
- Monitor class / Monitor
V
- val
- def, overriding with / Understanding the problem
- versus final val / C-penalty and margin
- variables
- about / What is a model?
- variance / Immutable statistics
- variance-bias trade-off / Bias-variance decomposition
- vector quantization
- about / K-mean clustering
- views / Lazy views
- Viterbi algorithm / The Viterbi algorithm
- psi matrix / The Viterbi algorithm
- qStar matrix / The Viterbi algorithm
- delta matrix / The Viterbi algorithm
W
- weight decay / Ln roughness penalty
- weighted moving average
- about / Weighted moving average
- While loop / Discrete Fourier transform (DFT)
- white noise / The transition equation
- WordNet / Basics information retrieval
- workflow
- instantiating / Instantiating the workflow
- modularizing / Modularizing
- workflow computational model
- about / Workflow computational model
Y
- 1-year Treasury bill (1yTB)
- Yahoo finances
- reference link / Financial data sources
Z
- Z-score / Z-score and Gauss