Index
A
- A/B tests
- abstraction
- activation function
- ActivationFunction interface
- activity recognition
- AdaBoost M1 method
- ADADELTA / Learning rate optimization
- ADAGRAD
- Adaline
- adaptive neural networks
- adaptive resonance theory (ART)
- advanced modelling
- affinity analysis
- agglomerative clustering
- AI
- AI transition
- Akaike Information Criteria (AIC) / How many clusters?
- AlphaGo
- Amazon Machine Learning / Machine learning as a service
- analysis types
- Android Device Monitor
- Android Studio
- anomalous behaviour detection
- anomalous pattern detection
- anomaly detection, in time series data
- anomaly detection, in website traffic
- Apache Mahout
- Apache Spark
- Application Portfolio Management (APM)
- Applied Machine Learning
- Apriori
- Apriori algorithm
- architecture, neural networks
- artificial neural networks
- artificial neural networks (ANN) / How learning helps solving problems
- artificial neural networks (ANNs)
- artificial neuron
- association rule learning
- association rule learning, basic concepts
- autoencoder
- automatic colorization
- automatic differentiation / Theano
B
- backpropagated error
- backpropagation algorithm
- backpropagation formula
- Backpropagation through Time (BPTT) / Recurrent neural networks
- bag-of-word (BoW)
- basic modelling
- basic naive Bayes classifier baseline
- Bayesian Information Criteria (BIC) / How many clusters?
- BBC
- BBC dataset
- benchmark tests
- Bernoulli RBM
- bias
- big data
- big data application
- BigML / Machine learning as a service
- bigram
- binary classes
- Boltzmann Machines (BMs)
- Book-Crossing dataset
- book-recommendation engine
- Bot Store
- Brazilian Institute of Meteorology (INMET)
- breadth-first search (BFS)
- breakdown-oriented approach, deep learning
C
- Caffe
- Canova library
- Cassandra
- categorical data
- cc.mallet.pipe package
- Chainer
- Chart class / Visualizing the SOMs
- Chebyshev distance
- classification
- classification algorithms
- classification problems
- classifier
- class implementation
- class unbalance / Class unbalance
- clustering
- clustering algorithms
- clustering tasks
- collaborative filtering
- comma-separated values (CSV) / Loading/selecting data
- Comma Separated Value (CSV)
- competitions
- competitive layer
- competitive learning
- computational differentiation / Theano
- confusion matrix
- conjugate gradient optimization algorithm
- constant error carousel (CEC) / Long short term memory networks
- content-based filtering
- Contrastive Divergence (CD)
- Contrastive Divergence algorithm
- convolutional neural network (CNN)
- Convolutional Neural Network (CNN)
- Convolutional neural networks (CNN)
- convolution layers / Convolution
- Core Motion framework, iOS
- correlation coefficient
- cosine distance
- cost function
- Coursera
- cross-industry applications, of affinity analysis
- cross-validation
- Cross Industry Standard Process for Data Mining (CRISP-DM)
- CrowdANALYTIX
- CSVLoader class
- curse of dimensionality
- customer profiling
- customer relationship database
- Cyc
D
- 2D competitive layer
- data
- Data and problem definition
- data and problem definition
- data cleaning
- data collection
- data collector
- data correlation
- data filtering
- Data Mining
- Data Mining Research
- data pre-processing
- data reduction
- data science
- Data Science Central
- Data Science CS109 (Harvard) by John A. Paulson
- data scientist
- DataSet class / Building auxiliary classes
- dataset rebalancing
- datasets
- data transformation
- Davies-Boudin index / Cluster evaluation and validation
- Decision and Predictive Analytics (ADAPA) / Predictive Model Markup Language
- decision boundary
- decision trees
- decision trees learning
- Decode
- deep architectures
- deep belief nets (DBN)
- Deep Belief Nets (DBNs)
- deep belief network
- deep belief network (DBN)
- deep belief networks
- Deep Belief Networks (DBNs)
- deep convolutional networks
- Deep Dream
- deep learning
- Deeplearning4j
- deeplearning4java
- deep learning algorithm
- deep learning algorithms
- deep learning group
- Deep Learning News
- DeepMind
- defined classes
- delta rule
- Denoising Autoencoders (DA)
- denormalize
- depth-first search (DFS)
- digit representation
- digits recognition
- dimensionality reduction
- directory
- Discrete Fourier Transform (DFT)
- disease diagnosis
- distance measures
- divide-and-conquer strategy
- DL4J
- double evaluateLeftToRight method
- DrivenData / Competitions
- DropConnect neural network
- dropout
- dropout algorithm
- DSGuide
- Dunn index / Cluster evaluation and validation
- dynamic time wrapping (DTW)
E
- Eclipse
- Eclipse IDE
- Edit distance
- elbow method
- email spam dataset
- email spam detection
- empirical design, neural networks
- encapsulation
- Encode
- energy efficiency dataset
- enrolment status prediction
- ensambleSel.setOptions () method
- ensemble learning
- ensembleLibrary package
- ensembles
- Ensemble Selection algorithm
- environmental sensors
- Euclidean distances
- evaluate() method, parameters
- evaluation
- Expectation Maximization (EM) clustering
- exploitation
- exploration
- external validation / External validation
- extreme learning machines (ELMs)
F
- Feature extraction
- feature map
- feature maps / Convolution
- feature selection
- feed-forward neural networks / Feed-forward neural networks for NLP
- feedback networks
- feedforward networks
- feedforward neural networks
- field-oriented approach, deep learning
- file
- fine-tuning
- forget gate / Long short term memory networks
- Fourier transform
- FP-Growth
- FP-growth algorithm
- FP-tree structure
- frame problem
- fraud detection, of insurance claims
- frequent pattern (FP)
G
- Geeking with Greg
- generalization
- Generalized Sequential Patterns (GSP)
- Generative Stochastic Networks (GSNs)
- Gibbs sampling
- GitHub
- GitXiv
- GNU General Public License (GNU GPL)
- Google Prediction API / Machine learning as a service
- gradient method
- Graphical User Interface (GUI) / Results and simulations
- Graphics Processing Unit (GPU)
- GraphX
H
- Hacker News
- Hadoop
- Hadoop Distributed File System (HDFS)
- Hamming distance
- HBase
- Hebbian learning
- Hidden layer
- Hidden layer, issues
- hidden layers
- Hidden Markov Model (HMM) / Hidden Markov Model (HMM)
- hidden Markov models (HMM)
- Hidden Markov Models (HMMs)
- hidden units
- hierarchical clustering
- histogram-based anomaly detection
- Hopfield network
- Hotspot
- hybrid approach
- hybrid neural network
- hybrid systems
I
- IBM Research team
- IBM Watson Analytics / Machine learning as a service
- image classification
- ImageNet
- Imagenet Large Scale Visual Recognition Challenge (ILSVRC)
- image recognition
- Inceptionism
- Infrastructure as a Service (IaaS) / Machine learning in the cloud
- inheritance
- input gate / Long short term memory networks
- Input layer
- input method editor (IME)
- input selection
- insurance claims
- interval data
- Intrusion Detection (ID)
- item-based analysis
- item-based collaborative filtering
J
- Jaccard distance
- Java
- Java-ML packages
- java -Xmx16g
- Java API packages, Weka
- weka.associations / Weka
- weka.classifiers / Weka
- weka.clusterers / Weka
- weka.core / Weka
- weka.datagenerators / Weka
- weka.estimators / Weka
- weka.experiment / Weka
- weka.filters / Weka
- weka.gui / Weka
- Java machine learning (Java-ML)
- JFreeChart package
K
- K-fold cross-validation
- k-means clustering
- k-nearest neighbors
- Kaggle / Competitions
- KDD Cup
- KDnuggets / Machine learning as a service
- kernel methods
- kernels / Convolution
- knowledge base
- Knowledge Representation (KR)
- known-knowns
- known-unknowns
- Kohonen neural network
- Kohonen self-organizing maps (SOMs)
L
- Latent Dirichlet
- Latent Dirichlet Allocation
- Latent Dirichlet Allocation (LDA)
- layer-wise training
- layers
- learning ability
- learning algorithms
- learning paradigms
- learning process
- learning rate
- leave-one-out validation
- Levenberg-Marquardt algorithm
- library/framework
- Linear Discriminant Analysis (LDA)
- linear regression
- linear separation
- Local Outlier Factor (LOF)
- LOF algorithm
- logistic regression
- long short term memory (LSTM) network / Long short term memory networks
- long short time memory (LSTM)
- LSTM block / Long short term memory networks
- LSTM memory block / Long short term memory networks
M
- machine
- machine learning
- machine learning application
- Machine Learning for Language Toolkit (MALLET)
- machine learning libraries
- Machine learning mastery
- Mahalanobis distance
- Mahout interfaces, abstractions
- Mahout libraries
- Mallet
- MALLET, packages
- Manhattan distance
- market basket analysis (MBA)
- Markov chain
- Markov model / Feed-forward neural networks for NLP
- Markov process
- matrix algebra
- Maven plugin
- maximizing the margin
- maximum likelihood estimation (MLE) / Feed-forward neural networks for NLP
- mean absolute error
- mean squared error
- mean squared error (MSE) / Calculating the cost function
- measurement scales
- Microsoft Azure Machine Learning / Machine learning as a service
- Microsoft Excel® / Building time series
- mini-batch
- mini-batch stochastic gradient descent (MSGD)
- Minkowski distance
- missing values
- MLlib API library
- MLP implementation
- MNIST classifications
- MNIST database
- MNIST dataset
- mobile app
- mobile phone
- mobile phone sensors
- model
- models
- momentum
- momentum coefficient / Learning rate optimization
- MongoDB
- monolayer networks
- motion sensors
- Mozilla Thunderbird
- multi-class logistic regression
- multi-layer neural networks (MLP)
- multi-layer perceptrons
- Multilayer Convolutional Network
- multilayer networks
- Multilayer Perceptrons (MLP) / Hybrid systems
- multiple classes
- myrunscollector package
N
- N-gram
- Naive Bayes
- naive Bayes baseline
- NaNs / Dropping NaNs
- natural language processing (NLP)
- ND4J
- neighborhood function
- Nervana
- Nesterov's Accelerated Gradient Descent / Learning rate optimization
- neural-storyteller
- NeuralLayer class
- neural network
- neural network class
- Neural Network Language Model (NLMM)
- neural networks
- neural networks, classes
- neuro-fuzzy
- neuro-genetic
- neuron class
- No Free Lunch Theorem (NFLT) / Image recognition
- nominal data
- non-Euclidean distance
- normalization
O
P
- p-norm distance
- paper, deep belief nets (DBN)
- PAPI
- part-of-speech (POS)
- pattern analysis
- pattern recognition
- Pearson coefficient
- Pearson correlation coefficient
- peephole connections / Long short term memory networks
- perceptron
- perceptron algorithm / Perceptrons (single-layer neural networks)
- perceptrons
- plan recognition
- polymorphism
- pooling layers / Pooling
- Portable Format for Analytics (PFA) / Predictive Model Markup Language
- position sensors
- Pre-processing phase
- pre-training
- precision
- Prediction.IO / Machine learning as a service
- predictive apriori
- Predictive Model Markup Language (PMML)
- pretraining
- Principal component analysis (PCA)
- Principal Component Analysis (PCA)
- Principal Components Analysis (PCA)
- probabilistic classifiers
- probabilistic statistical model
- proben1 dataset
- profiling
- protocol file
- pseudo-algorithm
- Pylearn2
R
S
- Scale Invariant Feature Transform (SIFT)
- score function
- scratch implementations
- Self-Organizing Maps (SOM) / Hybrid systems
- signe
- signifiant
- signifié
- similar items
- similarity calculation
- SimRank
- single layer regression model
- Singular value decomposition (SVD)
- Skymind
- softplus function / Dropout
- SOM learning algorithm
- Spark Streaming
- spatio-temporal patterns
- Spearman's footrule distance
- stacked autoencoders
- Stacked Denoising Autoencoders (SDA)
- standards and markup languages
- Statistics 110 (Harvard) by Joe Biltzstein
- stochastic gradient descent (SGD)
- stochastic online learning
- stratification
- strong AI
- structure selection
- sum transfer function
- supermarket dataset
- supervised learning
- Support Vector Machine (SVM) / Support Vector Machine (SVM)
- Support Vector Machine (SVM) model / Predictive Model Markup Language
- Support Vector Machines (SVM)
- support vectors
- survivorship bias
- suspicious behaviour detection
- suspicious pattern detection
- suspicious patterns, modelling
- SVM
- symbol content
- symbol grounding problem
- symbol representation
- Sample, Explore, Modify, Model, and Assess (SEMMA).
T
- target variables
- Tay
- Technical Singularity
- TensorFlow
- Tertius
- testing
- test set
- text classification
- text data
- text mining
- text recognition
- Theano
- time series data
- topic modeling
- topic modelling, for BBC news
- traditional machine learning
- training
- Training data
- training data
- train set
- transaction analysis
- TreeModel / Predictive Model Markup Language
- trigram
- truncated BPTT / Recurrent neural networks
- two-dimensional SOM
U
- UCI machine learning repository
- Udemy
- undefined classes
- underfits
- underfitting
- unigram
- Universal PMML Plug-in (UPPI) / Predictive Model Markup Language
- unknown-unknowns
- unsupervised learning
- unsupervised learning algorithms
- user-based analysis
- user-based collaborative filtering
V
- vanilla approach
- vanishing gradient problem
- visible layer
- visible units
W
- Waikato Environment for Knowledge Analysis (Weka)
- weak AI
- weather database
- weather forecasting
- data, loading / Loading/selecting data, Loading the data and beginning to play!
- data, selecting / Loading/selecting data
- output variable, selecting / Choosing input and output variables
- input variable, selecting / Choosing input and output variables
- preprocessing / Preprocessing
- normalization, implementing / Normalization
- normalization, handling with NeuralDataSet / Adapting NeuralDataSet to handle normalization
- learning algorithm, adapting for normalization / Adapting the learning algorithm to normalization
- Java implementation / Java implementation of weather forecasting
- weather data, collecting / Collecting weather data
- variables, delaying / Delaying variables
- executing / Loading the data and beginning to play!
- correlation analysis, performing / Let's perform a correlation analysis
- neural networks, creating / Creating neural networks
- training / Training and test
- testing / Training and test
- neural network, training / Training the neural network
- error, plotting / Plotting the error
- neural network output, viewing / Viewing the neural network output
- weather forecasting, data selection
- web resources and competitions
- website traffic
- weka.classifiers package
- weka.classifiers.bayes / Weka
- weka.classifiers.evaluation / Weka
- weka.classifiers.functions / Weka
- weka.classifiers.lazy / Weka
- weka.classifiers.meta / Weka
- weka.classifiers.mi / Weka
- weka.classifiers.rules / Weka
- weka.classifiers.trees / Weka
- Weka 3.6
- WEKA Packages
- winner-takes-all rule / Competitive learning
- word2vec
- workflow, Applied Machine Learning
X
Y
Z
..................Content has been hidden....................
You can't read the all page of ebook, please click
here login for view all page.