Index
A
- activation function / Artificial neuron
- actor-network / The actor-critic algorithm
- advanced transportation controller (ATC) / Smart traffic management
- AI-powered Industrial IoT
- about / Introduction to AI-powered industrial IoT
- use cases / Some interesting use cases
- AI-powered Smart Surveillance Solution / Home surveillance
- AI-powered startups
- Uptake Technologies Inc / Introduction to AI-powered industrial IoT
- C3.ai / Introduction to AI-powered industrial IoT
- Alluvium / Introduction to AI-powered industrial IoT
- Arundo Analytics / Introduction to AI-powered industrial IoT
- Canvass Analytics / Introduction to AI-powered industrial IoT
- AI controlled sailplanes / Some successful applications
- AI platforms / AI platforms and IoT platforms
- Air quality dataset / Air quality data
- Alexa / Digital assistants
- Alluvium
- reference / Introduction to AI-powered industrial IoT
- AlphaGo Zero / Some successful applications
- Amazon AWS IoT / AI platforms and IoT platforms
- Amazon Web Service (AWS) / AWS
- Apache MLlib
- about / Apache MLlib
- algorithms / Apache MLlib
- Apple watch
- reference / HAR using wearable sensors
- Array of Things (AoT) / Chicago Array of Things data
- Artificial Intelligence (AI) / Deep learning 101, Continuous glucose monitoring
- artificial neuron / Artificial neuron
- Arundo Analytics
- reference / Introduction to AI-powered industrial IoT
- asset management / Some interesting use cases
- asset tracking / Some interesting use cases
- audio file
- as input data / Audio files as input data
- autoencoders
- about / Autoencoders
- denoising autoencoders / Denoising autoencoders
- variational autoencoders (VAE) / Variational autoencoders
B
- backpropagation algorithm / The backpropagation algorithm
- backpropagation through time (BPTT) / Recurrent neural networks
- bagging / Bagging and pasting
- batch normalization / Some popular CNN model
- Bellman Equation / Q-learning
- Blender learning environment / Simulated environments
C
- C3.ai
- reference / Introduction to AI-powered industrial IoT
- Canvass Analytics
- reference / Introduction to AI-powered industrial IoT
- cardiac arrhythmia / Heart monitor
- catastrophic forgetting / Taxi drop-off using Q-Network
- Cauchy method / Gradient descent method
- CGM data
- using, in hypoglycemia prediction / Hypoglycemia prediction using CGM data
- channels / The convolution layer
- cities, with open data
- about / Cities with open data
- Atlanta city Metropolitan Atlanta Rapid Transit Authority data / Atlanta city Metropolitan Atlanta Rapid Transit Authority data
- Chicago Array of Things data / Chicago Array of Things data
- CitySense
- reference / Smart lighting
- classification
- with support vector machines / Classification using support vector machines
- in H2O / Classification in H20
- classification, MLlib / Classification in MLlib
- cloud platform providers
- about / Computing in the cloud
- Amazon Web Service (AWS) / AWS
- Google Cloud platform / Google Cloud Platform
- Microsoft Azure / Microsoft Azure
- cloud platforms
- IBM Watson IoT Platform / AI platforms and IoT platforms
- Microsoft IoT-Azure IoT suite / AI platforms and IoT platforms
- Google Cloud IoT / AI platforms and IoT platforms
- Amazon AWS IoT / AI platforms and IoT platforms
- CNN models
- LeNet / Some popular CNN model
- VGGNet / Some popular CNN model
- ResNet / Some popular CNN model
- GoogleNet / Some popular CNN model
- combined cycle power plant (CCPP) dataset
- about / The combined cycle power plant dataset
- features / The combined cycle power plant dataset
- Comma-separated value (CSV) / CSV format
- complex event processing (CEP) / Big data and IoT
- components, smart city
- smart traffic management / Smart traffic management
- smart parking / Smart parking
- smart waste management / Smart waste management
- smart policing / Smart policing
- smart lighting / Smart lighting
- smart governance / Smart governance
- components, Spark
- Resilient Distributed Datasets (RDDs) / Spark components
- distributed variables / Spark components
- DataFrames / Spark components
- libraries / Spark components
- components, TensorFlow
- computation graph / TensorFlow
- execution graph / TensorFlow
- continuous glucose monitoring (CGM) / Continuous glucose monitoring
- convex cost function
- example / Optimization
- Convolutional neural networks (CNN)
- about / Convolutional neural networks
- layers / Different layers of CNN
- convolution layer / The convolution layer
- pooling layer / Pooling layer
- convolution layers
- parameters / The convolution layer
- Cortana / Digital assistants
- critic-network / The actor-critic algorithm
- cross-entropy loss function / Cross-entropy loss function
- cross-industry standard process for data mining (CRISP-DM) / Cross-industry standard process for data mining
- cross-validation / Cross-validation
- crossover operation
- performing / Crossover
- CSV files
- using, with csv module / Working with CSV files with the csv module
- using, with pandas module / Working with CSV files with the pandas module
- using, with NumPy module / Working with CSV files with the NumPy module
- CycleGAN
D
- data
- processing / Processing different types of data
- data augmentation
- for images / Data augmentation for images
- Database Management System (DBMS) / SQL data
- data management (DM) / Cross-industry standard process for data mining
- datasets
- combined cycle power plant (CCPP) / The combined cycle power plant dataset
- Wine quality dataset / Wine quality dataset
- Air quality data / Air quality data
- about / HDF5 format
- dead neurons / Modelling single neuron in TensorFlow
- decision trees
- about / Decision trees
- in scikit / Decision trees in scikit
- working / Decision trees in action
- Deep Convolutional GAN (DCGAN) / Deep Convolutional GANs
- deep learning
- about / Deep learning 101, Deep learning—why now?
- artificial neuron / Artificial neuron
- Deep Q-Network (DQN) / Q-Network
- deep reinforcement learning / Deep reinforcement learning
- DeepSight AILabs
- reference / Home surveillance
- denoising autoencoders / Denoising autoencoders
- deterministic and analytic methods, optimization
- gradient descent method / Gradient descent method
- Newton-Raphson method / Newton-Raphson method
- digital assistants
- about / Digital assistants
- Siri / Digital assistants
- Cortana / Digital assistants
- Alexa / Digital assistants
- Google Assistant / Digital assistants
- directed acyclic graph (DAG) / Genetic algorithm for CNN architecture
- discriminative network / GANs
- Distributed Evolutionary Algorithms in Python (DEAP)
- about / Coding genetic algorithms using Distributed Evolutionary Algorithms in Python
- used, for coding genetic algorithms / Coding genetic algorithms using Distributed Evolutionary Algorithms in Python
- reference / Coding genetic algorithms using Distributed Evolutionary Algorithms in Python
- Double DQN / Double DQN
- DQN
- Atari game, playing / DQN to play an Atari game
- Duelling DQN / Dueling DQN
E
- electrical load forecasting
- in industry / Electrical load forecasting in industry
- ensemble learning
- about / Ensemble learning
- voting classifier / Voting classifier
- bagging / Bagging and pasting
- pasting / Bagging and pasting
- epsilon greedy algorithm / Q-learning
- experience replay / Taxi drop-off using Q-Network
- explicit generative models / Introduction
- exploration-exploitation trade-off / Q-learning
F
- feature scaling / Feature scaling to resolve uneven data scale
- feedforward network / Multilayered perceptrons for regression and classification
- Fitbit
- reference / HAR using wearable sensors
- fitness function / Optimization, The genetic algorithm
- fixed Q-targets / Taxi drop-off using Q-Network
- fleet management and maintenance / Some interesting use cases
- folds / Cross-validation
- Frontiers of Information Technology (FIT) / IoT verticals
- function call / HDFS
G
- GAN Zoo GitHub
- reference / Variants of GAN and its cool applications
- Gated recurrent unit (GRU) / Gated recurrent unit
- Gaussian kernel / Kernel trick
- Gaussian Naive Bayes
- for wine quality / Gaussian Naive Bayes for wine quality
- Gazebo / Simulated environments
- General Transit Feed Specification (GTFS) / Atlanta city Metropolitan Atlanta Rapid Transit Authority data
- Generative Adversarial Networks (GANs)
- about / GANs
- architecture / GANs
- learning steps / GANs
- applications / Applications of GANs
- generative models
- about / Introduction
- explicit generative models / Introduction
- implicit generative models / Introduction
- generative network / GANs
- genetic algorithms
- about / Genetic algorithms, Introduction to genetic algorithms
- implementing / The genetic algorithm
- crossover operation / Crossover
- mutation operator / Mutation
- advantages / Advantages
- disadvantages / Disadvantages
- coding, Distributed Evolutionary Algorithms in Python (DEAP) used / Coding genetic algorithms using Distributed Evolutionary Algorithms in Python
- used, for guessing word / Guess the word
- for CNN architecture / Genetic algorithm for CNN architecture
- for LSTM optimization / Genetic algorithm for LSTM optimization
- Google Assistant / Digital assistants
- Google Cloud IoT / AI platforms and IoT platforms
- Google Cloud platform
- reference / Google Cloud Platform
- Google Colaboratory / Deep learning—why now?
- Google File System
- reference / HDFS
- GoogleNet / Some popular CNN model
- gradient descent algorithm / Modelling single neuron in TensorFlow
- gradient descent method
- about / Gradient descent method
- drawbacks / Gradient descent method
- types / Gradient descent method
- graphical processing units (GPUs) / Deep learning—why now?
- grid search / Hyperparameter tuning and grid search
- group / HDF5 format
H
- H2O
- about / Introducing H2O.ai
- reference / H2O AutoML
- regression / Regression in H2O
- classification / Classification in H20
- H2O.ai / Introducing H2O.ai
- H2O AutoML
- about / H2O AutoML
- reference / H2O AutoML
- h5py
- about / HDF5 format, Using HDF5 with h5py
- HDF5, using with / Using HDF5 with h5py
- reference / Using HDF5 with h5py
- Hadoop Distributed File System (HDFS)
- about / HDFS
- NodeName / HDFS
- DataNode / HDFS
- hdfs3, using with / Using hdfs3 with HDFS
- PyArrow's filesystem interface, using for / Using PyArrow's filesystem interface for HDFS
- Hamilton Watch Company / Personal IoT
- HDF5
- reference / HDF5 format
- about / HDF5 format
- using, with PyTables / Using HDF5 with PyTables
- using, with pandas / Using HDF5 with pandas
- using, with h5py / Using HDF5 with h5py
- HDF5, in pandas
- reference / Using HDF5 with pandas
- hdfs3
- using, with HDFS / Using hdfs3 with HDFS
- reference / Using hdfs3 with HDFS
- Hierarchical Data Format (HDF) / HDF5 format
- home surveillance / Home surveillance
- human activity recognition (HAR)
- about / Human activity recognition
- with wearable sensors / HAR using wearable sensors
- from videos / HAR from videos
- hyperparameter tuning / Hyperparameter tuning and grid search
- hypoglycemia prediction
- with CGM data / Hypoglycemia prediction using CGM data
I
- IBM Watson IoT Platform / AI platforms and IoT platforms
- If This Then That (IFTTT) / Smart lighting
- images
- generating, VAEs used / Generating images using VAEs
- implicit generative models / Introduction
- independent and identically distributed (iid) / Naive Bayes
- Industry 4.0 / Predictive maintenance using AI
- Industry IoT (IIoT) / What is IoT 101?
- International Telecommunication Unit (ITU)
- reference / What is IoT 101?
- internet of everything / What is IoT 101?
- Internet of Things (IoT)
- reference / What is IoT 101?
- scenarios / What is IoT 101?
- big data / Big data and IoT
- data science / Infusion of AI – data science in IoT
- adopting, for smart city / Adapting IoT for smart cities and the necessary steps
- IoT, and smart homes
- about / IoT and smart homes
- human activity recognition (HAR) / Human activity recognition
- smart lighting / Smart lighting
- home surveillance / Home surveillance
- IoT 101 / What is IoT 101?
- IoT layers
- device layer / IoT reference model
- network layer / IoT reference model
- service layer / IoT reference model
- application layer / IoT reference model
- IoT platforms
- about / IoT platforms, AI platforms and IoT platforms
- selection criteria / IoT platforms
- IoT reference model / IoT reference model
- IoT verticals
- about / IoT verticals
- smart building / IoT verticals
- smart agriculture / IoT verticals
- smart city / IoT verticals
- connected healthcare / IoT verticals
J
- JavaScript Object Notation (JSON) / Working with the JSON format
- JSON files
- using, with JSON module / Using JSON files with the JSON module
- using, with pandas module / JSON files with the pandas module
- JSON format
- working with / Working with the JSON format
K
- Kaggle / Deep learning—why now?
- Keras / Keras
- kernel / Kernel trick
L
- Lasso regularization / Regularization
- LeNet
- reference / Some popular CNN model
- handwritten digits, recognizing / LeNet to recognize handwritten digits
- linear regression
- using, for prediction / Prediction using linear regression
- about / Prediction using linear regression
- electrical power output prediction / Electrical power output prediction using regression
- Locomotion behavior / Some successful applications
- logistic regression
- about / Logistic regression for classification
- for classification / Logistic regression for classification
- cross-entropy loss function / Cross-entropy loss function
- logistic regressor
- wine quality data, classifying / Classifying wine using logistic regressor
- logit function / Logistic regression for classification
- long-term load forecasting / Electrical load forecasting in industry
- Long Short-Term Memory (LSTM) / Predictive maintenance using Long Short-Term Memory
- long short-term memory (LSTM) / Long short-term memory
- loss function / Optimization
M
- machine-to-human (M2H) / Predictive maintenance using AI
- machine-to-machine (M2M) / Predictive maintenance using AI
- machine learning (ML)
- about / ML and IoT, Introduction to AI-powered industrial IoT
- IoT / ML and IoT
- paradigms / Learning paradigms
- Markov Decision Process (MDP) / RL terminology
- maximal margin separator / Classification using support vector machines
- medium-term load forecasting / Electrical load forecasting in industry
- Metropolitan Atlanta Rapid Transit Authority (MARTA) / Atlanta city Metropolitan Atlanta Rapid Transit Authority data
- Microsoft Azure / Microsoft Azure
- Microsoft Cloud services
- reference / Microsoft Azure
- Microsoft IoT-Azure IoT suite / AI platforms and IoT platforms
- min-max normalization / Feature scaling to resolve uneven data scale
- MLlib
- regression / Regression in MLlib
- classification / Classification in MLlib
- Multi-Point Crossover / Crossover
- multilayered perceptrons
- for regression / Multilayered perceptrons for regression and classification
- for classification / Multilayered perceptrons for regression and classification
- multiple linear regression / Prediction using linear regression
- mutation operator / Mutation
- MySQL database engine / The MySQL database engine
N
- Naive Bayes / Naive Bayes
- natural optimization methods
- about / Natural optimization methods
- simulated annealing / Simulated annealing
- Particle Swarm Optimization (PSO) / Particle Swarm Optimization
- natural optimization methodsatural optimization methods
- genetic algorithms / Genetic algorithms
- Newton-Raphson method / Newton-Raphson method
- nodes / Chicago Array of Things data
- No Free Lunch theory / No Free Lunch theorem
- normalization
- Z-score normalization / Feature scaling to resolve uneven data scale
- min-max normalization / Feature scaling to resolve uneven data scale
- Not Only Structured Query Language (NoSQL) database / NoSQL data
- NumPy module
- CSV files, using with / Working with CSV files with the NumPy module
O
- objective function / Optimization
- observations / Chicago Array of Things data
- one-point crossover / Crossover
- OpenAI gym
- about / Simulated environments, OpenAI gym
- supported environments / OpenAI gym
- OpenPyXl
- using, for XLSX files / Using OpenPyXl for XLSX files
- optimization
- about / Optimization
- tasks / Optimization
- deterministic and analytic methods / Deterministic and analytic methods
- natural optimization methods / Natural optimization methods
- optimizers, TensorFlow
- reference / Gradient descent method
- overfitting / Overfitting
- overfitting, solutions
- regularization / Regularization
- cross-validation / Cross-validation
P
- pandas
- using, with XLSX files / Using pandas with XLSX files
- HDF5, using with / Using HDF5 with pandas
- pandas module
- CSV files, using with / Working with CSV files with the pandas module
- JSON files, using with / JSON files with the pandas module
- parameters, single artificial neuron
- learning rate parameter / Modelling single neuron in TensorFlow
- activation function / Modelling single neuron in TensorFlow
- loss function / Modelling single neuron in TensorFlow
- Particle Swarm Optimization (PSO) / Particle Swarm Optimization
- pasting / Bagging and pasting
- perception layer / IoT reference model
- personal IoT
- about / Personal IoT
- SuperShoes by MIT / SuperShoes by MIT
- continuous glucose monitoring (CGM) / Continuous glucose monitoring
- heart monitor / Heart monitor
- digital assistants / Digital assistants
- policy gradients
- about / Policy gradients
- need for / Why policy gradients?
- used, for playing game of Pong / Pong using policy gradients
- actor-critic algorithm / The actor-critic algorithm
- prediction
- with linear regression / Prediction using linear regression
- predictive maintenance
- about / Some interesting use cases
- with AI / Predictive maintenance using AI
- with Long Short-Term Memory / Predictive maintenance using Long Short-Term Memory
- advantages / Predictive maintenance advantages and disadvantages
- disadvantages / Predictive maintenance advantages and disadvantages
- predictive maintenance template, Azure AI Gallery
- preventive maintenance / Introduction to AI-powered industrial IoT
- protocols, IoT
- reference / IoT verticals
- PyArrow's filesystem interface
- using, for HDFS / Using PyArrow's filesystem interface for HDFS
- PyTables
- about / HDF5 format
- HDF5, using with / Using HDF5 with PyTables
- Python
- TXT files, using / Using TXT files in Python
Q
- Q-learning / Q-learning
- Q-Network
- about / Q-Network
- taxi drop-off / Taxi drop-off using Q-Network
- Q-tables
- taxi drop-off / Taxi drop-off using Q-tables
- quasi-Newton methods / Newton-Raphson method
R
- radial basis function / Kernel trick
- random forest / Bagging and pasting
- random resetting / Mutation
- rectified linear units (ReLU) / Modelling single neuron in TensorFlow
- recurrent neural networks / Recurrent neural networks
- about / Recurrent neural networks
- long short-term memory (LSTM) / Long short-term memory
- Gated recurrent unit (GRU) / Gated recurrent unit
- regression
- in H2O / Regression in H2O
- regression, MLlib
- about / Regression in MLlib
- linear regression / Regression in MLlib
- generalized linear regression / Regression in MLlib
- decision tree regression / Regression in MLlib
- random forest regression / Regression in MLlib
- gradient boosted tree regression / Regression in MLlib
- regularization / Regularization
- reinforcement learning (RL)
- terminologies / RL terminology
- states / RL terminology
- actions / RL terminology
- reward / RL terminology
- policy / RL terminology
- value function / RL terminology
- environment model / RL terminology
- deep reinforcement learning / Deep reinforcement learning
- applications / Some successful applications
- relational database / SQL data
- remaining useful life (RUL) / Predictive maintenance using Long Short-Term Memory
- Remote Procedure Calls (RPCs) / HDFS
- replay buffer / Taxi drop-off using Q-Network
- residual learning / Some popular CNN model
- Resilient Distributed Datasets (RDDs) / Spark components
- ResNet / Some popular CNN model
- ridge regularization / Regularization
- RL algorithms
- value-based methods / Deep reinforcement learning
- policy-based methods / Deep reinforcement learning
- Robot Operating System (ROS) / Simulated environments
- root-mean-square error (RMSE) / Genetic algorithm for LSTM optimization
S
- San Francisco Municipal Transportation Agency (SAFTA) / Smart parking
- scikit
- decision trees / Decision trees in scikit
- seasonality / Time series modeling
- sensors / Chicago Array of Things data
- SFpark
- reference / Smart parking
- short-term load forecasting (STLF)
- about / Electrical load forecasting in industry
- with LSTM / STLF using LSTM
- sigmoid function / Logistic regression for classification
- simple linear regression / Prediction using linear regression
- simple perceptron / Multilayered perceptrons for regression and classification
- simulated annealing / Simulated annealing
- simulated environments
- OpenAI gym / Simulated environments, OpenAI gym
- Unity ML-Agents SDK / Simulated environments
- Gazebo / Simulated environments
- Blender learning environment / Simulated environments
- single artificial neuron
- parameters / Modelling single neuron in TensorFlow
- single neuron
- modelling, in TensorFlow / Modelling single neuron in TensorFlow
- Single Point Crossover / Crossover
- Siri / Digital assistants
- smart city
- need for / Why do we need smart cities?
- components / Components of a smart city
- IoT, adopting for / Adapting IoT for smart cities and the necessary steps
- crime, detecting with San Francisco crime data / Detecting crime using San Francisco crime data
- challenges / Challenges and benefits
- benefits / Challenges and benefits
- smart lighting / Smart lighting
- smart policing initiatives (SPI) / Smart policing
- SmartTrack / HAR using wearable sensors
- Snakebite
- reference / HDFS
- Spark
- components / Spark components
- working / Spark components
- reference / Spark components
- SparkDL
- using, in transfer learning / Transfer learning using SparkDL
- reference / Transfer learning using SparkDL
- Spark driver / Spark components
- Spark executors / Spark components
- Spark MLlib logistic regression classifier
- wine quality classification problem, implementing / Classification in MLlib
- SQL data / SQL data
- sqlite3 module / The SQLite database engine
- SQLite database engine
- reference / The SQLite database engine
- about / The SQLite database engine
- using / The SQLite database engine
- stationarity / Time series modeling
- stationary / Time series modeling
- Structured Query Language (SQL) / SQL data
- subroutine call / HDFS
- SuperShoes
- about / SuperShoes by MIT
- reference / SuperShoes by MIT
- supervised learning / Learning paradigms
- supported environments, OpenAI gym
- algorithms / OpenAI gym
- Atari / OpenAI gym
- Box2D / OpenAI gym
- Classic control / OpenAI gym
- MuJoCo / OpenAI gym
- robotics / OpenAI gym
- toy text / OpenAI gym
- Support Vector Machines (SVMs)
- about / Classification using support vector machines
- maximum margin hyperplane / Maximum margin hyperplane
- kernel / Kernel trick
- wine data, classifying / Classifying wine using SVM
- synaptic connections / Artificial neuron
T
- taxi drop-off
- with Q-tables / Taxi drop-off using Q-tables
- with Q-Network / Taxi drop-off using Q-Network
- Temporal Difference (TD) error / Taxi drop-off using Q-Network
- TensorFlow
- about / TensorFlow
- components / TensorFlow
- single neuron, modelling / Modelling single neuron in TensorFlow
- energy output prediction, with MLPs / Energy output prediction using MLPs in TensorFlow
- wine quality classification, with MLPs / Wine quality classification using MLPs in TensorFlow
- Variational Autoencoders (VAEs) / VAEs in TensorFlow
- vanilla GAN, implementing in / Implementing a vanilla GAN in TensorFlow
- TensorFlowOnSpark (TFoS) / Introduction
- Tensor Processing Unit (TPUs) / Deep learning—why now?
- textual data
- preprocessing / Preprocessing textual data
- time series modeling / Time series modeling
- transfer learning
- with SparkDL / Transfer learning using SparkDL
- about / Transfer learning using SparkDL
- trend / Time series modeling
- Turing test / Deep learning 101
- TXT files
- using, in Python / Using TXT files in Python
- TXT format / TXT format
U
- Unity ML-Agents SDK
- about / Simulated environments
- reference / Simulated environments
- unsupervised learning / Learning paradigms
- Uptake Technologies Inc
- reference / Introduction to AI-powered industrial IoT
- use cases, AI-powered Industrial IoT
- predictive maintenance / Some interesting use cases
- asset tracking / Some interesting use cases
- fleet management and maintenance / Some interesting use cases
V
- vanilla GAN
- implementing, in TensorFlow / Implementing a vanilla GAN in TensorFlow
- variational autoencoders (VAE) / Variational autoencoders
- Variational Autoencoders (VAEs)
- images, generating / Generating images using VAEs
- architecture / Generating images using VAEs
- in TensorFlow / VAEs in TensorFlow
- vehicle-to-infrastructure (V2I) / Smart traffic management
- vertical market / IoT verticals
- VGG16 / Some popular CNN model
- VGGNet
- reference / Some popular CNN model
- videos files
- handling / Handling videos files
- Virtual Singapore / Smart policing
- voting classifier / Voting classifier
W
- weights / Pooling layer
- Wine quality dataset / Wine quality dataset
X
- XLSX files
- OpenPyXl, using for / Using OpenPyXl for XLSX files
- pandas, using with / Using pandas with XLSX files
- XLSX format / XLSX format
Z
- Z-score normalization / Feature scaling to resolve uneven data scale