book

Applied Machine Learning and AI for Engineers

by Jeff Prosise

November 2022

Intermediate to advanced

425 pages

11h 25m

English

O'Reilly Media, Inc.

Book available

Read now

Unlock full access

Who Should Read This BookWhy I Wrote This BookRunning the Book’s Code SamplesNavigating This BookConventions Used in This BookUsing Code ExamplesO’Reilly Online LearningHow to Contact UsAcknowledgments
What Is Machine Learning?Machine Learning Versus Artificial IntelligenceSupervised Versus Unsupervised LearningUnsupervised Learning with k-Means ClusteringApplying k-Means Clustering to Customer DataSegmenting Customers Using More Than Two DimensionsSupervised Learningk-Nearest NeighborsUsing k-Nearest Neighbors to Classify FlowersSummary
Linear RegressionDecision TreesRandom ForestsGradient-Boosting MachinesSupport Vector MachinesAccuracy Measures for Regression ModelsUsing Regression to Predict Taxi FaresSummary
Logistic RegressionAccuracy Measures for Classification ModelsCategorical DataBinary ClassificationClassifying Passengers Who Sailed on the TitanicDetecting Credit Card FraudMulticlass ClassificationBuilding a Digit Recognition ModelSummary
Preparing Text for ClassificationSentiment AnalysisNaive BayesSpam FilteringRecommender SystemsCosine SimilarityBuilding a Movie Recommendation SystemSummary
How Support Vector Machines WorkKernelsKernel TricksHyperparameter TuningData NormalizationPipeliningUsing SVMs for Facial RecognitionSummary
Understanding Principal Component AnalysisFiltering NoiseAnonymizing DataVisualizing High-Dimensional DataAnomaly DetectionUsing PCA to Detect Credit Card FraudUsing PCA to Predict Bearing FailureMultivariate Anomaly DetectionSummary
Consuming a Python Model from a Python ClientVersioning Pickle FilesConsuming a Python Model from a C# ClientContainerizing a Machine Learning ModelUsing ONNX to Bridge the Language GapBuilding ML Models in C# with ML.NETSentiment Analysis with ML.NETSaving and Loading ML.NET ModelsAdding Machine Learning Capabilities to ExcelSummary

Understanding Neural NetworksTraining Neural NetworksSummary
Building Neural Networks with Keras and TensorFlowSizing a Neural NetworkUsing a Neural Network to Predict Taxi FaresBinary Classification with Neural NetworksMaking PredictionsTraining a Neural Network to Detect Credit Card FraudMulticlass Classification with Neural NetworksTraining a Neural Network to Recognize FacesDropoutSaving and Loading ModelsKeras CallbacksSummary
Understanding CNNsUsing Keras and TensorFlow to Build CNNsTraining a CNN to Recognize Arctic WildlifePretrained CNNsUsing ResNet50V2 to Classify ImagesTransfer LearningUsing Transfer Learning to Identify Arctic WildlifeData AugmentationImage Augmentation with ImageDataGeneratorImage Augmentation with Augmentation LayersApplying Image Augmentation to Arctic WildlifeGlobal PoolingAudio Classification with CNNsSummary
Face DetectionFace Detection with Viola-JonesUsing the OpenCV Implementation of Viola-JonesFace Detection with Convolutional Neural NetworksExtracting Faces from PhotosFacial RecognitionApplying Transfer Learning to Facial RecognitionBoosting Transfer Learning with Task-Specific WeightsArcFacePutting It All Together: Detecting and Recognizing Faces in PhotosHandling Unknown Faces: Closed-Set Versus Open-Set ClassificationSummary
R-CNNsMask R-CNNYOLOYOLOv3 and KerasCustom Object DetectionTraining a Custom Object Detection Model with the Custom Vision ServiceUsing the Exported ModelSummary
Text PreparationWord EmbeddingsText ClassificationAutomating Text VectorizationUsing TextVectorization in a Sentiment Analysis ModelFactoring Word Order into PredictionsRecurrent Neural Networks (RNNs)Using Pretrained Models to Classify TextNeural Machine TranslationLSTM Encoder-DecodersTransformer Encoder-DecodersBuilding a Transformer-Based NMT ModelUsing Pretrained Models to Translate TextBidirectional Encoder Representations from Transformers (BERT)Building a BERT-Based Question Answering SystemFine-Tuning BERT to Perform Sentiment AnalysisSummary
Introducing Azure Cognitive ServicesKeys and EndpointsCalling Azure Cognitive Services APIsAzure Cognitive Services ContainersThe Computer Vision ServiceThe Language ServiceThe Translator ServiceThe Speech ServicePutting It All Together: Contoso TravelSummary

Content preview from Applied Machine Learning and AI for Engineers

Chapter 13. Natural Language Processing

It’s not difficult to use Scikit-Learn to build machine learning models that analyze text for sentiment, identify spam, and classify text in other ways. But today, state-of-the-art text classification is most often performed with neural networks. You already know how to build neural networks that accept numbers and images as input. Let’s build on that to learn how to construct deep-learning models that process text—a segment of deep learning known as natural language processing, or NLP for short.

NLP encompasses a variety of activities including text classification, named-entity recognition, keyword extraction, question answering, and language translation. The accuracy of NLP models has improved in recent years for a variety of reasons, not the least of which are newer and better ways of converting words and sentences into dense vector representations that incorporate meaning, and a relatively new neural network architecture called the transformer that can zero in on the most meaningful words and even differentiate between different meanings of the same word.

One element that virtually all neural networks that process text have in common is an embedding layer, which uses word embeddings to transform arrays, or sequences, of scalar values representing words into arrays of floating-point numbers called word vectors. These vectors encode information about the meanings of words and the relationships between them. Output from an embedding layer ...