▶Book Description
Natural language processing (NLP) has found its application in various domains, such as web search, advertisements, and customer services, and with the help of deep learning, we can enhance its performances in these areas. Hands-On Natural Language Processing with Python teaches you how to leverage deep learning models for performing various NLP tasks, along with best practices in dealing with today’s NLP challenges.
To begin with, you will understand the core concepts of NLP and deep learning, such as Convolutional Neural Networks (CNNs), recurrent neural networks (RNNs), semantic embedding, Word2vec, and more. You will learn how to perform each and every task of NLP using neural networks, in which you will train and deploy neural networks in your NLP applications. You will get accustomed to using RNNs and CNNs in various application areas, such as text classification and sequence labeling, which are essential in the application of sentiment analysis, customer service chatbots, and anomaly detection. You will be equipped with practical knowledge in order to implement deep learning in your linguistic applications using Python's popular deep learning library, TensorFlow.
By the end of this book, you will be well versed in building deep learning-backed NLP applications, along with overcoming NLP challenges with best practices developed by domain experts.
▶What You Will Learn
⦁ Implement semantic embedding of words to classify and find entities
⦁ Convert words to vectors by training in order to perform arithmetic operations
⦁ Train a deep learning model to detect classification of tweets and news
⦁ Implement a question-answer model with search and RNN models
⦁ Train models for various text classification datasets using CNN
⦁ Implement WaveNet a deep generative model for producing a natural-sounding voice
⦁ Convert voice-to-text and text-to-voice
⦁ Train a model to convert speech-to-text using DeepSpeech
▶Key Features
⦁ Weave neural networks into linguistic applications across various platforms
⦁ Perform NLP tasks and train its models using NLTK and TensorFlow
⦁ Boost your NLP models with strong deep learning architectures such as CNNs and RNNs
▶Who This Book Is For
This book is intended for developers who want to leverage NLP techniques to develop intelligent applications with rich human-centric interfaces. The book assumes introductory knowledge of machine learning (ML) or deep learning and intermediate Python programming skills. Our aim is to introduce cutting-edge techniques for NLP tasks, such as sentiment detection, conversational systems, language translation, speech-to-text, and much more, using the TensorFlow framework and Python.
The reader will go from the basic concepts of deep learning to state-of-the-art algorithms and best practices for dealing with natural language. Our focus is on implementing applications using real-world data and deploying deep learning models to add human capabilities to commercial applications in a production environment.
▶What this book covers
⦁ Chapter 1, Getting Started, explores the basic concepts of NLP and the various problems it tries to solve. We also look at some of the real-world applications to give the reader the feeling of the wide range of applications that leverage NLP.
⦁ Chapter 2, Text Classification and POS Tagging Using NLTK, introduces the popular NLTK Python library. We will be using NLTK to describe basic NLP tasks, such as tokenizing, stemming, tagging, and classic text classification. We also explore POS tagging with NLTK. We provide the reader with the tools and techniques necessary to prepare data for input into deep learning models.
⦁ Chapter 3, Deep Learning and TensorFlow, introduces the basic concepts of deep learning. This chapter will also help the reader to set up the environment and tools such as TensorFlow. At the end of the chapter, the reader will get an understanding of basic deep learning concepts, such as CNN, RNN, LSTM, attention-based models, and problems in NLP.
⦁ Chapter 4, Semantic Embedding Using Shallow Models, explores how to identify semantic relationships between words in a document, and in the process, we obtain a vector representation for words in a corpus. The chapter describes developing word embedding models, such as CBOW using neural networks. It also describes techniques for developing neural network models to obtain document vectors. At the end of this chapter, the reader will get familiar with training embeddings for word, sentence, and document; and visualize simple networks.
⦁ Chapter 5, Text Classification Using LSTM, discusses various approaches for classifying text, a specific application of which is to classify sentiments of words or phrases in a document. The chapter introduces the problem of text classification. Following this, we describe techniques for developing deep learning models using CNNs and LSTMs. The chapter also explains transfer learning for text classification using pretrained word embeddings. At the end, the reader will get familiar with implementing deep learning models for sentiment classification, spam detection, and using pretrained word embeddings for his/her classification task.
⦁ Chapter 6, Searching and Deduplicating Using CNNs, covers the problems of searching, matching and deduplicating documents and approaches used in solving them. The chapter describes developing deep learning models for searching text in a corpus. At the end of this chapter, you will learn to implement a CNN-based deep learning model for searching and deduplicating text.
⦁ Chapter 7, Named Entity Recognition Using Character LSTM, describes methods and approaches to perform Named Entity Recognition (NER), a sub-task of information extraction, to locate and classify entities in text of a document. The chapter introduces the problem of NER and the applications where it can be used. We then explain the implementation of a deep learning model using character-based LSTM for identifying named entities trained using labeled datasets.
⦁ Chapter 8, Text Generation and Summarization Using GRUs, covers the methods used for the task of generating text, an extension of which can be used to create summaries from text data. We then explain the implementation of a deep learning model for generating text. This is followed by a description of implementing GRU-based deep learning models to summarize text. At the end of this chapter, the reader will learn the techniques of implementing deep learning models for text generation and summarization.
⦁ Chapter 9, Question-Answering and Chatbots Using Memory Networks, describes how to train a deep learning model to answer questions and extend it to build a chatbot. The chapter introduces the problem of question answering and the approaches used in building an answering engine using deep learning models. We then describe how to leverage a question-answering engine to build a chatbot capable of answering questions like a conversation. At the end of this chapter, you will be able to implement an interactive chatbot.
⦁ Chapter 10, Machine Translation Using Attention-Based Models, covers various methods for translating text from one language to another, without the need to learn the grammar structure of either language. The chapter introduces traditional machine translation approaches, such as Hidden Markov Model (HMM) based methods. We then explain the implementation of an encoder-decoder model with attention for translating text from French to the English language. At the end of this chapter, the reader will be able to implement deep learning models for translating text.
⦁ Chapter 11, Speech Recognition Using Deep Speech, describes the problem of converting voice to text, as a beginning of a conversational interface. The chapter begins with feature extraction from speech data. This is followed by a brief introduction of the deep speech architecture. We then explain the detailed implementation of the Deep Speech architecture to transcribe speech to text. At the end of this chapter, the reader will be equipped with the knowledge to implement a speech-to-text deep learning model.
⦁ Chapter 12, Text to Speech Using Tacotron, describes the problem of converting text to speech. The chapter describes the implementation of the Tacotron model to convert text to voice. At the end, the reader will get familiar with the implementation of a text-to-speech model based on the Tacotron architecture.
⦁ Chapter 13, Deploying Trained Models, is the concluding chapter and describes model deployments in various cloud and mobile platforms.
