توسعه برنامه‌های هوشمند IOS با سوئیفت: درک متون، طبقه‌بندی احساسات و تشخیص خودکار پاسخ‌ها در متن با استفاده از NLP

Build smart apps capable of analyzing language and performing language-specific tasks, such as script identification, tokenization, lemmatization, part-of-speech tagging, and named entity recognition. This book will get you started in the world of building literate, language understanding apps. Cutting edge ML tools from Apple like CreateML, CoreML, and TuriCreate will become natural parts of your development toolbox as you construct intelligent, text-based apps. You'll explore a wide range of text processing topics, including reprocessing text, training custom machine learning models, converting state-of-the-art NLP models to CoreML from Keras, evaluating models, and deploying models to your iOS apps. You’ll develop sample apps to learn by doing. These include apps with functions for detecting spam SMS, extracting text with OCR, generating sentences with AI, categorizing the sentiment of text, developing intelligent apps that read text and answers questions, converting speech to text, detecting parts of speech, and identifying people, places, and organizations in text. Smart app development involves mainly teaching apps to learn and understand input without explicit prompts from their users. These apps understand what is in images, predict future behavior, and analyze texts. Thanks to natural language processing, iOS can auto-fix typos and Siri can understand what you're saying. With Apple’s own easy-to-use tool, Create ML, they’ve brought accessible ML capabilities to developers. __Develop Intelligent iOS Apps with Swift__ will show you how to easily create text classification and numerous other kinds of models. What You'll Learn Incorporate Apple tools such as CreateML and CoreML into your Swift toolbox+ Convert state-of-the-art NLP models to CoreML from Keras + Teach your apps to predict words while users are typing with smart auto-complete Who This Book Is For Novice developers and programmers who wish to implement natural language processing in their iOS applications and those who want to learn Apple's native ML tools. Build smart apps capable of analyzing language and performing language-specific tasks, such as script identification, tokenization, lemmatization, part-of-speech tagging, and named entity recognition. This book will get you started in the world of building literate, language understanding apps. Cutting edge ML tools from Apple like CreateML, CoreML, and TuriCreate will become natural parts of your development toolbox as you construct intelligent, text-based apps. You'll explore a wide range of text processing topics, including reprocessing text, training custom machine learning models, converting state-of-the-art NLP models to CoreML from Keras, evaluating models, and deploying models to your iOS apps. You'll develop sample apps to learn by doing. These include apps with functions for detecting spam SMS, extracting text with OCR, generating sentences with AI, categorizing the sentiment of text, developing intelligent apps that read text and answers questions, converting speech to text, detecting parts of speech, and identifying people, places, and organizations in text. Smart app development involves mainly teaching apps to learn and understand input without explicit prompts from their users. These apps understand what is in images, predict future behavior, and analyze texts. Thanks to natural language processing, iOS can auto-fix typos and Siri can understand what you're saying. With Apple's own easy-to-use tool, Create ML, they've brought accessible ML capabilities to developers. Develop Intelligent iOS Apps with Swift will show you how to easily create text classification and numerous other kinds of models. What You'll Learn: Incorporate Apple tools such as CreateML and CoreML into your Swift toolbox ; Convert state-of-the-art NLP models to CoreML from Keras and TensorFlow ; Teach your apps to predict words while users are typing with smart auto-complete Table of Contents 5 About the Author 8 About the Technical Reviewer 9 Acknowledgments 10 Chapter 1: A Gentle Introduction to ML and NLP 11 What Is Machine Learning? 11 Supervised Learning 15 Unsupervised Learning 16 Basic Terminology of ML 17 What Is Deep Learning? 20 What Is Natural Language Processing 22 Summary 25 Chapter 2: Introduction to Apple ML Tools 26 Vision 26 Face and Body Detection 27 Image Analysis 28 Text Detection and Recognition 31 Other Capabilities of Vision 34 VisionKit 35 Natural Language 36 Language Identification 36 Tokenization 37 Part-of-Speech Tagging 39 Identifying People, Places, and Organizations 40 NLEmbedding 42 Speech 44 Core ML 45 Create ML 46 Turi Create 47 Chapter 3: Text Classification 49 Spam Classification with the Create ML Framework 49 Train a Model in macOS Playgrounds 51 Spam Classification with the Create ML App 65 Spam Classification with Turi Create 70 Turi Create Setup 70 Training a Text Classifier with Turi Create 72 Summary 75 Chapter 4: Text Generation 76 GPT-2 76 Let’s Build OCR and the Text Generator App 79 Using the Built-in OCR 81 Text Generation Using AI Model 85 Summary 92 Chapter 5: Finding Answers in a Text Document 93 BERT 93 Building a Question-Answering App 98 BERT-SQuAD 98 Examine the Core ML Model 99 Let’s Build the App 103 Using the BERT Model in iOS 104 Building the UI of the App 111 Speech Recognition with the Speech Framework 118 Summary 124 Chapter 6: Text Summarization 126 What Is Text Summarization? 126 Building the Text Summarizer App 128 Summary 140 Chapter 7: Integrating Keras Models 141 Converting the Keras Model into Core ML Format 141 Training the Text Classification Model in Keras 142 Testing the Core ML Model 151 Testing the Core ML Model in Jupyter Notebook 153 Testing the Core ML Model in Xcode 158 Using the Core ML Model in Xcode 161 Summary 168 Conclusion 168 Index 169 Front Matter ....Pages i-xiii A Gentle Introduction to ML and NLP (Özgür Sahin)....Pages 1-15 Introduction to Apple ML Tools (Özgür Sahin)....Pages 17-39 Text Classification (Özgür Sahin)....Pages 41-67 Text Generation (Özgür Sahin)....Pages 69-85 Finding Answers in a Text Document (Özgür Sahin)....Pages 87-119 Text Summarization (Özgür Sahin)....Pages 121-135 Integrating Keras Models (Özgür Sahin)....Pages 137-164 Back Matter ....Pages 165-169