Natural Language Processing For Hackers

Preface

Introduction

1. What is Natural Language Processing?
2. Challenges in Natural Language Processing
3. What makes this book different?
4. Part 1: Introduction to NLTK

NLTK Fundamentals

1. Installing NLTK
2. Splitting Text
3. Building a vocabulary
4. Fun with Bigrams and Trigrams
5. Pointwise Mutual Information
6. Part Of Speech Tagging
7. Named Entity Recognition

Getting started with Wordnet

1. Wordnet Structure
2. Lemma Operations

Lemmatizing and Stemming

1. How stemmers work
2. How lemmatizers work
3. Part 2: Create a Text Analysis service

Introduction to Machine Learning

1. A Practical Machine Learning Example

Getting Started with Scikit-Learn

1. About NumPy
2. Installing Scikit-Learn and building a dataset
3. Training a Scikit-Learn Model
4. Making Predictions

Finding the data

1. Existing corpora
2. Ideas for Gathering Data
3. Getting the Data

Learning to Classify Text

1. Text Feature Extractor
2. Scikit-Learn Feature Extraction
3. Text Classification with Naive Bayes

Persisting models

Building the API

1. Building a Flask API
2. Deploy to Heroku
3. Part 3: Create a Social Media Monitoring Service

Basics of Sentiment Analysis

1. Be Aware of Negations
2. Machine Learning doesn’t get Humour
3. Multiple and Mixed Sentiments
4. Non-Verbal Communication

Twitter Sentiment Data

1. Twitter Corpora
2. Other Sentiment Analysis Corpora
3. Building a Tweets Dataset
4. Sentiment Analysis - A First Attempt
5. Better Tokenization

Fine Tuning

1. Try a different classifier
2. Use Ngrams Instead of Words
3. Using a Pipeline
4. Cross Validation
5. Grid Search
6. Picking the Best Tokenizer

Building the Twitter Listener

Classification Metrics

1. Binary Classification

Multi-Class Metrics

1. The Confusion Matrix
2. Part 4: Build Your Own NLP Toolkit

Build Your Own Part-Of-Speech Tagger

1. Part-Of-Speech Corpora
2. Building Toy Models
3. About Feature Extraction
4. Using the NLTK Base Classes
5. Writing the Feature Extractor
6. Training the Tagger
7. Out-Of-Core Learning

Build a Chunker

1. IOB Tagging
2. Implementing the Chunk Parser
3. Chunker Feature Detection

Build a Named Entity Extractor

1. NER Corpora
2. The Groningen Meaning Bank Corpus
3. Feature Detection
4. NER Training

Build a Dependency Parser

1. Understanding the Problem
2. Greedy Transition-Based Parsing
3. Dependency Dataset
4. Writing the Dependency Parser Class

Adding Labels to the Parser

1. Learning to Label Dependencies
2. Training our Labelled Dependency Parser
3. Part 5: Build Your Own Chatbot Engine

General Architecture

1. Train the Platform via Examples
2. Action Handlers

Building the Core

1. Chatbot Base Class and Training Set
2. Training the Chatbot
3. Everything together

MovieBot

1. The Movie DB API
2. Small-Talk Handlers
3. Simple Handlers
4. Execution Handlers

MovieBot on Facebook

1. Installing ngrok
2. Setting up Facebook
3. Trying it Out
4. What Next?