Python Natural Language Processing Cookbook : Over 60 Recipes for Building Powerful NLP Solutions Using Python and LLM Libraries / Zhenya Antic and Saurabh Chakravarty.

Format:

Book

Author/Creator:

Antić, Zhenya, author.

Chakravarty, Saurabh, author.

Language:

English

Subjects (All):

Natural language processing (Computer science).

Python (Computer program language).

Physical Description:

1 online resource (312 pages)

Edition:

Second edition.

Place of Publication:

Birmingham, England : Packt Publishing Ltd., [2024]

Biography/History:

Antic Zhenya: Zhenya Antic, Ph. D. is an expert in AI and NLP. She is currently the Director of AI Automation at Arch Insurance, where she leads initiatives in Intelligent Document Processing and applies various AI solutions to complex problems. With extensive consulting experience, Zhenya has worked on numerous NLP projects with various companies. She holds a Ph. D. in Linguistics from the University of California, Berkeley, and a B. S. in Computer Science from the Massachusetts Institute of Technology. Chakravarty Saurabh: Saurabh Chakravarty, Ph. D. is a seasoned veteran in the software industry with over 20 years of experience in software development. A software developer at heart, he is passionate about programming. He has held various roles, including architect, lead engineer, and software developer, specializing in AI and large-scale distributed systems. Saurabh has worked with Microsoft, Rackspace, and Accenture, as well as with a few startups. He holds a Ph. D. in Computer Science with a specialization in NLP from Virginia Tech, USA. Saurabh lives in California with his wife, Tina, and daughter, Aaliya, and works for AWS in Santa Clara, California.

Summary:

Harness the power of Natural Language Processing to overcome real-world text analysis challenges with this recipe-based roadmap written by two seasoned NLP experts with vast experience transforming various industries with their NLP prowess. You’ll be able to make the most of the latest NLP advancements, including large language models (LLMs), and leverage their capabilities through Hugging Face transformers. Through a series of hands-on recipes, you’ll master essential techniques such as extracting entities and visualizing text data. The authors will expertly guide you through building pipelines for sentiment analysis, topic modeling, and question-answering using popular libraries like spaCy, Gensim, and NLTK. You’ll also learn to implement RAG pipelines to draw out precise answers from a text corpus using LLMs. This second edition expands your skillset with new chapters on cutting-edge LLMs like GPT-4, Natural Language Understanding (NLU), and Explainable AI (XAI)—fostering trust and transparency in your NLP models. By the end of this book, you'll be equipped with the skills to apply advanced text processing techniques, use pre-trained transformer models, build custom NLP pipelines to extract valuable insights from text data to drive informed decision-making.

Contents:

Cover

Title page

Copyright and credits

Dedication

Contributors

Table of Contents

Preface

Chapter 1: Learning NLP Basics

Technical requirements

Dividing text into sentences

Getting ready

How to do it…

There's more…

See also

Dividing sentences into words - tokenization

How to do it

Part of speech tagging

There's more

Combining similar words - lemmatization

Removing stopwords

Chapter 2: Playing with Grammar

Counting nouns - plural and singular nouns

Getting the dependency parse

Extracting noun chunks

Extracting subjects and objects of the sentence

Finding patterns in text using grammatical information

Chapter 3: Representing Text - Capturing Semantics

Creating a simple classifier

Putting documents into a bag of words

Constructing an N-gram model

Representing texts with TF-IDF

How it works…

Using word embeddings

Training your own embeddings model

Using BERT and OpenAI embeddings instead of word embeddings.

Getting ready

Retrieval augmented generation (RAG)

Chapter 4: Classifying Texts

Getting the dataset and evaluation ready

Performing rule-based text classification using keywords

Clustering sentences using K-Means - unsupervised text classification

Using SVMs for supervised text classification

Training a spaCy model for supervised text classification

Classifying texts using OpenAI models

Chapter 5: Getting Started with Information Extraction

Using regular expressions

Finding similar strings - Levenshtein distance

Extracting keywords

Performing named entity recognition using spaCy

Training your own NER model with spaCy

Fine-tuning BERT for NER

Chapter 6: Topic Modeling

LDA topic modeling with gensim

How to do it...

There's more...

Community detection clustering with SBERT

K-Means topic modeling with BERT

Topic modeling using BERTopic

Using contextualized topic models

Chapter 7: Visualizing Text Data

Technical requirements.

Visualizing the dependency parse

Visualizing parts of speech

Visualizing NER

Creating a confusion matrix plot

Constructing word clouds

Visualizing topics from Gensim

Visualizing topics from BERTopic

Chapter 8: Transformers and Their Applications

Loading a dataset

Tokenizing the text in your dataset

Classifying text

Using a zero-shot classifier

Generating text

Language translation

Chapter 9: Natural Language Understanding

Answering questions from a short text passage

Answering questions from a long text passage

Answering questions from a document corpus in an extractive manner

Answering questions from a document corpus in an abstractive manner

Summarizing text using pre-trained models based on Transformers

Detecting sentence entailment

Enhancing explainability via a classifier-invariant approach

Enhancing explainability via text generation

Getting ready.

How to do it

Chapter 10: Generative AI and Large Language Models

Model access

Running an LLM locally

Running an LLM to follow instructions

Augmenting an LLM with external data

Executing a simple prompt-to-LLM chain

Augmenting the LLM with external content

Creating a chatbot using an LLM

Generating code using an LLM

Generating a SQL query using human-defined requirements

Agents - making an LLM to reason and act

Using OpenAI models instead of local ones

Index

Other Books You May Enjoy.

Notes:

Includes index.

Description based on publisher supplied metadata and other sources.

Description based on print version record.

ISBN:

1-80324-144-6

OCLC:

1456999696