Blueprints for Text Analytics using Python: Machine Learning Based Solutions for Common Real World (NLP) Applications

This practical book provides data scientists and developers with blueprints for best practice solutions to common tasks in text analytics and natural language processing, with real-world case studies and detailed code examples in Python.

Format: Paperback / softback
Length: 350 pages
Publication date: 22 December 2020
Publisher: O'Reilly Media, Inc, USA

For businesses seeking to gain a competitive edge, the transformation of text into valuable information is paramount. The recent advancements in natural language processing (NLP) have opened up a multitude of options for addressing complex challenges. However, determining which NLP tools or libraries are best suited for a particular business's needs, as well as the appropriate techniques to employ and in what sequence, can be a daunting task. This comprehensive book aims to bridge this gap by offering data scientists and developers practical blueprints for achieving best-practice solutions in text analytics and NLP. Authored by Jens Albrecht, Sidharth Ramachandran, and Christian Winkler, the book provides real-world case studies and detailed code examples in Python, enabling readers to get started quickly.

In Chapter 1, the authors delve into the process of extracting data from APIs and web pages, highlighting the importance of preparing textual data for statistical analysis and machine learning. They introduce techniques such as text preprocessing, tokenization, and stemming, which are essential for effective text analysis. Chapter 2 explores the realm of machine learning, covering classification, topic modeling, and summarization. The authors introduce popular machine learning algorithms like Support Vector Machines (SVMs), Naive Bayes, and Long Short-Term Memory (LSTM) networks, and demonstrate their application in text analytics tasks. They also discuss the importance of evaluating and interpreting AI models and classification results, providing insights into how to assess the performance of text-based algorithms.

Chapter 3 delves into the exploration and visualization of semantic similarities with word embeddings. The authors introduce the concept of word embeddings, which represent words as vectors in a low-dimensional space. They discuss the use of word embeddings for tasks such as identifying related words, understanding sentence context, and generating natural language responses. The chapter includes practical examples of using word embeddings in text analysis and machine learning applications.

Chapter 4 focuses on identifying customer sentiment in product reviews. The authors introduce sentiment analysis, a technique that involves analyzing text data to determine the overall positivity, negativity, or neutrality of a sentiment. They discuss the use of sentiment analysis algorithms, such as Sentiment Analysis Toolbox (SANT), and provide insights into how to evaluate and interpret sentiment analysis results. The chapter also includes practical examples of analyzing customer reviews and identifying trends in customer sentiment.

Chapter 5 explores the creation of a knowledge graph based on named entities and their relations. Named entities refer to individuals, organizations, locations, and other relevant entities mentioned in text data. The knowledge graph represents these entities and their relationships, providing a structured representation of the information contained in the text. The authors introduce the concept of entity recognition and linking, and demonstrate how to build a knowledge graph using popular libraries such as GraphDB and Neo4j. The chapter includes practical examples of using knowledge graphs for tasks such as information retrieval, knowledge discovery, and recommendation systems.

In conclusion, this practical book offers data scientists and developers a comprehensive guide to text analytics and NLP. With its real-world case studies, detailed code examples, and best-practice solutions, it empowers readers to tackle complex challenges and extract valuable information from text. Whether you are a beginner or an experienced practitioner, this book will provide you with the tools and knowledge needed to excel in text analytics and NLP.

Weight: 732g
Dimension: 178 x 235 x 28 (mm)
ISBN-13: 9781492074083