Book Image

Mastering spaCy

By : Duygu Altınok
Book Image

Mastering spaCy

By: Duygu Altınok

Overview of this book

spaCy is an industrial-grade, efficient NLP Python library. It offers various pre-trained models and ready-to-use features. Mastering spaCy provides you with end-to-end coverage of spaCy's features and real-world applications. You'll begin by installing spaCy and downloading models, before progressing to spaCy's features and prototyping real-world NLP apps. Next, you'll get familiar with visualizing with spaCy's popular visualizer displaCy. The book also equips you with practical illustrations for pattern matching and helps you advance into the world of semantics with word vectors. Statistical information extraction methods are also explained in detail. Later, you'll cover an interactive business case study that shows you how to combine all spaCy features for creating a real-world NLP pipeline. You'll implement ML models such as sentiment analysis, intent recognition, and context resolution. The book further focuses on classification with popular frameworks such as TensorFlow's Keras API together with spaCy. You'll cover popular topics, including intent classification and sentiment analysis, and use them on popular datasets and interpret the classification results. By the end of this book, you'll be able to confidently use spaCy, including its linguistic features, word vectors, and classifiers, to create your own NLP apps.
Table of Contents (15 chapters)
1
Section 1: Getting Started with spaCy
4
Section 2: spaCy Features
9
Section 3: Machine Learning with spaCy

Introducing NER

We opened this chapter with a tagger, and we'll see another very handy tagger—the NER tagger of spaCy. As NER's name suggests, we are interested in finding named entities.

What is a named entity? A named entity is a real-world object that we can refer to by a proper name or a quantity of interest. It can be a person, a place (city, country, landmark, famous building), an organization, a company, a product, dates, times, percentages, monetary amounts, a drug, or a disease name. Some examples are Alicia Keys, Paris, France, Brandenburg Gate, WHO, Google, Porsche Cayenne, and so on.

A named entity always points to a specific object, and that object is distinguishable via the corresponding named entity. For instance, if we tag the sentence Paris is the capital of France, we parse Paris and France as named entities, but not the word capital. The reason is that capital does not point to a specific object; it's a general name for many objects.

...