Day 72 – spaCy Natural Language Processing in Python

Image Source: spaCy

What is spaCy?

spaCY is an open-source library for Natural Language Processing that can be used on dependency parsing, sentence segmentation, text classification, etc. It is one of the leading Industrial-Strength Natural Language Processing.  We can integrate with the machine learning framework and build custom components quickly.

Why we spaCy?

If one of your projects requires named-entity recognition, you don’t need to write code from scratch. You can call spaCy components and write a few lines, then test it.

spaCy Features:

• Supports more than 64 languages
• Pretrained word vectors
• Readymade components available named entity recognition, part-of-speech tagging, dependency parsing, sentence segmentation, text classification, lemmatization, morphological analysis, entity linking, etc
• Pretrained transformers
• Built-in visualizer for syntax and NER
• many more.

At the time of writing, the spaCy has a version 3.0

spaCy Projects:

spaCy has many projects built by its community, and you can check and use those projects.

• Pipeline
• Training
• Conversational
• Research
• Scientific
• Visualizers
• Containers & APIs
• Non-Python
• Standalone
• Models

What are available trained pipelines?

The trained pipelines available based on language selection.  For example, if you select English, then you can see below components and pipelines.

Components

• tok2vec
• tagger
• parser
• senter
• ner
• attribute_ruler
• lemmatizer

Pipeline

• tok2vec
• tagger
• parser
• ner
• attribute_ruler
• lemmatizer

We can write some codes in the next blogs.

Please comment what’s your view on the spaCy.

Have you used it in your project?