Text Analytics with Python

University of Canterbury via edX Professional Certificate

Overview

Learn the core techniques of text analytics and natural language processing (NLP) while discovering the cognitive science that makes it possible in this certificate Text Analytics with Python. On the practical side, you’ll learn how to actually do an analysis in Python: creating pipelines for text classification and text similarity using machine learning. These pipelines are automated workflows that go all the way from data collection to visualization. On the scientific side, you’ll learn what it means to understand language computationally. Artificial intelligence and humans don’t view text documents in the same way. Sometimes deep learning sees patterns that are invisible to us. But often deep learning misses the obvious. We have to understand the limits of a computational approach to language together with the ethical requirements that guide how we choose what data to use and how we protect the privacy of individuals.

Along the way, you’ll explore real-world case studies using pandas, numpy, scikit-learn, tensorflow, matplotlib, seaborn, gensim, and spacy within jupyter notebooks to gain useful insights from unstructured data.

Syllabus

Courses under this program:
Course 1: Text Analytics 1: Introduction to Natural Language Processing

Learn the core techniques of computational linguistics alongside the cognitive science that makes it all possible and the ethics we need to use it properly.

Course 2: Text Analytics 2: Visualizing Natural Language Processing

Extend your knowledge of the core techniques of computational linguistics by working through case-studies and visualizing their results.

Courses

0 reviews
6 weeks, 3-6 hours a week, 3-6 hours a week
View details

__ _ Visualizing Natural Language Processing _ is the second course in the Text Analytics with Python professional certificate (or you can study it as a stand-alone course). Natural language processing (NLP) is only useful when its results are meaningful to humans. This second course continues by looking at how to make sense of our results using real-world visualizations.

How can we understand the incredible amount of knowledge that has been stored as text data? This course is a practical and scientific introduction to text analytics. That means you’ll learn how it works and why it works at the same time.

On the practical side, you’ll learn how to visualize and interpret the output of text analytics. You’ll learn how to create visualizations ranging from word clouds, heatmaps, and line plots to distribution plots, choropleth maps, and facet grids. You’ll work through real case-studies using jupyter notebooks and to visualize the results of machine learning in Python using packages like pandas, matplotlib, and seaborn.

On the scientific side, you’ll learn what it means to understand language computationally. How do word embeddings and topic models relate to human cognition? Artificial intelligence and humans don’t view language in the same way. You’ll see how both deep learning and human beings interact with the meaning that is encoded in language.
0 reviews
6 weeks, 3-6 hours a week, 3-6 hours a week
View details

Introducing Natural Language Processing is part one of the Text Analytics with Python professional certificate (or you can study it as a stand-alone course). This first course introduces the core techniques of natural language processing (NLP) and computational linguistics. But we introduce these techniques from data science alongside the cognitive science that makes them possible.

How can we make sense out of the incredible amount of knowledge that has been stored as text data? This course is a practical and scientific introduction to natural language processing. That means you’ll learn how it works and why it works at the same time.

On the practical side, you’ll learn how to actually do an analysis in Python: creating pipelines for text classification and text similarity that use machine learning. These pipelines are automated workflows that go all the way from data collection to visualization. You’ll learn to use Python packages like pandas, scikit-learn, and tensorflow.

On the scientific side, you’ll learn what it means to understand language computationally. Artificial intelligence and humans don’t view documents in the same way. Sometimes AI sees patterns that are invisible to us. But other times AI can miss the obvious. We have to understand the limits of a computational approach to language and the ethical guidelines for applying it to real-world problems. For example, we can identify individuals from their tweets. But we could never predict future criminal behaviour using social media.

This course will cover topics you may have heard of, like text processing, text mining, sentiment analysis, and topic modeling.