README.md

Getting the most out of text - Using supervised and unsupervised approaches in NLP

📍 R-Ladies Tunis

About the talk

Text data provides an oasis of information for both researchers and non-researchers alike to explore. Natural Language Processing (NLP) helps to make sense of this data type. The talk will give you a smooth introduction to the quanteda package quanteda package. I will also showcase some visualizations of the results as well as supervised and unsupervised approaches in NLP. As part of a code demo, we will use text data from the UN as a working example to give you first insights into the structure of text data and how to work with it.

Material

You can access the material here:

Slides	Code (raw)	Code (deployed)	Recording

Overview repository

This repository contains:

├── code
│   ├── code_full.Rmd
│   └── code_full.html
├── data
│   ├── UN-data.RData
│   └── policy_agendas_english.lcd
├── figures
│   ├── stm-evaluation.png
│   └── terms.png
└── nlp-rladies-tunis.Rproj

More resources

• Kohei Watanabe and Stefan Müller: Quanteda Tutorials
• Cosima Meyer and Cornelius Puschmann: Advancing Text Mining with R and quanteda