Introduction to Text Mining in R

November 16, 2016

posts R

Introduction

This workshop will introduce you to text analysis techniques in R, an open source programming language. Some familiarity with R is expected as a requirement to attend this course. You can attend the Intro to R course we provide or take this online tutorial https://www.datacamp.com/courses/free-introduction-to-r from DataCamp.

Topics

Preprocessing text
Creating a DTM (document term matrix)
Working with DTMs
Word frequencies
Wordclouds - of course!
Discriminating / Distinctive Words
Time permitting: Dictionary Methods (i.e. Sentiment Analysis)

Dates

November 16, 2016 (9:00 AM - 12:00 PM)

Location

Biomedical Library Classroom 4

Things not covered

Document Similarity
Topic modeling
Clustering

Instructors

@u2ng - Reid Otsuji
@jt14den - Tim Dennis

Audience

All graduate students and researchers.

Setup

This lesson assumes you have the R, RStudio software installed on your computer.

R can be downloaded here.

We will use the following packages in R, if you can, install prior to class:

RStudio is an environment for developing using R. It can be downloaded here. You will need the Desktop version for your computer.

Required R Packages

tm # text mining in R
RTextTools # a machine learning package for text classification
qdap # quantiative discourse analysis
qdapDictionaries # for sentiment analysis, etc
entropy # tools applying Information Theory
dplyr # data preparation and pipes $>$
ggplot2 # for plotting
SnowballC # for stemming
matrixStats # for stats
data.table # for easier data manipulation
scales # to help us plot
lsa # latent semantic analysis
cluster # for clustering analysis
fpc # flexible procedures for clustering
mallet # a wrapper around the Java machine learning tool MALLET
wordcloud # to visualize wordclouds
rJava # dependency for mallet
Any dependencies to the packages above.