10) Text Mining 2
π Lecture
One popular topic in text mining involves Sentiment Analysis (aka Sentiment Scoring). Simply put, sentiment analysis is the process of extracting an authorβs emotional intent from text. At first glance, sentiment analysis may appear easy. But as we will see, this type of analysis is tricky to do well. There are various emotional frameworks that can be used for sentiment analysis. We will cover a simple approach that implies stating whether a document is positive, negative, or neutral. This is referred to as the polarity of a document.
π Reading
Read chapters 1, 2 and 4 of βText Mining with Rβ (by Julia Silge and David Robinson):
π¬ Lab
- Getting started with the
"tidytext"
package.
π― Objectives
- Define sentiment analysis.
- Describe a basic polarity scoring algorithm.
- Compute basic sentiments (positive, negative, neutral) of a text.
π Assignments
- Shiny App1, due 03/22