HappyR

HappyR (pronounced like "happy hour") is an application for analyzing sentiment and emotions in text using various lexicons and natural language processing techniques.

Features

• Text input via direct typing or loading a text file
• Overall sentiment scoring
• Positive/negative word breakdown
• Emotion analysis with 8 different emotions
• Word frequency visualization
• Interactive word cloud
• Cross-platform web interface

Technologies Used

• R: Primary programming language
• Shiny: Web application framework
• tidytext: Text analysis framework
• Plotly: Interactive data visualization
• wordcloud2: Word cloud generation

How It Works

The application analyzes text using multiple complementary sentiment lexicons:

1. AFINN Lexicon: Words scored on a scale from -5 (highly negative) to +5 (highly positive)
2. Bing Lexicon: Binary classification of words as positive or negative
3. NRC Emotion Lexicon: Categorization of words into emotions (anger, fear, joy, etc.)
4. Syuzhet: Alternative sentiment analysis method

Installation

1. Clone this repository.

git clone https://github.com/aileks/HappyR.git

2. Install the packages in your R Studio console.

# Install required packages
install.packages(c("shiny", "shinydashboard", "tidyverse", "tidytext", 
                  "textdata", "syuzhet", "plotly", "wordcloud2", 
                  "tm", "stringr", "DT"))

# Run the application
shiny::runApp("text-sentiment-analyzer")

Citations and Acknowledgments

This application uses the following sentiment lexicons:

AFINN Lexicon

Nielsen, F. Å. (2011). A new ANEW: Evaluation of a word list for sentiment analysis in microblogs. Proceedings of the ESWC2011 Workshop on 'Making Sense of Microposts': Big things come in small packages. Volume 718 in CEUR Workshop Proceedings: 93-98.

Bing Lexicon

Hu, M., & Liu, B. (2004). Mining and summarizing customer reviews. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD-2004).

NRC Emotion Lexicon

Mohammad, S. M., & Turney, P. D. (2013). Crowdsourcing a Word-Emotion Association Lexicon. Computational Intelligence, 29(3): 436-465.

Syuzhet

Jockers, M. (2017). Syuzhet: Extract Sentiment and Plot Arcs from Text. https://github.com/mjockers/syuzhet

R Packages

This application relies on several R packages, particularly:

• tidytext: Silge, J., & Robinson, D. (2016). tidytext: Text Mining and Analysis Using Tidy Data Principles in R. JOSS, 1(3). doi:10.21105/joss.00037
• textdata: Hvitfeldt, E. (2020). textdata: Download and Load Various Text Datasets. https://github.com/EmilHvitfeldt/textdata