The objective of this project is to generate word storms, multiples of word clouds to afford visual comparison of groups of documents. Just as a storm is a group of clouds, a word storm is a group of word clouds. Each cloud in the storm represents a subset of the corpus. For example, a storm might contain one cloud per document, or alternatively one cloud to represent all the documents written in each year, or one cloud to represent each track of an academic conference, etc.
In order to make the clouds easy to compare, words that appear in different clouds are placed in similar locations and more informative words are emphasized. In this way, similar documents are be represented by visually similar clouds.
This project allows you to create word storms from your texts, giving you options to personalize their appearance, so you can create beautiful figures to analyze your documents.
The algorithms to generate the storm are build on top of WordCram (web, gitHub), a flexible word cloud generator that creates beautiful clouds using the popular Wordle's layout.
- Set the path to the text files using 'io.DataPath.java'. The project takes the data from 'datapath/input/folder' and returns the storm in 'datapath/output/folder/parameters'
- Decide the cloud's size in 'AppletConf'.
- Decide the storm configuration using 'StormConf' (number of files, number of words, colors, angles...)
- Create a Storm!
See examples of usage in 'src/wslauncher':
- 'StormLauncher.java' to create a coordinated word storm.
- 'IndepLauncher.java' to create independent clouds.