Semantic-Preserving Word Clouds by Seam Carving

Yingcai Wu¹    Thomas Provan²     Furu Wei¹     Shixia Liu¹     Kwan-Liu Ma²
This project was conducted when Yingcai Wu worked in UC Davis.
¹Microsoft Research Asia      ²University of California, Davis

System overview: the system has three major components: a preprocessing component for extracting keywords and creating a preliminary layout, a word cloud generation component for creating compact and semantic-preserving word clouds, and a visualization component for visualizing the word clouds.

Word clouds are proliferating on the Internet and have received much attention in visual analytics. Although word clouds can help users understand the major content of a collection of documents quickly, their ability to visually compare documents is limited. This paper introduces a new method to create semantic-preserving word clouds by leveraging tailored seam carving, a well-established content-aware image resizing operator. The method can optimize a word cloud layout by removing a left-to-right or top-to-bottom seam iteratively and gracefully from the layout. Each seam is a connected path of low energy regions determined by a Gaussian-based energy function. With seam carving, we can pack the word cloud compactly and effectively, while preserving its overall semantic structure. Furthermore, we design a set of interactive visualization techniques for the created word clouds to facilitate visual text analysis and comparison. Case studies are conducted to demonstrate the effectiveness and usefulness of our techniques.

@article {YWu2011a,
author = {Yingcai Wu and Thomas Provan and Furu Wei and Shixia Liu and Kwan-Liu Ma},
title = {Semantic-Preserving Word Clouds by Seam Carving} ,
journal = {Computer Graphics Forum},
year = {2011},
volume = {30},
number = {3},
pages = {741--750}
}