Collection Understanding (Logging, Subcoll)
(Requires WinXP, IE5+ and Java)

Collection understanding shifts the traditional focus of image retrieval in large collections from locating specific images to gaining a comprehensive view of the collection. Users visualize the collection by changing metadata values dynamically to create their own understanding of the collection. As users narrow or expand their view, they acquire the general gist of why the images belong to the collection and what characteristics the images share.

When users interactively filter metadata and perceive varying viewpoints of the images, they are then inspired to derive their own relationships among the images and come to a fuller understanding of the collection.

Through novel visualizations, users will be able to efficiently understand what specific collections have to offer. Collection insight will increase and this will enhance information access, making digital collections more valuable knowledge assets.

JCDL 2004 Slides
Tucson, Arizona, June 7-11, 2004

Thesis (Word)

Thesis Defense Slides (Powerpoint)

Thesis Proposal: Collection Understanding (Word)

JCDL 2003 IVIRA Workshop

Houston, Texas, May 2003
Chang, Michelle; Leggett, John. “Collection Understanding Through Streaming Collage”

(Word, Adobe PDF)

Collection Understanding Presentation Slides


Figure 1: Collection Understanding: Ambient Picasso



Figure 2: Collection Understanding: Picasso Collage

Home | Location | HCSI | Thesis | CSCW | Information Retrieval | Hypertext | Recombinant Media | Intelligent User Interfaces

For questions or comments, please e-mail Michelle
Texas A&M University
College Station, TX.