Investigative analysts and researchers acquire clues and connect small bits of
evidence to uncover larger plans, stories, or narratives, and to
simply gain a better understanding of the information. Often, the
individual bits of evidence are short text documents or spreadsheets,
must examine large collections of such documents in order to "put the
pieces together" and formulate a well-supported hypothesis about
actions that may occur in the future. As the number of documents to
examine rises, it becomes more and more challenging for analysts to
understand the data and make judgments about it.
We are creating Jigsaw, a visual analytics system to help analysts and
better explore, analyze, and make sense of such document collections.
Our specific objective is to help analysts reach more timely and
accurate understandings of the larger stories and important concepts embedded throughout textual
reports. Jigsaw provides a collection of visualizations that each
portray different aspects of the documents. We particularly focus on
presenting the identifiable important entities (people, places,
organizations, etc.) and their direct or indirect connections. Textual
processing extracts the important entities from the documents and then
the visualizations help an analyst to explore the relationships and
connections among the entities. The system includes a variety of
visualizations such as list, graph, temproal and connection-based
views, as well as views
of individual document's text and the document collection as a whole.
Jigsaw essentially acts as a visual index onto the document
collection, helping analysts identify particular documents to read and
We have used Jigsaw to explore a wide variety of domains and document
collections including academic papers, grants, product reviews,
business press releases, news articles, intelligence and police
reports, statutes, and even books such as the Bible. Jigsaw is also
available for others to try and use.
For a good, short (2-page) overview of Jigsaw's goals and methods,
please see the Spring 2008 VacViews
article about the system.
To learn more about the system's details and its different
visualizations, please examine the Jigsaw views
page. And to see the system in action, explore the scenario
videos above or examine the video tutorial page.
Using Jigsaw as an analytic aid, we entered and won the university
division of the VAST 2007
Contest. To learn more about our contest entry, visit the Contest Summary page.
We presented a full paper about Jigsaw at the VAST 2007 Symposium and
an extended version of that paper in the Information
Visualization journal in 2008.
Two short papers also appearing in the VAST 2007 proceedings describe
our experiences working on the
contest. Furthermore, a video illustrating system capabilities is
available as well (links to the video and papers appear above). Please
contact Professor Stasko if you would like more information about the system.
The picture below shows Jigsaw being used on a computer with four monitors
presenting the different system views. The multitude of views in the
system makes a mutliple monitor computer like this desirable for analysis.
If four monitors aren't enough, Jigsaw can really hum on a display
wall. The picture below shows Carsten demonstrating Jigsaw running on
the PowerWall at Konstanz University (Germany) in Daniel Keim's lab.
Below is a picture of John demonstrating Jigsaw to Tim Collins of
Purdue at the 2009 NVAC Consortium meeting at PNNL.
This research is supported in part by the VACCINE DHS Center of
Excellence and NSF Awards IIS-0915788 and CCF-0808863 (FODAVA Lead).
Past support for the project came from a grant from
the Dept. of Homeland Security's NVAC
Program and NSF Award IIS-0414667.
modified: August 18, 2012