Making Sense of VAST Data

: We view the task of sensemaking in intelligence as that of abducing a story whose plot explains the current data and makes verifiable predictions about the future and the past. We have developed a computational system, called STAB, that abduces stories from data. The story plots in STAB are represented as processes with goals and states, nized in an abstraction hierarchy. STAB analyzes the VAST dataset generated by PNNL. This dataset pertains to normal and typical activities, as well as illegal and unethical activities, in a fictitious town in the United States. Given the VAST data incrementally, STAB retrieves and invokes multiple story plots as explanatory hypotheses and generates expectations about future data. It uses supporting and contradicting evidence to build justifications for its final conclusions.

• Type: Conference
• Year: 2007
• Where: ISI
• Authors: Summer Adams, Ashok K. Goel
• Keywords: ISI 2007, Multiple Story Plots, Security Privacy, Story Plots, VAST Dataset