Skip to Main Content
This work presents a significant extension of the information flocking concept and algorithms originally presented by G. Proctor and C. Winter (1998). It introduces a novel way of visualizing time-varying datasets using the emergent characteristics of self organization and dynamic behavior simulation. The current prototype uses both spatial clustering and behavioral animation to represent temporal data similarities by simulating the time-varying evolution of dynamic datasets. Instead of presenting exact data values, the way how the data values change over time is being visualized. In addition, the current information flocking method is capable of visualizing short-term temporal events or long-term time-varying data evolutions by automatically generating different recognizable motion typologies. This research show how artificial life principles have been merged with the field of information visualization. Several aspects of the original information flocking algorithms have been improved to incorporate the real-time evaluation of continuous dynamic data value streams and to generate multiple stable, recognizable atomic as well as collective dynamic behaviors that reflect time-varying dataset changes and relative data value evolutions. The main information flocking principles are demonstrated through a visualization of historical stock market quotes.