A survey of techniques to visualize streaming textual datasets

In this paper we examine the applicability of several well know document visualization programs to the newly arising problem of visualizing the processing of streams of textual datasets. In the military intelligence community, many new forms of information are arriving as textual streams. In addition to the traditional wire services and intelligence summaries available in electronic form, the Internet and 24/7 international news coverage present huge quantities of open source data to the analyst in streaming textual form. When dealing with the modern asymmetric threat, modern analysts are forced to utilize open sources of information more and more. These analysts are hard pressed to handle this new influx of information using traditional relational data base search and aggregate techniques. We identify several view points into a modern textual stream processing flow where the operation and efficiency of the system can be improved by the appropriate visualization technique. We examine some of the physical scalability limits of visualizing large quantities of information. We then put forth several criteria and evaluate some well known text visualization systems for use in this new environment