Keywords: text data analytics, visualization, cluster analysis, emergence, dimensionality reduction, fusion
This talk will discuss work performed over the last 14 years which has been focused on the application of text data analytics to better understand semi-structured document collections. These document collections have included scientific paper meta-data, patents, and other textual datasets. The talk will discuss previous work in the area of visualization of cluster solutions involving thousands of clusters, methods for temporal analysis including identification of emergent and pre-emergent topic areas, and strategies for the fusion of multiple indicators as calculated on the clusters in order to produce meaningful orderings of the clusters within the solution.