Infomaps: Integrated data and document visualization and analysis

Analysts use visual analytical systems for exploring and analyzing structured datasets. Large multi-dimensional data have been visualized effectively and interactively with numerous classic and modern visualizations such as scatterplots, parallel coordinates, line graphs, etc.. The analysts also require tools to access supporting documents, research papers and news reports for their research work and they typically depend on information retrieval systems for gathering related documents. Information retrieval systems have been built to help users gain access to ever growing information sources and to analyze that information. Any information retrieval system involves two major processes: retrieving information and presenting the information to the users. While traditionally the retrieved information is displayed in a list-based textual format, there have been many examples using visualization techniques to render the returned information. However, the visual analytical systems built for text corpora typically concentrate on techniques for exploring only document collections. We have developed a system for visualizing and analyzing both document collections and structured datasets. This dissertation focuses on the design and implementation of an architecture for two-way query generation. It includes a visual interface for information retrieval called InfoMaps that is integrated within a visual analytical framework. Queries are generated from structured data for information retrieval and from documents for visual data selection. We describe the architecture design and framework developed in integrating the information retrieval and document management system with the open source visual analytic system Weave. We discuss the methodology applied in generating queries from structured datasets and establishing two-way linking between the structured data sets and the document collections retrieved from the queries. We will also present our work on visualizing the query results and tools for exploring and analyzing the document collections with the structured data sets and provide example applications.