Web-based Visualization and Querying of Food and Beverage Endorsements by Celebrities

This paper explores the development of a web-based dashboard for the exploration of the nature of food and beverage endorsements by celebrities in the United States. While traditional methods of analysis require the translation of user requests to queries in one of the many languages that analysts use, the end user with no formal training in these languages is usually unable to get meaningful information without assistance. In this work, we propose the creation of a web-based dashboard that allows the user to use widgets to submit queries that can be executed on the dataset. To meet the design requirements of visualizing this data, two visual encodings are matched to the dataset under consideration, which illustrate the hierarchical relationships that exist between the various entities, within the confines of limited visual space. Finally, the dashboard concept is extended to a dataset larger than machine RAM using out-of-core processing that allows users to work on such data without resorting to distributed resources.