The goal of this PhD project is to investigate Multidimensional Visualization techniques to support exploratory data analysis tasks related to identifying relevant attributes to understand the behavior of phenomena involving multiple variables. We seek to investigate and propose a visual framework for analyzing multidimensional data that describe a phenomenon´s behavior over time, assisting data analysts in identifying the relevant attributes. The proposed framework must be interactive and allow users to drive the analysis process and input his/her domain knowledge. This project involves a partnership with researchers from the Department of Cartography of the Faculty of Science and Technology of UNESP at Presidente Prudente. These researchers study the effects of ionospheric scintillation on GPS measurements, and are developing a database that records observations concerning this purpose. This problem features an ideal application domain to identify the requirements and validate the techniques to be proposed inthis project, as it involves the investigation of the role of different attributes in the temporal behavior of the phenomenon. The proposed framework must be applicable to this and other problems of a similar nature.
News published in Agência FAPESP Newsletter about the scholarship: