by Luc Anselin
Abstract: This talk focuses on a number of important computational aspects associated with the implementation of spatial data science functionality in software. It provides a brief overview of the evolution of software, particularly my own efforts which include SpaceStat in the 1990s and GeoDa and PySAL in the more recent past. Particular attention is paid to our current work on moving such capability from the desktop to a cyberinfrastructure environment, or CyberGIS. This involves refining/developing new algorithms, parallelization, provenance and metadata tracking and managing a web services infrastructure. The talk concludes with an illustration of a prototype for the latest GeoDa-Web cloud implementation for exploratory spatial data analysis and spatial regression.