Report on New Methods for Representing and Interacting with Qualitative Geographic Information, Stage 2: Task Group 4 Message-Focused Use Case.

摘要

This report documents research in the GeoVISTA Center at The Pennsylvania State University (PSU) to develop visual analytics tools for leveraging data from microblogs, with a specific focus on Twitter as a primary data source. The tools are implemented within SensePlace2, a web-based application that provides a visual interface and display methods for query and exploration of large repositories of microblog data. Our emphasis is on revealing the where, when, what, and who components of microblog data to support situational awareness in natural disasters and other emergency events. This report summarizes project research to date and focuses specifically on outcomes for Task 4 in this research, which is directed to adding capabilities for message-focused queries and visual analysis of the query results. This augments the existing place-, tweeter-, and social-focused methods and tools completed for Tasks 1, 2 & 3, respectively. The report begins with a brief overview of work completed for Tasks 1, 2, and 3 to highlight achievements of this project to date and set the context for adding message-focused capabilities. This overview is followed by a comprehensive review of work completed in Task 4, divided into four sections that report on: (a) the sources of spatial information that can be used to map spatial aspects of Twitter information diffusion, (b) system extensions to collect data on Twitter-related message propagation and to store and enable quick queries of this information, (c) description of new SensePlace2 functionality for depicting retweet connections, and (d) advances in system architecture developed and implemented to support quick access to the growing volumes of microblog data. Finally, we present conclusions and outline some future research challenges.