SOVAS: A Scalable Online Visual Analytic System
Most existing online processing and analytics systems for climate studies only support fixed user interface with predefined functions. These systems are often not scalable to handle massive climate data that could easily accumulate terabytes daily. To address the major limitations of existing online systems for climate studies, we develop a scalable online visual analytic system, known as SOVAS, to balance both usability and flexibility. SOVAS, enabled by a set of key techniques, supports large-scale climate data analytics and knowledge discovery in a scalable and sharable environment.
SOVAS Website: https://gidbusc.github.io/SCOVAS
SOVAS Portal: http://gis.cas.sc.edu/scovas
Li Z., Huang Q., Jiang Y., Hu F. (2019) , SOVAS: A Scalable Online Visual Analytic System for Big Climate Data Analysis, International Journal of Geographic Information Science, DOI: 10.1080/13658816.2019.1605073
Six types of User Defined Spatiotemporal Functions (UDSFs) and their syntax/usage examples
Spatiotemporal Exploration of Big Geotagged Tweets
Big geotagged tweets coupled with innovative spatial computing platform offer enormous opportunities for disaster management by examining the physical infrastructure (e.g., road damage), environment (e.g., flood extent), and nature-human interaction (e.g., evacuation) from spatial, temporal, and social dimensions.
This tool is used to support exploratory analysis of billions of tweets in near real-time. Currently, the public version of the tool is limited to query a time period of less than a month and returns up to 5,000 tweets.
High Resolution Population Grid in CONUS from Microsoft Building Footprints
This tool shows a new 100m population grid in CONUS, disaggregated from ACS 5-year estimates (2013-2017) using 125 million building footprints released by Microsoft. Land use dataset from OSM, a crowdsourced platform, was applied to trim raw footprints that are not likely residential. Layers derived from trimmed footprint statistics were designed and considered as weighting scenarios for dasymatric method, which was further applied to disaggregate ACS census tract estimates into 100m population grid.
Tool link: http://arcg.is/19S4qK
Huang X., Wang C., Li Z., High-Resolution Population Grid in the CONUS using Microsoft Building Footprints: a feasibility study, in Proceedings of the 3rd ACM SIGSPATIAL International Workshop on Geospatial Humanities, November 5, Chicago, Illinois, USA
A visual-textual fused CNN architecture for identifying disaster related tweets
Two convolutional neural networks (CNNs), the Inception-V3 CNN and word embedded CNN, are applied to extract visual and textual features, respectively, from social media posts. This implementation is based on Python coding environment. A high-level neural network API, Keras (https://keras.io/), is used to enable fast experimentation. The code requires python 3.6 compiler and the installation of Keras with TensorFlow backend. The authors recommend the usage of GPU-supported TensorFlow to reduce the training time.
Huang, X., Li, Z., Wang, C., & Ning, H. (2019). Identifying disaster related social media for rapid response: a visual-textual fused CNN architecture. International Journal of Digital Earth, 1-23.
Twitter Census: Mining and mapping geotagged tweets to reveal human dynamics
The study of population stocks and human movements has historically been severely limited by the absence of reliable data or the temporal sparsity of the available data. Using geospatial digital trace data, the study of population movements can be much more precisely and dynamically measured. Our research seeks to develop a near real-time Twitter census that gives a more temporally granular picture of local and non-local population at the county level.
Martin Y., Li Z., Ge Y., Towards real-time population estimates: introducing Twitter daily estimates of residents and non-residents at the county level, Annals of the American Association of Geographers (under review)