Masters Thesis Defense “Multi-scale and multi-modal streaming data aggregation and processing for decision support during natural disasters” By Shruti Kar

Committee:  Drs. Krishnaprasad Thirunarayan, Advisor, Amit Sheth, and Valerie Shalin (Psychology)

ABSTRACT:

With the surge in digital information systems, there is a humongous amount of data from various sources that can be synthesized to produce intelligent systems.  These intelligent systems can provide functionality that caters the user with related and reliable information, preserving low-level details and surfacing essential information for better decision making.

Therefore, we employ multi-modal data (i.e., unstructured text, gazetteers, and imagery) for an aggregate level of analysis and location-centric demand/request matching in the context of disaster relief. After classifying the Need expressed in a tweet (the WHAT), we leverage OpenStreetMap to geolocate thatNeedon a computationally accessible map of the local terrain (the WHERE) populated with location features such as hospitals and housing. Further, our novel use of flood mapping based on satellite images of the affected area supports the elimination of candidate resources that are not accessible by road transportation. 

The resulting map-based visualization serves two levels of users.  A community-level user (first-responders) can visualize aggregated summary of a selected geographical area and an individual level user can identify current needs and available resources in their geographic proximity. Additionally, our pluggable modularized pipeline (DisasterRecord)readily allows for extending the functionality and the addition of any number of layers overlaid on top of the map. The integration of disaster-related tweets, imagery and pre-existing knowledge-base resources (gazetteers) reduce decision-making latency and enhance resiliency by assisting decision-makers and first responders for relief effort coordination.