User Tools

Site Tools


Action unknown: copypageplugin__copy
projects:year9:9a.001.ul

9a.001.UL - Solving Data Integration and Inter-data Relationships from a Wide Variety of Data Sources

Project - Summary

OBJECTIVES

  • Identify and implement latest Deep Learning and data mining approaches to establishing ontology-based data integration for variables with different spatial and temporal resolution’s (e.g., raster vs. vector, hourly model outputs vs. satellite overpass).
  • Identify and implement methods to maximize AI model learning efficiency through identifying the significant variables, inter-data relationships, and the elimination of redundant information.

Project - Team

Team Member Role Email Phone Number Academic Site/IAB
Emad Habib PI emad.habib@louisiana.edu (337) 482-6513 UL Lafayette
Mohamed ElSaadani Co-PI mohamed.elsaadani@louisiana.edu (337) 789-9828 UL Lafayette
Magdy Bayoumi Co-PI mab0778@louisiana.edu (337) 482-5365 UL Lafayette
Ahmed Abdelhameed Student ahmed.abdelhameed1@Louisiana.edu N/A UL Lafayette
Sumit Shah Project Mentor sumit.shah@cgifederal.com (202) 309-8790 CGI
Funded by: CGI

Project - Novelty of Approach

  • Efficiently quantify the added-value of including a given available dataset in decreasing the models’ learning time and increasing the overall performance, this is needed to avoid redundancy and un-needed information.
  • Due to the large amounts of the available data, a single data source or a combination of data sources with similar features (e.g., raster only data) are often used to derive the DL models. We will over come this issue by using innovative data merging approaches to get the best out of the available data.

Project - Deliverables

Deliverables
1 Develop methods to fuse data with different spatial and temporal resolutions and eliminate un-needed information.
2 Produce and document a systematic evaluation methodology that quantifies the added-value of each input variable in terms of its contribution to the DL model computational efficiency, the significance of the information it provides, and the overall quality of the DL model output.

Project - Benefits to IAB

  • The methods and approaches produced and documented throughout project’s span are generic and can be applied for a multitude of applications (e.g., flood hazard, road hazard, and vulnerability analysis of areas at risk from natural and anthropogenic induced disasters).
  • The goals of the project directly align with the mission of our potential IAB member, CGI Federal, to protect America’s assets.

Project - Documents

projects/year9/9a.001.ul.txt · Last modified: 2022/02/08 08:28 by sally.johnson