In Partial Fulfillment of the Requirements for the Degree of Master of Science
will defend her thesis
Design and Implementation of a DAG-based Water Level Prediction Approach
Flood is a hazardous disaster that threats the United States and its territories and also millions of people every year. In this context, having good approaches for predicting flood threats in the future may help humans to reduce property losses and to prepare in advance. There has been a lot of research centering on how to reduce the flood damage to human life and properties including research that centers on developing early warning systems that simulate the dynamics of water flows by using hydrological models; these models are using complex mathematical and physics equations to forecast water levels.. Moreover, many cities developed sensor-based flood monitoring systems that collect flood-related data, such as amount of rainfall and water level along streams. The goal of this thesis is to take advantage of these data and develop a data driven water level prediction approach that interpolates the past into the future. In particular, a DAG-based multi-target prediction (DBMTP) approach is proposed for this purpose. DBMTP not only uses historical rainfall and water level information to predict water levels in the future, but also learn dependencies between neighboring sensors. In detail, this approach chains and feeds predecessors’ water level predictions into the prediction models for downstream locations. We evaluate the performance of this approach in two case studies involving watersheds in Harris County. The experiments show that DBMTP improves prediction accuracy in some cases over traditional regression approaches.
Date: Monday, November 27, 2017
Time: 3:00 PM
Place: PGH 550
Advisor: Dr. Christoph F. Eick, Dr. Shishir Shah, Dr. Jian Chen
Faculty, students, and the general public are invited.