Bhargava Receives Best Poster at the M.S. Data Science Poster Fair
Data Science student Akhil Bhargava received the best poster award at this year’s Data Science M.S. Poster Fair. He was honored for his poster, “Fully Convolutional Network in Performing Dense Pixelwise Classification for Satellite Data.”
His research focuses on using a neural network to determine whether a picture contains water or land pixels. The idea is that if an algorithm can be created to perfectly classify each pixel, analysts will be able to use this algorithm to create accurate water/land maps for the entire globe at any time that has photographic evidence. For example, climate scientists could visualize and compare data of the Dead Sea and learn greater detail into how it has been affected by climate change.
Bhargava hopes his capstone project will allow climate scientists further understand climate change and the water dynamics of Earth. Once his machine learning algorithm is perfected, it may also be possible to aid in predicting the location and time of a natural disaster such as a hurricane.
For this project, Bhargava focused on applying machine learning models to satellite data, but a lot of his work is focused in the health care field. He has used the power of data science to help hospitals assess the risk of a patient and physicians decide what drug regimen cancer patients should take.
The Data Science Poster Fair highlighted capstone projects being carried out by M.S. students as part of their degree. Topics were wide ranging, including social web, biology, medicine, statistical modeling, and scalable algorithms.
Visit datascience.umn.edu for more information on the program.