Algorithms today dominate the digital ecosystem, since their “guidance” allows users to be given personalized recommendations, based on their search results, or on their data. But not only these algorithms that help us decide on a movie Netflix either amazon has that function, since they could also contribute to the improvement of satellites.
The findings are published in International Journal of Applied Earth Observation and Geoinformation and show how with these algorithms, the researchers were able to decipher areas free of cloud cover and blind spots for satellite transmission.
By adapting a recommendation algorithm first built for Netflix, Ruo-Qian (Roger) Wang, an assistant professor of civil and environmental engineering at Rutgers School of Engineering, created a system that is more accurate and faster at predicting landscapes covered in debris. clouds in coastal areas than conventional data filling tools.
This Netflix algorithm is called Funk-SVD, and it was created by Simon Funk to plot consumer reviews in a matrix.
“E-service platforms like Alibaba and Amazon use recommendation systems, which leverage large data sets to provide personalized product recommendations to help customers make decisions,” Wang said. “Interestingly, the way that Recommendation systems process the data is not unlike the process for predicting cloud-obscured coastal landscapes.”
Cloud fill algorithms used in remote sensing measure continuous data, such as water temperature, color, and algae content, to make predictions of what is hidden.
It’s a similar process for cloud filling: each coordinate on a map is represented by a pixel on a photograph, and that pixel can be water or land, with clouds representing unrecorded data. Wang’s adaptation of Funk-SVD makes educated guesses about what is below the clouds based on other data points.