There are few things that truly scare me. Snakes might be one or the surprise of a snake appearing, at least. The other is sinkholes. The unexpected opening of the ground swallowing up people, cars and houses feeds a deep-seeded fear in me that some therapist will have to work on some day. For now, I live with the worry – especially in a state that seems to have them quite frequently and with a husband who loves to send me sinkhole videos. To hear that research is taking place that is utilizing machine learning to identify and predict sinkhole activity, gives me a little hope. This interesting topic came to us from the University of Kentucky (UK) in their article, “KGS, UK Computer Science Collaborate on Sinkhole Machine-Learning Research.”
Since 2015, Junfeng Zhu, a hydrogeologist with the Kentucky Geological Survey, a research center within the University of Kentucky, has been investigating the use of machine learning to identify Kentucky sinkholes in aerial LiDAR (light detection and ranging) data.
Machine learning uses an array of tools that help a computer learn and improve its abilities at a task without being programmed specifically for the task. Because machine learning can be applied to just about anything with data, this assignment also serves as a good teaching tool to the UK students.
Melody K. Smith
Sponsored by Data Harmony, a unit of Access Innovations, the world leader in indexing and making content findable.
