One Concern's mission is to combine weather science with machine learning in order to predict and prepare for natural disasters and catastrophes.
As a data analyst, I annotated large amounts of geographical data in order to improve the accuracy of the product, a data model which concisely visualizes disaster risk in Earth's ever-changing climate.
My location of focus was Japan, where I used QGIS to assess maps, rivers, water gauges, lakes, and dams and rectify discrepancies generated by the model.
I learned a great deal about how machine learning can be practically applied to solve real-world problems, as well as the geographical factors which make can make areas "risky" in the event of a natural disaster.
Assisted supervisor with graduate-level research. Used Deeplabv3+ deep learning model to analyze MRI scans in order to diagnose patients with Temporomandibular joint dysfunction (TMJ). Helped to tune model hyperparameters and gained lots of knowledge and working experience with ML/Deep Learning.
