How to Deal with Class Imbalance, Detection Thresholds and Species Co-Occurrence in Data-Driven Detection of Harmful Algal Blooms from Satellite Data
How to Deal with Class Imbalance, Detection Thresholds and Species Co-Occurrence in Data-Driven Detection of Harmful Algal Blooms from Satellite Data
Conor McGlinchey1, Mortimer Werther2, Jesus Torres Palenzuela3, Luis Gonzalez Vilas3, Maximo Frangopulos4, Gemita Pizarro5, Yolanda Pazos6, Shenglei Wang7, Matthew Blake8, Junsheng Li7, Andrew N. Tyler1, Violeta Slabakova9, Nataliya Slabakova10, Nina Dzembekova11 and Evagelis Spyrakos1, (1)University of Stirling, Earth and Planetary Observation Sciences (EPOS), Department of Biological and Environmental Sciences, Stirling, United Kingdom, (2)EAWAG Swiss Federal Institute of Aquatic Science and Technology, Duebendorf, Switzerland, (3)University of Vigo, Ourense, Spain, (4)University of Magallanes, Punta Arenas, Chile, (5)Fisheries Development Institute (IFOP), Punta Arenas, Chile, (6)Instituto Tecnolóxico para o Control do Medio Mariño de Galicia, Pontevedra, Galicia, Spain, (7)Aerospace Information Research Institute Chinese Academy of Sciences, Beijing, China, (8)University of Stirling, Biological and Environmental Sciences, Stirling, United Kingdom, (9)Institute of Oceanology - Bulgarian Academy of Sciences, Department of Ocean Technologies, Varna, Bulgaria, (10)Institute of Oceanology - Bulgarian Academy of Sciences, Varna, Bulgaria, (11)Institute of Oceanology, Bulgarian Academy of Sciences, Varna, Bulgaria