DeepData: Machine learning in the marine ecosystems

Abstract

This work presents a web-based machine learning tool to facilitate biologists work of building species distribution models. The DeepData web-based tool takes into account the way biologists deal with species distribution models nowadays. Biologists mostly use probabilistic algorithms, such as maximum entropy, generalized linear models and generalized addictive models. We propose the use of machine learning algorithms, such as classification and regression trees, random forest and support vector machines. Other steps involved in the species distribution models, such as data preparation and model evaluation, are also discussed. A concrete explanation of the use of the web-based tool is made, as well as the details of implementation and evaluation.