The Biomonitoring of Ireland's River Network Using a 1D Convolution Neural Network

Cleanliness and continuity of water resources are crucial for humans, the environment, and animals. However, with developing technology and increasing industrialization, water resources are decreasing and getting polluted daily. To prevent this pollution, the regular monitoring of water resources is essential. New and innovative technology to monitor water quality and sources are also critical in agricultural processes for increased yields, efficiencies, and product quality. This study was carried out with the aim of making a classification of water pollution based on the population of macroinvertebrates in Irish rivers. The population of macroinvertebrates is a good indicator to define the pollution level in the rivers. The study determines whether the river waters were polluted using an advanced classifier named 1D Convolution neural network with a public data set. It has been proven by experimental studies that the proposed framework produces better results than traditional machine learning classifiers, including decision trees, support vector machines, k-neighbors algorithms, naive Bayes, and neural networks. The obtained findings are also supported statistically. Our study is an example of a more effective use of developing artificial intelligence techniques for the benefit of humanity.