A dataset for internet of things based fish farm monitoring and notification system

Abstract

Water quality depends on many factors. Some of them are essential for maintaining the minimum sustainability of water. Because of the great dependence of fishes on the condition of the aquatic environment, the water quality can directly affect their activity. Therefore monitoring water quality is a very important issue to consider, especially in the fish farming industry. In this paper a digital fish farm monitoring system is introduced and a collection of experimental data of water quality monitoring was presented, which were directly collected from a fish pond. As the quality factor of water affects its aquatic life form sustainability, therefore the quality factors of the water were measured using digital sensors. Temperature, pH factor and Turbidity were selected as the basic quality factors to measure. The dataset contains data recorded from two different water levels to analyze the aquatic environment more efficiently. Each level has 9623 sets of data of the selected parameters. Collection was continued all day long for several days. Later collected sensor data were analyzed as short period time series to find its properties. Machine Learning regression method was used to predict near future conditions. Moreover data were processed to find any repetitive patterns in its properties. This dataset represents the exact condition of the environment of the fish pond. Therefore it can be used to develop a system to monitor fish farms digitally. Using these data in machine learning, predicting the future is possible for advance monitoring of a fish farm. The dataset is available in Mendeley Data[1].