An IoT-Based Efficient Water Quality Prediction System for Aquaponics Farming

Abstract

An IoT-based smart water monitoring system is of prime importance to control the threats related with aquaponics farming. Thus, it helps to provide a remarkable boost to improve the yield and productivity. Water quality directly impacts growth rates, feed efficiency, and the overall health of the fish, plants, and bacteria. The major issue in the aquaponics farming business is the lack of knowledge about species selection based on the water quality parameters. The proposed system provides a farming prediction for cold water, warm water fish, plants, and bacteria to improve the aquaponics farming business. Initially, the proposed system collects data using IoT sensors. After that, data cleaning is performed by removing missing values and outliers. Next, features correlated with the sensed data are obtained, and unwanted features are removed. Then, we propose a novel M-SMOTE algorithm to address the imbalanced class problem. Finally, the proposed approach employs the multi-model classification for the aquaponic ecosystem. The proposed method utilizes the mechanism of optimal prediction based on voting to evaluate the performance of six classifiers. The proposed method chooses the XGBoost and the random forest (are the best classifiers) based on the voting principle. The experimental results reveal that the proposed method’s results offer a new state-of-the-art aquaponics farming prediction model with an accuracy of 99.13%.