Investigations of Naive Bayes, Random Forest and Decision Tree Algorithm for building aeroponics system

Abstract

Urban farming has becoming popular and most adopted recently as society has become more conscious about the food quality they consume and conservation of resources. The Internet of Things (IoT) and machine learning algorithms are employed to build and implement an automated aeroponics system. This system contains Arduino as a main processing unit and uses Node MCU for cloud storage.  It gets inputs from sensors like ultrasonic sensor, light sensor, humidity sensor, temperature sensor and exhibit the output in a LCD unit.  In response to the inputs, water pump, cooler fan and light acts automatically. The data are stored in cloud as a record for future use.  Using machine learning algorithms Naive Bayes, Random Forest and Decision Tree the datasets from cloud are trained and tested to know the crop suitable for the given condition. According to the results, the system has the potential to save more labour and reducing water consumption than conventional methods.