A Study on the Application of Bayesian Learning and Decision Trees IoT-Enabled System in Postharvest Storage

Abstract

AbstractLately, IoT has attracted much attention for its numerous potentials for applications in smart environments. In smart agriculture, IoT may be targeted toward optimum farm yield and minimization of postharvest waste and man-hour maximization. Food and Agriculture Organization (FAO) has regarded postharvest waste as a global menace and threat to food sustainability. IoT-enabled systems use various sensors and sensing technology for data gathering, processing, and storage. The quality of data in an IoT-enabled system is tied to the precision of the sensors and sensing technology deployed in the system. Innovative aspects of IoT system include the integration of intelligence where decisions can take place based on learning from the previous data acquired by the system or from related data sources. Different machine learning algorithms perform differently in terms of accuracy, speed, and efficiency during training and the actual prediction. In this chapter, the Bayesian learning and decision trees are presented in respect of their ability to entrench optimum intelligent prediction in IoT-enabled domain. Succinct elucidation of the potential application of an intelligent IoT-driven system is presented as a possible panacea to address some of the problems in food production cycle especially in postharvest storage and wastage.KeywordsBayesian learningDecision tressClassificationClusteringIoTSmart farming