Sourav Kumar Bhoi & Kalyan Kumar Jena
Department of Computer Science and Engineering, Parala Maharaja Engineering College (Govt.), Berhampur 761003, Odisha, India
Chittaranjan Mallick
Department of Basic Science (Mathematics), Parala Maharaja Engineering College (Govt.), Berhampur 761003, Odisha, India
Abstract
Ozone layer depletion is a very serious issue for human life and environment. It is depleting day by day due to the increase of pollution in the atmosphere. The polluted gases in the atmosphere are responsible for reducing the content of ozone (O ) gas in the atmosphere. Ozone 3 helps the environment in absorbing the ultraviolet rays emitting from sun. Therefore, it is necessary to monitor the ozone depletion to save the environment from ultraviolet rays. A framework is designed in this work to detect the ozone layer depletion at different zones using a machine learning based system architecture and Internet of Things (IoT) platform. Here, at a zone the IoT devices connected with ultraviolet ray detection sensor and other necessary sensors are responsible for sensing. The sensor readings are then forwarded to the cloud through the base station. The cloud uses a machine learning based approach to detect the ozone depletion from the readings received. This machine learning based approach is called as stacking classifier that is build using hybridization of some standard machine learning models. From the results it can be stated that the proposed stacking classifier outperforms other standard machine learning models by showing a classification accuracy (CA) of 94.80% in detecting the ozone depletion zone.
Keywords- Ozone Layer Depletion Monitoring, IoT, Machine Learning, Accuracy