Review of Inference Model for Environment Detection
Miss. Dhawane Shrutika Bhausaheb · International Journal for Research in Applied Science and Engineering Technology · 2019
The Internet of Things (IoTs) refers to the interconnection of billions of smart devices. The steadily increasing number of IoT devices with heterogeneous characteristics requires that future networks evolve to provide a new architecture to cope with the expected increase in data generation. The IoT paradigm has to be connected to the Cloud to increase the reach the scale at which it can be implemented. Today due to introduction of 4G the devices connected to the internet are growing rapidly which increases the scope of useful hardware connecting to internet. The IoT can be applied to many day today useful applications such as Home automation, Healthcare and Environment. But as the hardware generates a lot of data per second to the cloud, it can cause a headache to maintain and analyze such a large data on the cloud. So to solve this problem we have thought of developing a application which will analyze the data coming from the connected devices and with the help of Machine learning's Support vector Machine (SVM) algorithm we can distinguish the data in two categories harmful and safe. After recognizing the safe and unsafe environments using MQ2 and MQ7 gas sensors we are going to send only harmful data to the cloud. This will decrease the amount of data going on the cloud. We will also send an alert which will help the concerned authorities to take the steps necessary for reducing the pollution in the environment. The concerned authority can view the data on his mobile device remotely with the help of cloud.