This document presents a research paper that proposes an inference model for environment detection using IoT and support vector machines (SVM). The model is intended to reduce the large amount of data generated by IoT devices and sent to the cloud. It does this by using SVM machine learning to analyze sensor data from IoT devices and only send data categorized as "harmful" to the cloud. The harmful data can then be viewed on a smartphone by the relevant authorities, allowing them to take steps to improve environmental conditions. The document reviews related literature on IoT and discusses the proposed system architecture, SVM algorithm, and results analyzing sensor data on a desktop. It is concluded that the model develops a unique approach combining IoT, machine learning, cloud