Abstract: With the arrival of Web of Thing (IoT), and ubiquitous data collected each moment by either portable (good phone) or mounted (sensor) gadgets, it is crucial to achieve insights and meaningful info from the sensor data in context-conscious computing environments. Many researches have been applied by scientists in several fields, to analyze such knowledge for the aim of safety, power effectivity, constructing reliability and good environments. One examine, that many researchers are keen on, is to utilize Machine Studying techniques for occupancy detection where the aforementioned sensors collect info concerning the atmosphere. This paper supplies a solution to detect occupancy utilizing sensor data through the use of and testing several variables. Additionally we show the evaluation performed over the gathered information using Machine Studying and pattern recognition mechanisms is possible to determine the occupancy of indoor environments. Seven well-known algorithms in Machine Studying, namely as Determination Tree, pastelink.net Random Forest, Gradient Boosting Machine, Logistic Regression, Naive Bayes, Kernelized SVM and Ok-Nearest Neighbors are examined and in contrast on this research.