Data Mining

Predicting Occupancy in
an office

The data set used for this project was presented with an article in science direct, the link to it is here: https://www.sciencedirect.com/science/article/pii/S0378778815304357.

Details regarding the collection of the data indicate that a microcontroller was placed inside an office in Belgium in February (during winter) to collect environmental data. A camera was also used to capture occupancy in the office.

We will use this data set to make a model capable of predicting whether the office is occupied or not. The data set contains 8,143 rows of data collected over a 24-hour period in an office setting. The features included in this data set are id, Temperature, Humidity, Light, CO2, HumidityRatio, and Occupied.

The target variable for this project is Occupied. The possible values (prior to cleaning) for the target variable, Occupied, are N, no, Y, yes, and blank. After cleaning the possible values for Occupied are yes and no.

The remaining features, such as light, humidity, and temperature can be used to predict the Occupied feature.