A DATASET AT A GLANCE...

Room Occupancy Estimation Dataset

The abruptly changing weather into an extremely hot or extremely cold condition, the unexpected rising levels of air pollution and the arrival of new life-threatening diseases have imposed a need for what is called the Heating, Ventilation and Air-Conditioning (HVAC) systems.  The HVAC systems are highly known for its ability to filter the outdoor air and to provide a moderately humid, thermal-comfort and fresh air environment to the users within a building. However, this HVAC benefit also adds a cost, in terms of energy utilization. Consider the scenario, where the HVAC system is consuming enormous energy to deliver heating or ventilation or air-conditioning to a few or no occupants in the room. Optimal and intelligent usage of HVAC becomes mandate at this instant to conserve energy. Estimating the number of occupants in a room can primarily support to meet this HVAC usage requirement.  Are you a researcher interested in room occupancy estimation for optimal energy utilization of HVAC devices? Then, you can avail the dataset publicly from the UCI Machine Learning Repository (http://archive.ics.uci.edu/ml/datasets/Room+Occupancy+Estimation#), but with a citation request as “Adarsh Pal Singh, Vivek Jain, Sachin Chaudhari, Frank Alexander Kraemer, Stefan Werner and Vishal Garg, “Machine Learning-Based Occupancy Estimation Using Multivariate Sensor Nodes”, in 2018 IEEE Globecom Workshops (GC Wkshps), 2018, DOI: 10.1109/GLOCOMW.2018.8644432”. This dataset is a multivariate time-series dataset with 10129 instances, which are acquired through five non-intrusive sensors (CO2, temperature, sound, light and digital Passive Infrared (PIR)). The data collection involved monitoring the occupancy of a room by 0 to 3 individuals for about 4 days in a controlled manner, with no HVAC utilization, to aid the researchers in performing classification tasks. Any researcher intending to perform research on energy-aware Internet of Things (IoT)-based applications in civil or medical field can make use of this dataset.  

Image courtesy: www.freepik.com