Abstract
Air quality is degrading over time due to industrial and technological growth. Indoor air quality can be 5-10% poor than that of outdoor. The classroom is one of the most susceptible locations of indoor environments where these pollutants affect the children physically as well as cognitively. Dealing with such scenarios requires knowledge of the concentration of pollutants in each classroom. There are air quality monitoring devices available in the market, which help keep track of the air quality inside a room. However, the available devices are costly, and their setup requires effort, time, and other resources. To address these problems, we first develop machine learning models that can calculate the air quality inside a room using readily available features and data from a simple thermo-hygrometer placed inside the room. Then we developed a mobile application with an easy-to-use graphical interface that helps the user feasibly get the air quality information of each room in a building. These resulted in an end-to-end system that provides a user the knowledge of the air pollutants in an indoor environment. A thorough analysis of the system by deploying it in a local area comprising multiple classrooms inside an educational institution indicates that the system works with an accuracy of 78%. Additionally, we have conducted a Post-Study System User Questionnaire (PSSUQ) survey with the help of 53 users. We have achieved an overall score of 1.75 from the survey indicating the high usability of the system.

This publication has 17 references indexed in Scilit: