As the proportion of time spent by humans in indoor environment increases, it becomes challenging to maintain good air quality for healthy and productive life. The need to develop a context aware, reliable system capable of providing real time information and alerts on indoor air quality is addressed in this article. The proposed Internet-of-Things (IoT) system serves to collect data, predict ventilation states, and provide alerts and recommendations to the end user. A novel method for determination of ventilation states using three indoor pollutants PM <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2.5</sub> , PM <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">10</sub> , and CO is proposed. Multilevel logistic regression is first used to define indoor ventilation states using ventilation rate which is calculated with the help of indoor CO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> concentration. K-NN classification technique then predicts indoor ventilation state with the help of three input attributes, PM <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2.5</sub> , PM <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">10</sub> , and CO. Context-aware information about indoor environment and current ventilation state is conveyed to the end-user in form of an alert, through a smartphone application. The system is found to determine the poor ventilation state with accuracy, precision, recall and F1 score values of 94.34%, 0.91, 0.88, and 0.89, respectively.