Publication Date


Document Type


Committee Members

Fathi Amsaad, Ph.D. (Advisor); Kenneth Hopkinson, Ph.D. (Committee Member); Travis E. Doom, Ph.D. (Committee Member); Thomas Wischgoll, Ph.D. (Committee Member)

Degree Name

Master of Science (MS)


Smart cities have emerged to tackle many critical problems that can thwart the overwhelming urbanization process, such as traffic jams, environmental pollution, expensive health care, and increasing energy demand. This Master thesis proposes efficient and high-quality cloud-based machine-learning solutions for efficient and sustainable smart cities environment. Different supervised machine-learning models for air quality predication (AQP) in efficient and sustainable smart cities environment is developed. For that, ML-based techniques are implemented using cloud-based solutions. For example, regression and classification methods are implemented using distributed cloud computing to forecast air execution time and accuracy of the implemented ML solution. These models are utilized to forecast AQI (air quality index) with the help of pollutants in the air such as particulate matter (PM10), ozone (O3), nitric oxide (NO), particulate matter (PM2.5), nitric x-oxide (NOx), sulphur dioxide (SO2), benzene, toluene, carbon monoxide (CO), xylene, ammonia (NH3), and nitrogen dioxide (NO2). Various formals like mean squared error, R2 score, mean absolute error, and root mean squared error are utilized to validate or test the designed models. As classification models, we perform the support vector machine and random forest algorithms, which are measured using the accuracy score and confusion matrix. Execution times and accuracy of the developed models are computed and contrasted with the times for the cloud-based versions of these models. The results reveal that lasso regression outperforms linear regression among the regression algorithms. Also, out of the classification models, the random forest algorithm performs better than the support vector machine approach. In conclusion, our findings demonstrate that run-time is minimized when models are executed on a cloud platform compared to a desktop machine. Moreover, the accuracy of our models is maintained with reduced execution time.

Page Count


Department or Program

Department of Computer Science and Engineering

Year Degree Awarded