Machine Learning Model for Message Queuing Telemetry Transport Data Analytics

Abstract

In the realm of MQTT data analytics, this project aims to accomplish the primary objective of developing and evaluating machine learning models for the classification of MQTT messages into distinct categories or message types. Two prominent classifiers, the Random Forest Classifier and K-Nearest Neighbors (KNN) Classifier, have been effectively implemented and rigorously assessed for their classification capabilities. The assessment primarily involves measuring the accuracy and precision of both models, employing comprehensive classification reports to ascertain their competence in categorizing MQTT data. The results of this evaluation reveal that both the Random Forest and KNN classifiers excel in effectively classifying MQTT messages with remarkable accuracy and precision. A standout feature of this project is the incorporation of confusion matrices to visually represent the performance of these models. These matrices offer a clear and intuitive depiction of the models' effectiveness by showcasing key metrics such as true positives, true negatives, false positives, and false negatives. This visual representation provides invaluable insights into the strengths and weaknesses of the classification models, aiding in a deeper understanding of their performance. In summary, this project sheds light on the potential of machine learning models, specifically Random Forest and KNN classifiers, in the context of MQTT data analytics. It highlights their proficiency in accurately categorizing MQTT messages, further enhanced by the visual clarity offered by confusion matrices. The findings presented in this study serve as a foundation for future research and applications in the domain of MQTT data analysis and classification