Air is one of the essential elements in human life besides water and soil. However, currently the air quality in Indonesia is getting worse. Therefore, an air quality classification system using Random Forest algorithm based on CO and CO2 levels using IoT-based MQ7 and MQ13 sensors is needed as a smart solution. The workflow of this system begins with detecting air quality using the MQ-7 sensor for CO gas and the MQ-135 sensor for CO2 gas. Then, the classification process is carried out with the Machine Learning Random Forest algorithm by utilizing a number of training data that has been stored in the program to classify the gas sensor detection results into three types of classes, namely “Good”, “Bad”, or “Toxic”. The final output of this system is a website display that can be accessed on a PC/Laptop monitor in real time. From the results of the Random Forest machine learning algorithm classification testing process, 1 unsuitable data was found from a total of 100 trials that have been carried out. Therefore, the Random Forest machine learning algorithm can be said to be successful in detecting air levels in the surrounding environment well because it provides an accuracy value of 99%.
Copyrights © 2024