Addressing Missing Data in Environmental Technologies: Economic and Environmental Optimizing Air Quality Monitoring with Random Forest and MissForest

Nengsih, Titin Agustin
Wardhana, Indrawata
M. Nazori Madjid, M. Nazori

Jurnal Riset Teknologi Pencegahan Pencemaran Industri
Vol. 16 No. 1 (2025): May

Air quality monitoring often encounters missing data issues due to technical glitches, equipment malfunctions, or other causes. This study employs PM2.5 and PM10 datasets from station 6, calculating multiple weighted probabilities for imputation. With missing values introduced at rates of 10, 40, and 70 percents through different amputation methods, the Random Forest and missForest techniques are utilized for imputation. Notably, missForest consistently outperforms Random Forest across all scenarios, yielding accuracy exceeding 96% even with high missing data levels. MissForest achieves remarkable accuracy above 96% for PM2.5 and PM10 across left, middle, and right multiple weight probabilities amputations. Overall, missForest attains the highest accuracy (over 97%) for Air Quality Index at lower and middle missing value proportions.