Predicting Pollution Level Using Random Forest: A Case Study of Marilao River in Bulacan Province, Philippines

• Published in: International Journal of Computing Sciences Research Vol. 3; no. 1; pp. 151 - 162
• Main Authors: Jayson M Victoriano, Manuel Luis C. Delos Santos, Albert A. Vinluan, Jennifer T. Carpio
• Format: Journal Article
• Language: English
• Published: STEP Academic Publisher 01.03.2019
• Subjects: data mining; machine learning; random forest; river pollution
• ISSN: 2546-0552
• DOI: 10.25147/ijcsr.2017.001.1.30

Purpose–This study aims to predict the pollution level that threatens the Marilao River, located in the province of Bulacan, Philippines. The inhabitants of this area are now being exposed to pollution. Contamination of this waterwaycomesfrom both formal and informal industries, such as a used lead-acid battery, open dumpsites metal refining, and other toxic metals. Using various water quality parameters like Dissolved Oxygen (DO), Potential of Hydrogen (pH), Biochemical Oxygen Demand(BOD) and Total Suspended Solids(TSS) were the basis for predicting the pollution level.Method–This study used the Data Mining technique based on the sample data collected from January of 2013 to November of 2017. These were used as a training data and test results to predict the river conditionwith itscorresponding pollution level classification indicated with the used of colorssuchas “Green” for “Normal”, “Yellow” for “Average”, “Orange” for “Polluted”and “Red” for “Highly Polluted”.The model got anaccuracy of 91.75% witha Kappavalue of 0.8115, interpreted as “Strong” in terms ofthe levelof agreement.Results–The predicted model using the Random Foresthavescored 91.75% in terms of correctly classified instances and were able to generate 0.8115 Kappa valueswhich indicatethat the model used to produce a stronglevel of agreement.Conclusion–From 2013 to 2017basedon the data sampling provided by the EnvironmentalManagement Bureau(EMB),an attached agency of the Department of Environment and Natural Resources (DENR)inthe Philippines mandated to protect and restore the environment,shows that the river is highly polluted. Several issues like, underestimationof the water parameterresults have been identified, issues which can be addressed by incorporating more observations to the training process and by validating the resulting model on the different training set.The discretion on decisions about the prediction ofthe model is attributed to DENR-EMBunit as they have more hands-on experience with regards to monitoring, restoring, protecting the environment.