Comparing the Performance of Machine Learning and Deep Learning Algorithms in Wastewater Treatment Process
This study assessed the performance of single and modified algorithms based on machine learning and deep learning for wastewater treatment process. More specifically, this study adopted support vector machine (SVM), random forest (RF), and artificial neural network (ANN) for machine learning as well...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Korean Society of Environmental Engineers
2023-12-01
|
| Series: | 대한환경공학회지 |
| Subjects: | |
| Online Access: | http://www.jksee.or.kr/upload/pdf/KSEE-2023-45-12-587.pdf |
| _version_ | 1797311435843305472 |
|---|---|
| author | Jaeil Kim Hyo Sub Lee Jinuk Jang Yongtae Ahn Seo Jin Ki Hyun-Geoun Park |
| author_facet | Jaeil Kim Hyo Sub Lee Jinuk Jang Yongtae Ahn Seo Jin Ki Hyun-Geoun Park |
| author_sort | Jaeil Kim |
| collection | DOAJ |
| description | This study assessed the performance of single and modified algorithms based on machine learning and deep learning for wastewater treatment process. More specifically, this study adopted support vector machine (SVM), random forest (RF), and artificial neural network (ANN) for machine learning as well as long short-term memory (LSTM) for deep learning. The performance of these (single) algorithms were compared with that of modified ones processed through hyperparameter tuning, ensemble learning (only for machine learning), and multi-layer stacking (i.e., two layers of LSTM units). The daily effluent of wastewater treatment process observed between 2017 and 2022 in the Cheong-Ju National Industrial Complex was used as input to all tested algorithms, which was evaluated with respect to mean squared error. For the model performance evaluation, discharge and biochemical oxygen demand are selected as dependent variables out of nine measured parameters. Results showed that the performance of any machine learning algorithms was superior to their competitor LSTM. This is mainly attributed to a small amount of input data provided to the LSTM algorithm and unstable effluent wastewater characteristics. Meanwhile, hyperparameter tuning improved the performance of all tested algorithms. However, ensemble learning for machine learning and two-layer stacking for LSTM generally resulted in performance degradation as compared to that of single algorithms, regardless of dependent variables. Therefore, this calls for a careful design and evaluation of modified algorithms, specifically for model architecture and performance improvement processes. |
| first_indexed | 2024-03-08T01:58:40Z |
| format | Article |
| id | doaj.art-3c5add5bcb844118a19635c5292f302a |
| institution | Directory Open Access Journal |
| issn | 1225-5025 2383-7810 |
| language | English |
| last_indexed | 2024-03-08T01:58:40Z |
| publishDate | 2023-12-01 |
| publisher | Korean Society of Environmental Engineers |
| record_format | Article |
| series | 대한환경공학회지 |
| spelling | doaj.art-3c5add5bcb844118a19635c5292f302a2024-02-14T06:49:31ZengKorean Society of Environmental Engineers대한환경공학회지1225-50252383-78102023-12-01451258759310.4491/KSEE.2023.45.12.5874490Comparing the Performance of Machine Learning and Deep Learning Algorithms in Wastewater Treatment ProcessJaeil Kim0Hyo Sub Lee1Jinuk Jang2Yongtae Ahn3Seo Jin Ki4Hyun-Geoun Park5Department of Energy Engineering, Gyeongsang National University, Republic of KoreaResidual Agrochemical Assessment Division, National Institute of Agricultural Sciences, Rural Development Administration, Republic of KoreaDepartment of Energy System Engineering, Gyeongsang National University, Republic of KoreaDepartment of Energy Engineering, Gyeongsang National University, Republic of KoreaDepartment of Environmental Engineering, Gyeongsang National University, Republic of KoreaDepartment of Environmental Engineering, Gyeongsang National University, Republic of KoreaThis study assessed the performance of single and modified algorithms based on machine learning and deep learning for wastewater treatment process. More specifically, this study adopted support vector machine (SVM), random forest (RF), and artificial neural network (ANN) for machine learning as well as long short-term memory (LSTM) for deep learning. The performance of these (single) algorithms were compared with that of modified ones processed through hyperparameter tuning, ensemble learning (only for machine learning), and multi-layer stacking (i.e., two layers of LSTM units). The daily effluent of wastewater treatment process observed between 2017 and 2022 in the Cheong-Ju National Industrial Complex was used as input to all tested algorithms, which was evaluated with respect to mean squared error. For the model performance evaluation, discharge and biochemical oxygen demand are selected as dependent variables out of nine measured parameters. Results showed that the performance of any machine learning algorithms was superior to their competitor LSTM. This is mainly attributed to a small amount of input data provided to the LSTM algorithm and unstable effluent wastewater characteristics. Meanwhile, hyperparameter tuning improved the performance of all tested algorithms. However, ensemble learning for machine learning and two-layer stacking for LSTM generally resulted in performance degradation as compared to that of single algorithms, regardless of dependent variables. Therefore, this calls for a careful design and evaluation of modified algorithms, specifically for model architecture and performance improvement processes.http://www.jksee.or.kr/upload/pdf/KSEE-2023-45-12-587.pdfmachine learningdeep learninghyperparameter tuningensemble learningmulti-layer stacking |
| spellingShingle | Jaeil Kim Hyo Sub Lee Jinuk Jang Yongtae Ahn Seo Jin Ki Hyun-Geoun Park Comparing the Performance of Machine Learning and Deep Learning Algorithms in Wastewater Treatment Process 대한환경공학회지 machine learning deep learning hyperparameter tuning ensemble learning multi-layer stacking |
| title | Comparing the Performance of Machine Learning and Deep Learning Algorithms in Wastewater Treatment Process |
| title_full | Comparing the Performance of Machine Learning and Deep Learning Algorithms in Wastewater Treatment Process |
| title_fullStr | Comparing the Performance of Machine Learning and Deep Learning Algorithms in Wastewater Treatment Process |
| title_full_unstemmed | Comparing the Performance of Machine Learning and Deep Learning Algorithms in Wastewater Treatment Process |
| title_short | Comparing the Performance of Machine Learning and Deep Learning Algorithms in Wastewater Treatment Process |
| title_sort | comparing the performance of machine learning and deep learning algorithms in wastewater treatment process |
| topic | machine learning deep learning hyperparameter tuning ensemble learning multi-layer stacking |
| url | http://www.jksee.or.kr/upload/pdf/KSEE-2023-45-12-587.pdf |
| work_keys_str_mv | AT jaeilkim comparingtheperformanceofmachinelearninganddeeplearningalgorithmsinwastewatertreatmentprocess AT hyosublee comparingtheperformanceofmachinelearninganddeeplearningalgorithmsinwastewatertreatmentprocess AT jinukjang comparingtheperformanceofmachinelearninganddeeplearningalgorithmsinwastewatertreatmentprocess AT yongtaeahn comparingtheperformanceofmachinelearninganddeeplearningalgorithmsinwastewatertreatmentprocess AT seojinki comparingtheperformanceofmachinelearninganddeeplearningalgorithmsinwastewatertreatmentprocess AT hyungeounpark comparingtheperformanceofmachinelearninganddeeplearningalgorithmsinwastewatertreatmentprocess |