Developing a new approach for design support of subsurface constructed wetland using machine learning algorithms

Nguyen, Xuan Cuong; Nguyen, Thi Thanh Huyen; Le, Quyet Van; Cuong, Le Phuoc; Srivastav, Arun Lal; Pham, Quoc Bao; Nguyen, Phuong Minh; La, D. Duong; Rene, Eldon R.; Ngo, Huu Hao; Chang, Soon Woong; Trinh, Nguyen Duy

doi:10.1016/j.jenvman.2021.113868

Cited by 24 publications

(13 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…XGB has reputation of been the best performer ML algorithms in most microbiological regression studies compared with others 30 . Cubist has been demonstrated to outperformed partial least squares, RF, and MARS in predicting soil property including soil total nitrogen, organic carbon, total sulphur, exchangeable calcium clay; sand, and cation exchange capacity, and pH and RF, classification, and regression trees, SVM, and KNN predicting NH 4 –N and COD in subsurface constructed wetlands effluents 56 , 57 . In forecasting daily dissemination of COVID-19 vaccination, Cubist outperformed ENR, Gaussian Process, Slab (SPIKES), and Spikes ML algorithms 58 .…”

Section: Discussionmentioning

confidence: 99%

Machine learning-guided determination of Acinetobacter density in waterbodies receiving municipal and hospital wastewater effluents

Ekundayo

Adewoyin

Ijabadeniyi

et al. 2023

Sci Rep

View full text Add to dashboard Cite

A smart artificial intelligent system (SAIS) for Acinetobacter density (AD) enumeration in waterbodies represents an invaluable strategy for avoidance of repetitive, laborious, and time-consuming routines associated with its determination. This study aimed to predict AD in waterbodies using machine learning (ML). AD and physicochemical variables (PVs) data from three rivers monitored via standard protocols in a year-long study were fitted to 18 ML algorithms. The models’ performance was assayed using regression metrics. The average pH, EC, TDS, salinity, temperature, TSS, TBS, DO, BOD, and AD was 7.76 ± 0.02, 218.66 ± 4.76 µS/cm, 110.53 ± 2.36 mg/L, 0.10 ± 0.00 PSU, 17.29 ± 0.21 °C, 80.17 ± 5.09 mg/L, 87.51 ± 5.41 NTU, 8.82 ± 0.04 mg/L, 4.00 ± 0.10 mg/L, and 3.19 ± 0.03 log CFU/100 mL respectively. While the contributions of PVs differed in values, AD predicted value by XGB [3.1792 (1.1040–4.5828)] and Cubist [3.1736 (1.1012–4.5300)] outshined other algorithms. Also, XGB (MSE = 0.0059, RMSE = 0.0770; R2 = 0.9912; MAD = 0.0440) and Cubist (MSE = 0.0117, RMSE = 0.1081, R2 = 0.9827; MAD = 0.0437) ranked first and second respectively, in predicting AD. Temperature was the most important feature in predicting AD and ranked first by 10/18 ML-algorithms accounting for 43.00–83.30% mean dropout RMSE loss after 1000 permutations. The two models' partial dependence and residual diagnostics sensitivity revealed their efficient AD prognosticating accuracies in waterbodies. In conclusion, a fully developed XGB/Cubist/XGB-Cubist ensemble/web SAIS app for AD monitoring in waterbodies could be deployed to shorten turnaround time in deciding microbiological quality of waterbodies for irrigation and other purposes.

show abstract

Section: Discussionmentioning

confidence: 99%

Machine learning-guided determination of Acinetobacter density in waterbodies receiving municipal and hospital wastewater effluents

Ekundayo

Adewoyin

Ijabadeniyi

et al. 2023

Sci Rep

View full text Add to dashboard Cite

show abstract

“…This statement is in agreement with previous studies on Cubist's performance. 50,70,71 For example, Cubist was found to be the most accurate model among other ML algorithms for predicting pharmaceuticals onto biochars with an R 2 of 0.95. 49 The accuracy of Cubist has been also witnessed in the evaluation of soil organic matter contents with an R 2 of 0.88.…”

Section: Estimation Of Adsorptionmentioning

confidence: 99%

“…The selection was based on our initial pre-evaluation (Figure S2) as well as our experience that the Cubist is a simple ML model that could give satisfactory prediction performance. 49,50 The basic principle and development of the Cubist can be found in Text S3.…”

Section: Machine Learning Techniquementioning

confidence: 99%

Insight into the Adsorption of Nutrients from Water by Pyrogenic Carbonaceous Adsorbents Using a Bootstrap Method and Machine Learning

Nguyen

Hằng

et al. 2022

ACS EST Water

Self Cite

View full text Add to dashboard Cite

Herein, findings from 98 research papers [52 biochar (BC) and 46 activated carbon (AC) papers] were analyzed using a bootstrap method and a 95% confidence interval (CI) to provide insights into nutrient adsorption. The results indicated the solution temperature, pore volume, specific surface area, and pyrolysis temperature were correlated significantly to the maximum adsorption capacity (Q m ), achieving r values of 0.37, 0.23, 0.16, and 0.14, respectively. A statistically significant difference was witnessed for nitrate Q m between AC and BC with median values of 17.64 and 0.51 mg/g, respectively, and CIs of 11.01−27.56 and 0.39−0.82 mg/g, respectively. In contrast, the difference in Q m between AC and BC for phosphate and ammonium contained a high probability of chance. On average, engineered biochar achieved higher Q m values of ammonium and phosphate (12.13 and 24.73 mg/g, respectively) in comparison with that of the original biochar. However, there is a good probability of no difference in nitrate between the two, albeit greater nitrate adsorption on the modified biochar with a mean Q m of 3.03 mg/g. The tuned Cubist estimated equilibrium adsorption capacity with an R 2 of ∼0.90−0.91. The median and bootstrap 95% CI can be used as an average standard for designing adsorbate−adsorbent systems.

show abstract

“…The skilled AI-based models in geoengineering problems were used to tackle the difficulties in handling the big data [43] and provide physically meaningful relationships within geo-data [44][45][46]. Accordingly, applicability of the AI techniques in the form of artificial neural network, machine/deep learning, evolutionary algorithms, and hybrid structures in producing the predictive 3D subsurface models have been highlighted [46][47][48][49][50][51]. Due to characterized features in creating transferable solutions and learnability from high-level data attributes [52] the feasibility of AI techniques in geothermal modeling [53,54] and compared performance by field prospecting methods [55,56] have been notified in several studies dealing with predicting the location of hot spot structures [57][58][59][60], estimating the temperature distribution [61,62], and potential of geothermal production associated with geological data [63,64].…”

Section: Introductionmentioning

confidence: 99%

Generating 3D Geothermal Maps in Catalonia, Spain Using a Hybrid Adaptive Multitask Deep Learning Procedure

et al. 2022

View full text Add to dashboard Cite

Mapping the subsurface temperatures can efficiently lead to identifying the geothermal distribution heat flow and potential hot spots at different depths. In this paper, an advanced adaptive multitask deep learning procedure for 3D spatial mapping of the subsurface temperature was proposed. As a result, predictive 3D spatial subsurface temperatures at different depths were successfully generated using geolocation of 494 exploratory boreholes data in Catalonia (Spain). To increase the accuracy of the achieved results, hybridization with a new modified firefly algorithm was carried out. Subsequently, uncertainty analysis using a novel automated ensemble deep learning approach for the predicted temperatures and generated spatial 3D maps were executed. Comparing the accuracy performances in terms of correct classification rate (CCR) and the area under the precision–recall curves for validation and whole datasets with at least 4.93% and 2.76% improvement indicated for superiority of the hybridized model. According to the results, the efficiency of the proposed hybrid multitask deep learning in 3D geothermal characterization to enhance the understanding and predictability of subsurface spatial distribution of temperatures is inferred. This implies that the applicability and cost effectiveness of the adaptive procedure in producing 3D high resolution depth dependent temperatures can lead to locate prospective geothermally hotspot active regions.

show abstract

Developing a new approach for design support of subsurface constructed wetland using machine learning algorithms

Cited by 24 publications

References 44 publications

Machine learning-guided determination of Acinetobacter density in waterbodies receiving municipal and hospital wastewater effluents

Machine learning-guided determination of Acinetobacter density in waterbodies receiving municipal and hospital wastewater effluents

Insight into the Adsorption of Nutrients from Water by Pyrogenic Carbonaceous Adsorbents Using a Bootstrap Method and Machine Learning

Generating 3D Geothermal Maps in Catalonia, Spain Using a Hybrid Adaptive Multitask Deep Learning Procedure

Contact Info

Product

Resources

About