본 논문에서는 인공 신경망의 일종인 Extreme Learning Machine의 학습 알고리즘을 기반으로 하여 노이즈에 강한 특성을 보이는 퍼지 집합 이론을 이용한 새로운 패턴 분류기를 제안 한다. 기존 인공 신경망에 비해 학습속도가 매우 빠르며, 모델의 일반화 성능이 우수하다고 알려진 Extreme Learning Machine의 학습 알고리즘을 퍼지 패턴 분류기에 적용하여 퍼지 패턴 분류기의 학습 속도와 패턴 분류 일반화 성능을 개선 한다. 제안된 퍼지 패턴 분류기의 학습 속도와 일반화 성능을 평가하기 위하여, 다양한 머신 러닝 데이터 집합을 사용한다.

Extreme Learning Machine 기반 퍼지 패턴 분류기 설계

Drinking water quality is one of the greatest factors affecting human health. However, drinking water quality in many countries, especially in developing countries is not desirable and poor drinking water quality has induced many waterborne diseases. This special issue of Exposure and Health was edited to gain a better understanding of the impacts of drinking water quality on public health so that proper actions can be taken to improve the drinking water quality conditions in many countries. This editorial introduction reviewed some latest research on drinking water quality and public health, summarized briefly the main points of each contribution in this issue, and then some research fields/directions were proposed to boost further scientific research in drinking water quality and public health. The papers in this issue are interesting and cover many aspects of this research topic, and will be meaningful for the sustainable drinking water quality protection.

Drinking Water Quality and Public Health

Artificial intelligence for sustainability: Challenges, opportunities, and a research agenda

https://eartharxiv.org/repository/object/149/download/296/

A comprehensive review of deep learning applications in hydrology and water resources

Hydrological droughts are characterized based on their duration, severity, and magnitude. Among the most critical factors, precipitation, evapotranspiration, and runoff  are essential in modeling t...

https://www.tandfonline.com/doi/pdf/10.1080/19942060.2020.1715844

Predicting Standardized Streamflow index for hydrological drought using machine learning models

This article explores the suitability of a long short-term memory recurrent neural network (LSTM-RNN) and artificial intelligence (AI) method for low-flow time series forecasting. The long short-term memory works on the sequential framework which considers all of the predecessor data. This forecasting method used daily discharged data collected from the Basantapur gauging station located on the Mahanadi River basin, India. Different metrics [root-mean-square error (RMSE), Nash–Sutcliffe efficiency (ENS), correlation coefficient (R) and mean absolute error] were selected to assess the performance of the model. Additionally, recurrent neural network (RNN) model is also used to compare the adaptability of LSTM-RNN over RNN and naive method. The results conclude that the LSTM-RNN model (R = 0.943, ENS = 0.878, RMSE = 0.487) outperformed RNN model (R = 0.935, ENS = 0.843, RMSE = 0.516) and naive method (R = 0.866, ENS = 0.704, RMSE = 0.793). The finding of this research concludes that LSTM-RNN can be used as new reliable AI technique for low-flow forecasting.

Long short term memory (LSTM) recurrent neural network for low flow hydrological time series forecasting

In the hydrological cycle, rainfall is a major component and plays a vital role in planning and managing water resources. In this study, new generation deep learning models, recurrent neural networ...

Forecasting monthly precipitation using sequential modelling

Nitrate (
 $$ {\mathrm{NO}}_3^{-} $$
 ) pollution is a global concern as it affects the whole ecosystem: human, livestock, economy, and environment. The elevated levels of nitrate in groundwater can directly pose risks to population. A total of 156 representative groundwater samples were collected from groundwater sources such as hand pumps and bore wells across the study area. To identify the source of nitrate with its associated attributes, multivariate statistical methods (factor analysis (FA), sparse principal component analysis (SPCA)) were used in this study. In addition, empirical Bayesian kriging (EBK) modeling was used to predict the nitrate at ungauged locations of the study area. From the analysis of results, it was found that 5% of the groundwater samples exceeded the acceptable limit (50 mg l−1) of nitrate as specified by the World Health Organization (WHO). The first principal component (PC) indicated by the SPCA was salinity factor, which was significantly contributed by electrical conductivity followed by sulfate. The fourth PC represented the nitrate as a factor and positive loading of nitrate was strongly associated with chloride, sulfate, and calcium. The associated loading of nitrate with water quality attributes indicated that elevated level of nitrate in groundwater may be due to external sources that came through anthropogenic activity. A similar conclusion was drawn from factor analysis as well, indicating that SPCA can be applied as a new method for groundwater geochemistry. Hazard index calculations showed that infants of the study region were at a higher risk compared to the adults and children.

Source characterization and human health risk assessment of nitrate in groundwater of middle Gangetic Plain, India

Hydrologic time series modeling using historical records plays a crucial role in forecasting different hydrological processes. The objective of this study is to analyze the suitability of Support Vector Regression (SVR) for modeling monthly low flows time series for three stations in Mahanadi river basin, India. The ‘low flow’ threshold was taken as the Q75 discharge, i.e., the flow is equal to or surpassed for the duration of 75% of the observation period which was obtained from the daily discharge data. The potential applicability of SVR model is assessed with two different framework models (ANN-ELM, GPR) based on various statistical measures (r2, RMSE, MAE, Nash-Sutcliffe coefficient, objective function (OBJ), Scatter Index (SI) and BIAS). The model selection was based on lowest OBJ value for each station amongst three models (SVR, ANN-ELM, GPR). The SVR model was trained using the Radial Basis Function (RBF). Using the same inputs, the other two models (ANN-ELM and GPR) was also tested. From results, among all the stations, the SVR outperformed GPR and ANN-ELM with lowest OBJ value for the three stations a (1.378, 1.202, 1.570). In addition, the accuracy of the three models were checked using mean forecasting error which were (0.474, 0.421, 0.509) for SVR, (0.507, 0.489 0.500) for GPR and (0.564, 0.603, 0.772) ANN-ELM for the three stations. The results confirm that SVR can be used satisfactorily for modeling monthly low flows in the Mahanadi river basin, India. Hence, the SVR model could be employed as a new reliable and accurate data intelligent approach for predicting the ‘low flow’ (Q75 discharge) based on precedent data in water resources and its allied field.

Application of Support Vector Regression for Modeling Low Flow Time Series

The occurrence of uranium in groundwater is of particular interest due to its toxicological and radiological properties. It has been considered as a relevant contaminant for drinking water even at a low concentration. Uranium is a ubiquitously occurring radionuclide in the environment. Four hundred and fifty-six (456) groundwater samples from different locations of five districts of South Bihar (SB) were collected and concentrations of uranium (U) were analyzed using a light-emitting diode (LED) fluorimetric technique. Uranium concentrations in groundwater samples varied from 0.1 µg l−1 to 238.2 µg l−1 with an average value of 12.3 µg l−1 in five districts of Bihar in the mid-eastern Gangetic plain. This study used hot spot spatial statistics to identify the distribution of elevated uranium concentration in groundwater. The hypothesis whether spatial distribution of high value and low value of U is more likely spatially clustered due to random process near a uranium hotspot in groundwater was tested based on z score and Getis-Ord Gi* statistics. The method implemented in this study, can be utilized in the field of risk assessment and decision making to locate potential areas of contamination.

Anshuman Singh

Papers

Long short term memory (LSTM) recurrent neural network for low flow hydrological time series forecasting

Forecasting monthly precipitation using sequential modelling

Source characterization and human health risk assessment of nitrate in groundwater of middle Gangetic Plain, India

Application of Support Vector Regression for Modeling Low Flow Time Series

Using spatial statistics to identify the uranium hotspot in groundwater in the mid-eastern Gangetic plain, India