Search

Accurate monthly flow discharge prediction can yield significant evidence for sustainable management of water resources systems, optimal water allocation and use, mitigating flood events, and warning against famine. Inspiration to explore and develop skillful prediction models is a continuing attempt for various hydrologic assessments. The main aim of present study is to explore the potential of novel hybrid PSR-SVM-FFA model (integration of phase space reconstruction with support vector machine and firefly algorithm) and assess its performance against conventional radial basis function network, SVM, and hybrid SVM-FFA to predict flow discharge considering data from four gauge stations of Mahanadi River basin, India. PSR is applied to extract information and characteristics from flow time series and improve accuracy of hybrid SVM-FFA model. For assessing the model's enactment, Nash–Sutcliffe coefficient root-mean-square error and Willmott's Index (WI) indicators are calculated. The results showed that PSR-SVM-FFA model generated improved monthly flow predictions than other applied methods. The result indicates that best values of WI are 0.912–0.929, 0.949–0.956, 0.961–0.967, and 0.98–0.984 for RBFN, SVM, SVM-FFA, and PSR-SVM-FFA, respectively. This demonstrates that PSR-SVM-FFA provides prominent predictions compared to the other three approaches. [ABSTRACT FROM AUTHOR]

Flooding events are determining a significant amount of damages, in terms of economic loss and also casualties in Asia and Pacific areas. Due to complexity and ferocity of severe flooding, predicting flood-prone areas is a difficult task. Thus, creating flood susceptibility maps at local level is though challenging but an inevitable task. In order to implement a flood management plan for the Balrampur district, an agricultural dominant landscape of India, and strengthen its resilience, flood susceptibility modeling and mapping are carried out. In the present study, three hybrid machine learning (ML) models, namely, fuzzy-ANN (artificial neural network), fuzzy-RBF (radial basis function), and fuzzy-SVM (support vector machine) with 12 topographic, hydrological, and other flood influencing factors were used to determine flood-susceptible zones. To ascertain the relationship between the occurrences and flood influencing factors, correlation attribute evaluation (CAE) and multicollinearity diagnostic tests were used. The predictive power of these models was validated and compared using a variety of statistical techniques, including Wilcoxon signed-rank, t-paired tests and receiver operating characteristic (ROC) curves. Results show that fuzzy-RBF model outperformed other hybrid ML models for modeling flood susceptibility, followed by fuzzy-ANN and fuzzy-SVM. Overall, these models have shown promise in identifying flood-prone areas in the basin and other basins around the world. The outcomes of the work would benefit policymakers and government bodies to capture the flood-affected areas for necessary planning, action, and implementation. [ABSTRACT FROM AUTHOR]

[Display omitted] • Seventy Centella asiatica accessions were collected from Eastern states of India. • ANN model was used to predict and optimize centelloside content in C. asiatica. • Influence of soil and climatic data on centelloside content was determined. • MLP algorithm was found to have better predictive performance than RBF algorithm. • Sensitivity analysis revealed soil nitrogen and phosphorus as influential factors. Centella asiatica consists of centellosides that impart medicinal properties to the plant. Diverse geographical regions lead to variation in centelloside content due to the influence of environment and soil conditions. Therefore it is imperative to analyze C. asiatica from different geographical regions for influence of various environmental factors on the yield of centellosides. Considering these factors, the current work attempted to conduct chemotyping of C. asiatica to find out an elite genotype, design an artificial neural network (ANN) model to optimize and predict the drug yield in C. asiatica from various regions. High Performance Liquid Chromatography (HPLC) was used to analyze the centelloside content in the C. asiatica samples collected from 70 different geographical locations of eastern and north eastern states of India. The elite germplasms were selected based on the higher percentage (>8%) of centellosides (sum of major biomarkers i.e, asiaticoside, madecassoside, asiatic acid, and madecassic acid). HPLC analysis showed a significant variation in the centelloside contents that ranged from 0.37% to 10.91%. Hence, 3 germplasm accessions CA-18 (10.91%), CA-37 (9.60%), and CA-38 (9.04%) were selected as elite germplasms on the basis of their drug yielding potential. To develop the ANN model, the soil and climatic data of all the C. asiatica samples collected from 70 accessions were used. Based on its correlation coefficient, MLP (R2=0.94) was superior to RBF (R2=0.87), as it predicted the centelloside content very precisely. Additionally, the sensitivity analysis showed that soil nitrogen, phosphorus, and maximum temperature were the most influential factors that affected centelloside content. The developed ANN model was tested to forecast the centelloside content of a new site and the prediction efficiency was found to be 94.63%. Therefore, the developed ANN model seems to be a promising approach and could be of great importance in predicting locations for optimizing centelloside content in C. asiatica. [ABSTRACT FROM AUTHOR]

Severe Thunderstorms are the extreme weather convective features. It causes local calamities in various ways. Proper prediction with lead time is an important factor to prevent such calamities from saving people. Here, both probabilistic and machine learning techniques are applied to weather data to obtain proper predictions. Traditional methodologies are already available for such prediction purposes. However, Naïve Bayes and RBFN (Radial Basis Function Network) methodology have been introduced here with some specific weather parameters that has not done before remarkably. A comparative study was performed on weather data including Naïve Bayes, Multilayer Perceptron (MLP), K-nearest neighbor (KNN) and Radial Basis Function Network (RBFN). All these data have been procured from Kolkata located in north-east India. The result obtained by applying the Radial Basis Function Network is better among the three methods, yielding a correct prediction of 95% for severe "squall-storms" and 94% for "no storm". The predictions have a sufficient lead time of 10- 12 h. [ABSTRACT FROM AUTHOR]

Suspended sediment concentration (SSC) is one of the primary reasons with respect to watersheds or river basins, which must be assessed in a correct manner so that it will help decision makers to make right decisions regarding hydraulic structure, flash-flood, flood-mitigation of the basin. The present research evaluated efficacy of a hybrid model integrating Support Vector Machine with Whale optimization algorithm (SVM-WOA) for predicting SSC at Sundargarh and Salebhata stations in Mahanadi River, India. Various quantitative statistical evaluation constrains are applied to evacuate the model performance. Also, model performance of SVM-WOA is compared with SVM-PSO (Particle Swarm Optimization) and conventional SVM and RBFN (Radial Basis Function Network) models. The results reveal that, SVM-WOA performed superiorly in comparison to SVM-PSO, SVM and RBFN models for five different input scenarios during both training and testing phases. Hence, it is recommended to apply SVM-WOA as an appropriate technique for hydrological simulation at the basin. [ABSTRACT FROM AUTHOR]

River Ganga is one of the most significant rivers in the country. This river is the adobe for numerous aquatic species and microorganisms. The color of the river suddenly changed to green due to the rise of algal bloom in the Varanasi and nearby regions of the river Ganga during May–June 2021. These algal blooms can be detrimental to the aquatic animals of the river. This study analyzes the occurrence and the possible reasons for the algal bloom generation in the river for the considered stretch. Several factors like nutrient accumulation in the river through agricultural run-off, warm river temperature, low flow condition of the river, thermal stratification, and less turbid river water can be considered as possible reasons for algal bloom development. In this work, the optical remote sensing-based Sentinel 2 datasets have been used for the duration of mid-May 2021 to mid-June 2021. These datasets have been processed in the Google Earth Engine (GEE) platform, and chlorophyll concentration has been calculated using different satellite-based indices or band ratios. The chlorophyll concentration measurements have quantified the algal bloom growth. These indices or band ratios have been analyzed using several artificial neural network (ANN) architectures like multilayer perceptron (MLP) and radial basis function (RBF) along with the in situ values. It has been found that chlorophyll concentration has been highest for the mid-June 2021 time period in the considered river stretch. [ABSTRACT FROM AUTHOR]

The spatial distribution of soil pH, electrical conductivity, available nitrogen, phosphorus and potassium concentrations at Anaimalai block of Coimbatore district, India was interpolated by geostatistical analysis to create digital maps of soil NPK that renders numerous information for agro based policy making. 1627 soil samples in 30,700 acres were gathered by composite soil sampling method and rendered interpolation analysis to predict the soil parameters at unsampled locations. Based on geostatistical, deterministic and interpolation with barriers methods 13 interpolation models were generated and compared to validate the precision. Interpolation reveals the deficiency of soil available nitrogen and phosphorus whereas available potassium was in sufficient range. Radial basis function, inverse distance weighting, probability kriging, empirical Bayesian kriging and local polynomial interpolations were validated by linear correlation method that reveals the precisions of models with R2 values of 0.74, 0.82, 0.98, 0.89 and 0.79 (soil pH, EC, nitrogen, phosphorus and potassium respectively). [ABSTRACT FROM AUTHOR]

The changes in weather have beneficial and harmful effects on crop yields. There will be a loss of yield because of the diseases in crops. With the growing population, the fundamental want of food is growing. That is why agriculture gains a prominent position all around the world. It eventually ends up by a massive defeat for the farmers and the financial boom of India. The article's primary goalis to bring together farmers and cutting-edge technologies to minimise diseases in plant leaves. To enforce the idea, 'Tomato' is selected in which leaf sicknesses are expected and identified by the Artificial Intelligence algorithms, CNN (Convolution Neural Network) with pc technological know-how. Tomato is a mere consumable vegetable in India. In this investigation, seven types of tomato leaf disorders were sensed, including one wholesome elegance. The farmers are able to check the symptoms with the shapes of images of the tomato leaves with those expecting diseases. Its comparison of various classification and filters/methods with different techniques, such as K-Means classifier, SVM (Support Vector), RBF(Radial Basis Function) Kernel, Optimised MLP(Multilayer perceptron), NN classifier, BPNN (back-propagation neural network) and CNN Classifier. The classification accuracy of the existing method after experiment is RBF − 89%, k-means – 85.3%, SVM – 88.8%, Optimised MLP – 91.4%, NN – 97, BPNN – 85.5%, CNN – 94.4%. The proposed architecture can achieve the desired accuracy of 99.4%. [ABSTRACT FROM AUTHOR]

When it comes to cancer in Indian women, breast cancer is one of the most common and frequently occurring cancer types. In most of the developing countries like India, the overhead charges for breast cancer are very high which leads to a poor mortality rate when compared to western countries. As of now, large numbers of researches have been carried out based on deep learning and machine learning prediction of Breast cancer. This paper reviews some of the existing machine learning algorithms like Decision Tree (DT), Multi-Layer Perceptron (MLP), Random Forest (RF) and Support Vector Machines (SVMs) in linear (SVML), polynomial (SVM-P), and radial basis function (SVM-RBF) on the Breast Cancer Wisconsin (BCW) data set using Python. An experimental result shows that SVM-L outperforms all the above algorithms with 96.5% accuracy. [ABSTRACT FROM AUTHOR]

Apple orchards constitute an important agro-ecosystem across Kashmir Himalayan Region (KHR), but estimates of their soil organic carbon (SOC) are unavailable. We investigated 174 apple orchards for estimating SOC spatial distribution across KHR at regional scale and evaluate accuracy of five interpolation methods. The selected orchards were representative of varied size, age, and management practices prevalent across KHR. Soil samples were collected from both tree rows and alleys between tree rows by digging pits at three depth intervals of 0–10; 10–20 and 20–30 cm. A total of 1044 soil samples were collected. The interpolation methods used included, ordinary kriging (OK), inverse distance weighing, empirical bayesian kriging, radial basis functions and local polynomial interpolation. Cross validation was used to assess the comparative performance of each method by measuring interpolation bias and accuracy. The soils were structurally less stony, with weak granular to moderate crumb at surface and sub-angular to angular blocky structure at sub-surface. The soils exhibited lowest co-efficient of variation for soil acidity and highest for soil electrical conductivity. With mean value of 12.33 ± 3.98 g kg−1 on concentration basis and 48.45 Mg C ha−1 on stock basis, the SOC declined with increasing soil depth, so much that near surface (0–10 cm) SOC content could explain well over 64% and 35% of second (20–30 cm) and third layer (20–30 cm) variation respectively. Exponential model best described SOC content across all depths. Semi-variograms of SOC at topsoil exhibited larger nugget effect while nugget sill ratios of 43% suggested moderate SOC spatial dependence. Cross validation exhibited better accuracy for OK in terms of indicating good match between observed and predicted SOC. This study exemplifies that apple orchards across KHR are significant carbon pool, improves our understanding of spatial distribution of SOC and help in evaluating soil health by providing site specific maps. [ABSTRACT FROM AUTHOR]

Forecasting of irrigation demand is important for decision-making, and reference evapotranspiration (ETo) is a key determinant in evaluating water demand in advance. However, the precise determination of ETo is fairly difficult, and complex machine learning approaches are often used for this. This study, carried out in Veeranam tank, India, determines the multivariate analysis of correlated variables involved in the estimation and modelling of ETo from 1995 to 2016. A reduced-feature data model was constructed with the most significant variables of the model extracted by principal component analysis. This work also explores the effectiveness of a deep learning neural network (DLNN) with the reduced-feature model in predicting ETo in comparison with the conventional Food and Agriculture Organization of the United Nations (FAO-56) Penman-Monteith equation and the radial basis function neural network (RBFNN) as a baseline machine learning method. The input variable dimensionality was reduced from six to three most significant variables in ETo modelling. Among machine learning methods, DLNN proved to be effective in ETo prediction with the reduced-feature data model. [ABSTRACT FROM AUTHOR]

The commoditization of voice, saturation of the urban market, fierce competition, and the increased cost of the spectrum have forced the mobile telecom network operators to promote and garner revenue from non-voice services (NVS) namely value-added services (VAS) and data services (DS). It is a fact that monthly revenue from different segments of a mobile telecom service provider is non-linearly related to its previous revenue. Hence, existing linear prediction models such as regression and linear combiner do not exhibit accurate prediction performance. This article proposes one linear adaptive linear combiner (ALC) and three nonlinear (trigonometric expansion based neural network (TENN), multi-layer perceptron (MLP) and radial basis function (RBF)) models for prediction of revenue from voice services (VS), DS and VAS segments. The real-life revenue data of a mobile telecom service provider in a licensing area of India is utilized for this study. The predictor performance has been obtained from the simulation study of the models and analyzed. It is observed that the prediction performance of linear ALC model is the worst. The TENN model outperforms the MLP and RBF models from amongst the proposed nonlinear models based on the available historical data. In essence, the study demonstrates that the overall ranking of the four models based on prediction performance are TENN, RBF, MLP, and ALC. [ABSTRACT FROM AUTHOR]

Probabilistic methods are the most efficient methods to account for different types of uncertainties encountered in the estimated rock properties required for the stability analysis of rock slopes and tunnels. These methods require estimation of various parameters of probability distributions like mean, standard deviation (SD) and distributions types of rock properties, which requires large amount of data from laboratory and field investigations. However, in rock mechanics, the data available on rock properties for a project are often limited since the extents of projects are usually large and the test data are minimal due to cost constraints. Due to the unavailability of adequate test data, parameters (mean and SD) of probability distributions of rock properties themselves contain uncertainties. Since traditional reliability analysis uses these uncertain parameters (mean and SD) of probability distributions of rock properties, they may give incorrect estimation of the reliability of rock slope stability. This paper presents a method to overcome this limitation of traditional reliability analysis and outlines a new approach of rock mass characterization for the cases with limited data. This approach uses Sobol's global sensitivity analysis and bootstrap method coupled with augmented radial basis function based response surface. This method is capable of handling the uncertainties in the parameters (mean and SD) of probability distributions of rock properties and can include their effect in the stability estimates of rock slopes. The proposed method is more practical and efficient, since it considers uncertainty in the statistical parameters of most commonly and easily available rock properties, i.e. uniaxial compressive strength and Geological Strength Index. Further, computational effort involved in the reliability analysis of rock slopes of large dimensions is comparatively smaller in this method. Present study also demonstrates this method through reliability analysis of a large rock slope of an open pit gold mine in Karnataka region of India. Results are compared with the results from traditional reliability analysis to highlight the advantages of the proposed method. It is observed that uncertainties in probability distribution type and its parameters (mean and SD) of rock properties have considerable effect on the estimated reliability index of the rock slope and hence traditional reliability methods based on the parameters of probability distributions estimated using limited data can make incorrect estimation of rock slope stability. Further, stability of the rock slope determined from proposed approach based on bootstrap method is represented by confidence interval of reliability index instead of a fixed value of reliability index as in traditional methods, providing more realistic estimates of rock slope stability. [ABSTRACT FROM AUTHOR]

Precise estimation of evapotranspiration (ET) is extremely essential for efficient utilization of available water resources. Among the empirical models, FAO-Penman-Monteith equation (FAO-PM) is considered as standard method to determine reference evapotranspiration (ET0 ). In developing countries like India, application of FAO-PM equation for ET0 estimation has certain limitations due to unavailability of specific data requirements. Several empirical models such as Hargreaves, Turc, Blaney-Criddle etc., are also considered for ET0 estimation. However, ET0 estimates obtain with these models are not comparable with benchmark FAO-PM ET0. To address this issue, potential of radial basis function neural network (RBFNN) is investigated to estimate FAO-PM ET0. Result obtained with proposed RBFNN models are compared with equivalent multi-layer artificial neural network (MLANN) and empirical approach of Hargreaves, Turc and Blaney-Criddle. Lower RMSE values obtained with RBFNN and MLANN models is an indication of improved performance over empirical models. Similarly, higher R² and Efficiency Factor obtained with RBFNN and MLANN models also approves the superiority of machine learning techniques over empirical models. Among the two machine learning techniques, RBFNN models performed better as compared to MLANN. In a nut shell, proposed RBFNN models can simulate FAO-PM ET0 even with limited meteorological parameters and consistence degree of accuracy level. [ABSTRACT FROM AUTHOR]

Rainfall forecast is a hotspot in meteorological studies in the last few years. The key difficulty in forecast accuracy lies in the nonlinearity of rainfall data. Considering the potential of support vector machine (SVM) to solve nonlinear time series, this paper develops an SVM-based model based on the monthly rainfall data from 1901 to 2015 in Northwest India. The forecast accuracies of four different kernels were compared, including linear, polynomial, radial basis function (RBF) and sigmoid kernels. The comparison shows that the SVM-based model with the RBF kernel achieved the smallest root mean squared error (RMSE) at the lookback of 25, C=1, γ =0.01 and 𝜀=0.1. Our research provides an effective tool to predict rainfall in regions with similar meteorological conditions to northwest India. [ABSTRACT FROM AUTHOR]

• Regular breast health examination is the only way to reduce the breast cancer mortality rate. • Infrared Breast Thermography can pinpoint the minute temperature changes related to breast health. • The non-radiating Infrared breast thermography can be used as a routine breast health checkup tool. • By identifying the breast abnormalities in early stages, infrared breast thermography provides more treatment options. The high incidence and mortality rate of breast cancer in India and the limitations of gold standard method X-ray mammography to be used as a screening and diagnostic modality in young women tempted us to evaluate the efficiency of highly sensitive and non-radiating Infrared Breast Thermography (IBT) in early breast abnormality detection. This study investigates the efficiency of IBT by doing Temperature based analysis (TBA), Intensity based analysis (IBA), and Tumor Location Matching (TLM). In TBA and IBA, several temperature and intensity features were extracted from each thermogram to characterize healthy, benign and malignant breast thermograms. In TLM, the locations of suspicious regions in thermograms were matched with the tumor locations in mammograms/Fine Needle Aspiration Cytology images to prove the efficiency of IBT. Thirteen different sets of features have been created from the extracted temperature and intensity features and their classification performances have been evaluated by using Support Vector Machine with Radial basis function kernel. Among all feature sets, the feature set comprising the statistically significant (p < 0.05) features provides the highest classification accuracy of 83.22% with sensitivity 85.56% and specificity 73.23%. Based on the results of this study, IBT is found to be potential enough to be used as a proactive technique for early breast abnormality detection in asymptomatic population and hence, capable of identifying the subjects that need urgent medical attention. [ABSTRACT FROM AUTHOR]

Visceral leishmaniasis or Kala-azar (KA) is a Vector-Borne Disease (VBD) that remains the second-largest parasitic killer across the globe (mortality rate: 75–95%). More than 60% of KA cases originate in South Asia, wherein India accounts for 2/3rd of the cases, and Bihar, a state in India, alone accounts for more than 50% of the Indian cases. Past studies suspected climate change vulnerabilities as a driving cause of KA outbreaks. The VBDs-based epidemic prediction systems have been developed to mitigate recurrent outbreaks; however, Machine Learning (ML) based approaches still need to be explored for modeling changing climate impacts on KA cases. This study, for the first time, develops a Radial Basis Function (RBF) kernel-based Support Vector Regression (SVR), hereinafter RBF-kernel-based-SVR model for the most-affected endemic districts of Bihar (northern-India), using the data from 2016 and 2021. Forward selection, backward elimination, and stepwise regression procedures were adopted while selecting influential climatic variables, followed by the k -fold cross-validation technique and, then, the RBF-kernel-based-SVR algorithm for classification. Results suggested that temperature, wind speed, rainfall, and population density significantly contributed to the KA outbreaks. This study also developed Multiple Linear Regression (MLR) and Multilayer Perceptron (MLP) models to compare SVR with other classification models. Findings indicated that the proposed RBF-kernel-based-SVR model [Correlation Coefficient (CC) = 0.82, Root-Mean-Square Error (RMSE) = 12.20, and Nash–Sutcliffe Efficiency (NSE) = 0.66] outperformed MLR (0.81, 14.20, 0.48) and MLP (0.81, 12.95, 0.61). Study recommends using the RBF-kernel-based-SVR model as a quick and efficient model capable of detecting KA cases with high predictability even under limited data availability. Such models can assist public health authorities, given monitoring KA spread, learning the climate impacts of outbreaks, and ensuring timelier health services. [ABSTRACT FROM AUTHOR]

This paper examines the quality of groundwater of Sabour block, Bhagalpur district of Bihar state, which lies on the southern region of Indo-Gangetic plains in India. Fifty-nine samples from different sources of water in the block have been collected to determine its suitability for drinking and irrigational purposes. From the samples electrical conductivity (EC), pH and concentrations of Calcium (Ca), Magnesium (Mg), Sodium (Na), Potassium (K), carbonate ion (CO), Bicarbonate ion (HCO), Chloride ion (Cl), and Fluoride (F) were determined. Surface maps of all the groundwater quality parameters have been prepared using radial basis function (RBF) method. RBF model was used to interpolate data points in a group of multi-dimensional space. Root Mean Square Error (RMSE) is employed to scrutinize the best fit of the model to compare the obtained value. The mean value of pH, EC, Ca, Mg, Na, K, HCO , Cl, and F are found to be 7.26, 0.69, 38.98, 34.20, 16.92, 1.19, 0.02, and 0.28, respectively. Distribution of calcium concentration is increasing to the eastern part and K concentrations raise to the downstream area in the southwestern part. Low pH concentrations (less than 6.71) occur in eastern part of the block. Spatial variations of hardness in Sabour block portraying maximum concentration in the western part and maximum SAR (more than 4.23) were recorded in the southern part. These results are not exceeding for drinking and irrigation uses recommended by World Health Organization. Therefore, the majority of groundwater samples are found to be safe for drinking and irrigation management practices. [ABSTRACT FROM AUTHOR]

India being one of the major producers of fish contributes 5.5 percent of global fish production and ranks second in the world after China. The production of aquaculture mainly depends on the quality of land selected for aqua farming. Neural Network algorithms have been applied to classify the aquaculture sites based on 6 input variables viz., water, soil, support, infrastructure, input and risk factor. An artificial neural network (ANN) consists of huge number of interconnected elements called neurons that work together to solve a specific problem. An Artificial Neural network can be used for classification, prediction, pattern recognition etc., through a learning process. In this paper, the models were constructed using three Neural Network algorithms viz., Back Propagation Network (BPN), Radial Basis Function (RBF) and Linear Vector Quantization (LVQ). The models classify each aquaculture site into 3 classes viz., suitable, moderate and unsuitable. From the results of the three models, it has been found that Radial Basis Function model not only gives accurate results but also time taken for training the dataset is less when compared with the other two Neural Network models. The results obtained from the neural network models were validated with the results of the fuzzy model. [ABSTRACT FROM AUTHOR]

Breast cancer has been and continues to be a cause of major health concern for women. It is more prevalent in old age, but its incidence has increased in recent years in groups below 50 years old, as in India. According to the Indian Council of Medical Research (ICMR) 2020, 50% of all the cases are in the age group of 25–50 where the numbers are staggering and constantly rising. The increase in incidence over the years indicates an urging need for innovative approaches to enhance breast cancer detection early. Thermography is non-contact imaging modalities and has potential to detect breast cancer at an early stage. Though thermography has capable of detecting breast cancer early, the challenge lies in the interpretation of the breast thermograms with respect to features and subsequent analysis. The present work discusses image acquisition, image processing related pre-processing, segmentation, and feature extraction. The extracted features were analyzed using ANOVA (Analysis of variance) statistical analysis. Statistical analyses were done in order to find the appropriate feature on the whole and quadrant breast. Statistical analysis results clearly reveal existence of thermal symmetry for the healthy subjects (p value >.05) in both whole and quadrant breast regions. In the case of abnormal subjects, whole breast analyses revealed the significance (p value <.05) for features like mean, variance, standard deviation, kurtosis, skewness, entropy, energy, homogeneity and contrast whereas upper outer quadrant analyses showed significance for all above features except contrast. The well correlated features of upper outer quadrant and whole breast were given as input for the Support Vector Machine – Radial Basis Function (SVM – RBF) classifier with grid search method. The results revealed that whole breast analysis has achieved 92.86% accuracy and upper outer quadrant breast analysis has achieved 85.71% accuracy. The results clearly indicate the involvement of upper outer quadrant and whole breast in early detection of breast cancer using thermal imaging. • Quadrant and whole breast significance on breast cancer detection was investigated. • Use of GLCM and statistical features for classification. • Use of statistical analysis (ANOVA) for the extracted features. • Use of grid search method to arrive at optimal hyperparameters. [ABSTRACT FROM AUTHOR]

BACKGROUND: Coronary artery disease due to atherosclerosis is an epidemic in India. An estimated 1.3 million Indians died from this in 2000. The projected death from coronary artery disease by 2016 is 2.98 million.OBJECTIVE: To build an effective model which assorts the individuals, whether they belong to the normal group, risk group and pathologic group regarding atherosclerosis in real time by doing necessary preprocessing techniques and to compare the performance with other state-of-the-art machine learning techniques. METHODS: In this work we have employed STULONG dataset. We have made a deep case study in selecting the attributes which contributes for higher accuracy in predicting the target. The selected attributes includes missing values. Initially our work includes imputation of missing values using Iterative Principal Component Analysis (IPCA). The second step includes selecting best features using Fast Correlation Based Filter (FCBF). Finally the classifier Multiclass Support Vector Machine (SVM) with kernel Radial Basis Function (RBF) is used for classification of atherosclerotic community. RESULTS: For the subjects belonging to the classes of normal, risk and pathologic, our methodology has outperformed with an accuracy of 99.85%, 99.80% and 99.46% respectively. CONCLUSION: The combined optimization methods such as Iterative Principal Component Analysis (IPCA) for missing value imputation, Multiclass SVM for classifying normal, risk and pathologic community in real time has performed with overall accuracy of about 98.97%. The essential pre-processing technique, Fast Correlation Based Filter (FCBF) was employed to further intensifying the target. [ABSTRACT FROM AUTHOR]

The present study demonstrates the capability of two preprocessing techniques such as wavelets and moving average (MA) methods in combination with feed-forward neural networks-namely, back propagation (BP) and radial basis (RB) and multiple linear regression (MLR) models-in the prediction of the daily inflow values of the Malaprabha reservoir in Belgaum, India. Daily data on 11 years of rainfall, inflow, and streamflow at an upstream gauging station have been used. The observed inputs are decomposed into subseries using discrete wavelet transform with different mother wavelet functions, and then the appropriate subseries is used as input to the neural networks for forecasting reservoir inflow. Model parameters are calibrated using 7 years of data, and the remaining data are used for model validation. More statistical indices have been used to determine the optimal models. Optimum architectures of the wavelet neural network (WNN) models are selected according to the obtained evaluation criteria in terms of the Nash-Sutcliffe efficiency coefficient, root mean squared error, and correlation coefficient. The result of this study has been compared by developing two standard neural network models and a multiple linear regression (MLR) model and MA. The results of this study indicate that the WNN model performs better compared to artificial neural network (ANN) and MLR models in forecasting the inflow hydrograph effectively. The study only used reservoir inflow data from one area, and further studies using data from various areas may be required to strengthen these conclusions. [ABSTRACT FROM AUTHOR]

Summary: Groundwater is a prominent source of drinking and domestic water in the world. In this context a reliable water supply policy, specifically during the dry season necessitates accurately acceptable predictions of water table depth fluctuations. Owing to the difficulties of identifying non-linear model structure and estimating the associated parameters, Back Propagation Neural Network (BPNN) and Radial Basis function network (RBFN) model is taken into account for study. Back propagation neural network model with delta algorithm is calibrated using historical groundwater level records and related hydro-meteorological data to simulate water table fluctuations in the study area. Similarly RBFN network has been used to analyze the water table depth prediction for four different stations. In the present investigation comparative assessment of water table depth for four different stations as well as the sensitivity of above two different models have been identified. [Copyright &y& Elsevier]

Multiple regression analysis (MRA) and two artificial neural networks (ANN) – multilayer perceptron (MLP) and radial basis function network (RBF) – were compared in terms of their accuracy and efficiency when applied to prediction of the biochemical oxygen demand (BOD) concentration at effluent and intermediate points of subsurface flow constructed treatment wetlands (CTW). The data used in this work was obtained from various hydraulic and BOD loading of a pilot CTW located in India and comprised of 91 patterns. The dataset was normalized and transformed using principal component analysis (PCA) in order to increase the efficiency of the modelling. At the modelling stage the most adequate models were determined by using systematic approach. The candidate ANN models were cross-validated to find optimal network architectures and values of training algorithm parameters. MRA performance was maximized by utilizing 14-fold cross-validation. MRA as well as ANN models were found to provide an efficient and robust tool in predicting CTW performance. MLP and RBF produced the most accurate results indicating strong potential for modelling of wastewater treatment processes. [Copyright &y& Elsevier]

• TSI increased from 18–22 to 44–54 showing deterioration of trophic state. • 30% of area comes under high vulnerable zone in the pre-dam period. • 54% of area comes under high vulnerable zone in the post-dam period. • RBF model is the best model of habitat vulnerability (AUC-0.877 and 0.868). • A strong spatial relation between TSI and habitat vulnerability is found. Damming over the river remarkably alters the downstream hydrological and ecological state of the river and riparian wetlands. Considering this, the present work intended to explore the habitat quality and the trophic state index (TSI) of the wetland in Tangan river basin of India and Bangladesh. The spatial linkages between habitat quality and TSI was also assessed. Random forest (RF), Radial basis function neural network (RBF) and Artificial neural network (ANN) algorithms based habitat vulnerability models were constructed based on eight proxy data layers, which are mainly related to hydrology. The ANN and RBF models were found as the best representative. In the pre-dam period, about 30% area was under high and very high habitat vulnerability, while it was increased to 54% after damming. TSI value in the post-dam period was found significantly higher (44–54) in comparison to pre-dam phase (18–22). It was also identified that vulnerable habitat areas were more susceptible to poorer water quality. Hence, in water quality and ecosystem related data sparse conditions, the proxy parameters can be used effectively for habitat vulnerability modelling. Furthermore, results clearly demonstrated that damming has deteriorated the habitability and trophic state of the wetland. To restrain the growing impact, ecological flow maintenance to the dam downstream river and riparian wetland is essential. [ABSTRACT FROM AUTHOR]

Groundwater crisis across the world is a thought-provoking issue and for resolving the problem, it is highly necessary to identify the potential groundwater zones and estimate water yield. The present work intends to identify potential groundwater zone based on ensemble modeling assembling advance machine learning algorithm like Random Forest (RF), Radial Basis Function (RBF nn) and Artificial Neural Network (ANN) and set theories like union and intersection based modeling using 15proxyconditioning parameters for developing sustainable water resource management plan. The work is based on Tangon river basin of Barind tract of Eastern Indian and Bangladesh, suffers from water scarcity. Groundwater potentiality models have identified 34.93–35.67% area to total basin area (2388.88 km2) at the proximity of lower reach of the main river as very high to highly potential for groundwater. Among the employed parameters elevation, slope, land use/land cover, distance from perennial segment of stream are identified as the dominant in this case. For assessing the accuracy level of the models, Receiver operating characteristics (RoC), proximity test and aquifer breadth data are used. Both the ensemble models and set theory centric models show very good to excellent performances suitably identifying groundwater potentiality. However, the performance level of ensemble modeling is more satisfactory. As the groundwater potentiality zones of the study area are well delimitated, this study may be useful for adopting a suitable water resource management plan. Image 1 • 29% of the total area at the lower part of the basin is highly potential. • Performance of Machine learning models is better over set theory-based models. • Ensemble prediction (RF_RBFnn_ANN) is found as the best predictive model. • Groundwater crisis exits but the further scope of its development admissible. [ABSTRACT FROM AUTHOR]

Banks play a vital role in strengthening the financial system of a country; hence, their survival is decisive for the stability of national economies. Therefore, analyzing the survival probability of the banks is an essential and continuing research activity. However, the current literature available indicates that research is currently limited on banks' stress quantification in countries like India where there have been fewer failed banks. The literature also indicates a lack of scientific and quantitative approaches that can be used to predict bank survival and failure probabilities. Against this backdrop, the present study attempts to establish a bankruptcy prediction model using a machine learning approach and to compute and compare the financial stress that the banks face. The study uses the data of failed and surviving private and public sector banks in India for the period January 2000 through December 2017. The explanatory features of bank failure are chosen by using a two-step feature selection technique. First, a relief algorithm is used for primary screening of useful features, and in the second step, important features are fed into the support vector machine to create a forecasting model. The threshold values of the features for the decision boundary which separates failed banks from survival banks are calculated using the decision boundary of the support vector machine with a linear kernel. The results reveal, inter alia, that support vector machine with linear kernel shows 92.86% forecasting accuracy, while a support vector machine with radial basis function kernel shows 71.43% accuracy. The study helps to carry out comparative analyses of financial stress of the banks and has significant implications for their decisions of various stakeholders such as shareholders, management of the banks, analysts, and policymakers. [ABSTRACT FROM AUTHOR]

The two main challenges of predicting the wind speed depend on various atmospheric factors and random variables. This paper explores the possibility of developing a wind speed prediction model using different Artificial Neural Networks (ANNs) and Categorical Regression empirical model which could be used to estimate the wind speed in Coimbatore, Tamil Nadu, India using SPSS software. The proposed Neural Network models are tested on real time wind data and enhanced with statistical capabilities. The objective is to predict accurate wind speed and to perform better in terms of minimization of errors using Multi Layer Perception Neural Network (MLPNN), Radial Basis Function Neural Network (RBFNN) and Categorical Regression (CATREG). Results from the paper have shown good agreement between the estimated and measured values of wind speed. According to the result, it can be concluded that ANN model with MLPNN could produce the acceptable prediction of the wind speed for given on wind direction. • To predict the wind speed (m/sec) at 65 m height. • To build the ANN Models and Categorical Regression (CATREG) for wind speed prediction. • To produce the validated and acceptable prediction. [ABSTRACT FROM AUTHOR]