Search

The present study aimed to develop a methodology to estimate jute (Corchorus sp.) cultivated areas in the Indian subcontinent using sentinel-1A Synthetic aperture radar (SAR) C band dual polarization data. We developed a hierarchical decision tree (DT) model to estimate jute area and compared it with other supervised and unsupervised methods. The proposed DT model showed improved overall accuracy of 82–83% (Kappa coefficient, k = 0.78–0.79) over supervised or unsupervised methods and validated against Directorate of Economics and Statistics (DES) data, Government of India during 2016–2019. The VV backscatter record improved accuracy over VH backscatter. [ABSTRACT FROM AUTHOR]

Renewable energy sources (RESs) are increasingly used to meet the world's growing electrical needs, especially for the economic benefits and environmental problems associated with fossil fuel use. Small-scale renewable energy sources, controllable loads, energy storage devices, and other nonrenewable sources are effectively integrated to form a virtual power plant (VPP). Uncertainty in forecasting renewable energy generation due to the intermittent nature of renewable energy sources is one of the biggest challenges in VPPs. Power generation by RESs changes with the day of the week, season, location, climate, and resource availability. Also, load demand and utility price vary with time and need to be forecasted for proper energy management of VPPs. However, the dispatching and planning of VPPs are significantly impacted by the volatile nature of RESs, load demand, and utility price. Predicting these uncertainties with high accuracy is essential to balance the electrical power generation and the load demand. In this article, a decision tree (DT) algorithm is proposed, to predict the uncertainty parameters, such as the day-ahead power from the RES, load demand, and utility prices of VPPs. The efficiency of the proposed model and the predicted results are compared with other complex models, such as the artificial neural network (ANN) and auto-regressive integrated moving average (ARIMA) algorithms. Root-mean square error (RMSE), mean square error (MSE), coefficient of determination (R2), and mean absolute error (MAE) are the statistical metrics used to evaluate the accuracy of the prediction. One-year meteorological data of the Chennai zone in India is considered for predicting the uncertainty parameters. IEEE 16-bus and 33-bus test systems are used to validate the forecasting model. It is evident from the results that the proposed DT algorithm can predict the uncertainty parameters more accurately and use lesser time than the ANN and ARIMA algorithms. [ABSTRACT FROM AUTHOR]

Introduction: The World Health Organization (WHO) has released the updated cardiovascular disease (CVD) risk prediction charts in 2019 for each of the 21 Global Burden of Disease regions. The WHO advocates countries to implement population-based CVD risk assessment and management using these updated charts for preventing and controlling CVDs. Objective: To assess the cost-effectiveness of implementing risk-based CVD management using updated WHO CVD risk prediction charts in India Methods: We developed a decision tree combined with Markov Model to simulate implementing two community-based CVD risk screening strategies (interventions) compared with the current no-screening scenario. In the first strategy, the whole population is initially screened using the WHO non-lab-based CVD risk assessment method, and those with ≥10% CVD risk are subjected to WHO lab-based CVD risk assessment (two-stage screening). In the second strategy, the whole population is subjected only to the lab-based CVD risk assessment (single-stage screening). A mathematical cohort of those aged ≥40 years with no history of CVD events was simulated over a lifetime horizon with three months of cycle length. Data for the model were derived from a primary study and secondary sources. Incremental cost-effectiveness ratios (ICERs) were determined for the screening strategies and sensitivity analyses. Results: The discounted Incremental cost-effectiveness ratio per QALY gained for both the two-stage (US$ 105; ₹ 8,656) and single-stage (US$ 1073; ₹ 88,588) screening strategies were cost-effective at an implementation effect of 40% when compared with no screening scenario. Implementing CVD screening strategies are estimated to cause substantial reduction in the number of CVD events in the population compared to the no screening scenario. Conclusion: In India, both CVD screening strategies would be cost-effective, and implementing the two-staged screening would be more cost-effective. Our findings support implementing population-based CVD screening in India. Future studies shall assess the budget impact of these strategies at different implementation coverage levels. [ABSTRACT FROM AUTHOR]

Background: Gujarat has implemented an mHealth program, technology for community health operations-plus (TeCHO+) in 2019. TeCHO+ is a mobile and web-based application that essentially enables data entry by the frontline workers providing service at the time and place of service delivery to improve the coverage and data quality. It also facilitates early identification of morbid condition and timely treatment. This study assessed the cost-effectiveness of TeCHO+ program in Gujarat. Materials and Methods: The study compared key program outcome indicators before and after the launch of TeCHO+ program. As the program was launched across the State, eMamta, the previous version of mother and child tracking system was used for comparison. A decision tree was parameterized to estimate change in disability-adjusted life year (DALY) and cost as a result of implementing TeCHO+ from a health system perspective. Results: TeCHO+ incurred a cost of Rs. 2,624 per beneficiary against Rs. 1,075 per beneficiary under the previous eMamta program. TeCHO+ has resulted in significant DALY averted through early identification of high-risk cases both among pregnant women and children. Overall, cost-effectiveness analysis indicated that TeCHO+ incurred an incremental cost of Rs. 1802.84 per DALY averted, which is 1.19% of the GDP per capita of India (year 2020). Conclusion: This study concludes that TeCHO+ is cost-effective for mother and child care and can be considered for replicating. [ABSTRACT FROM AUTHOR]

About 10.5 percent of global adult population is living with diabetes. India has 77 million diabetic patients, it is the second highest in the world. Developing countries such as India face a huge burden of diabetes and its complications. Even children at the age of five are suffering from this disease. It is high time that people understand the gravity of the situation and make themselves fit to fight the disease than to suffer with it. If diabetes is not identified and treated in right time, it may lead to chronical health issues. In this paper a machine learning based prediction model is built to find the factors leading to complicated health issues such as cardio vascular disease. This model identifies the attributes that highly contribute to cardio vascular disease and compare various machine learning algorithms to predict cardio vascular disease among diabetic patients. It identifies the best algorithm from a set of supervised learning algorithms such as KNN, Decision Tree, Random Forest, Naïve Bayes and Gradient Boosting for prediction based on several performance metrics. The algorithms are compared based on the performance metrics such as accuracy, precision, recall, F1 score, time taken to train and time taken to test. We identified that Decision Tree with entropy as the split criterion achieved the highest accuracy. [ABSTRACT FROM AUTHOR]

This study investigated the influence of climate factors on malaria incidence in the Sundargarh district, Odisha, India. The WEKA machine learning tool was used with two classifier techniques, Multi-Layer Perceptron (MLP) and J48, with three test options, 10-fold cross-validation, percentile split, and supplied test. A comparative analysis was carried out to ascertain the superior model among malaria prediction accuracy techniques in varying climate contexts. The results suggested that J48 had exhibited better skill than MLP with the 10-fold cross-validation method over the percentile split and supplied test options. J48 demonstrated less error (RMSE = 0.6), better kappa = 0.63, and higher accuracy = 0.71), suggesting it as most suitable model. Seasonal variation of temperature and humidity had a better association with malaria incidents than rainfall, and the performance was better during the monsoon and post-monsoon when the incidents are at the peak. [ABSTRACT FROM AUTHOR]

Over recent decades, ambiguous ground vibration induced by blasting operation can cause extensive damage to structures, lives and fields in and around mine premises. As a consequence, it is indispensable to measure the ambiguous ground vibration intensity levels for assessing and reduce their perilous impact. In this investigation, estimation and evaluation of blast-induced ground vibration in terms of peak particle velocity (PPV) through the ensemble machine learning intelligent algorithms were carried out. One hundred and 21 experimental and blasting events were monitored to collect the realtime field data in Mine-A, India. The collected data was randomly split into training and testing to generate models. Eight input parameters include number of holes, burden, spacing, hole diameter, hole depth, top stemming, maximum explosive charge per delay and the distance were selected for development of ensemble machine learning algorithms. An eXtreme gradient boosting (XGBoost) and random forest (RF) ensemble model, Decision Tree were developed to assess the PPV levels. In addition to that, four empirical predictor models proposed by the US Bureau of Mines, Langefors-Kihlstrom, Central Mining Research Institute, and Bureau of Indian Standards were applied to derive a relation between PPV and its influencing parameters. The accuracy and efficiency of developed models can be determined by performance evaluation metrics chosen as the coefficient of determination (R²), and root mean square error (RMSE). Among all models, yielded results evidence that the Decision Tree ensemble model with the R² of 0.9549, and RMSE of 0.0444 was more precise optimum model to assess the PPV. Besides, a sensitivity analysis method was applied in this current study to know the role of the input parameters in estimating PPV. The determined results inferred that burden, number of holes and top stemming are more influenced parameters on the intensity of PPV levels. [ABSTRACT FROM AUTHOR]

Tyre condition monitoring system (TCMS) is an emerging electronic safety system present in most of the new generation vehicles. It is intended to warn the driver when wheels are unbalanced or inflation pressure in a tyre goes below a certain recommended value. In India only limited vehicles having this critical safety system installed because it is highly expensive. The existing TCMS monitor only the tyre pressure using direct or indirect methods. In recent days the indirect TCMS using vibration signals and machine learning techniques are gaining momentum. This paper reports the effect of wheel balancing on a proposed affordable indirect TCMS. The present study goes one step forward to include the wheel balancing in the TCMS, thus increasing the scope of TCMS. The study was carried out using vibration signal data acquired from a low cost accelerometer placed on the wheel hub. The features required were taken out from the acquired data with the help of statistical feature extraction techniques. Features selection and feature classification were processed with J48 decision tree algorithm. The effect of wheel balancing in classification accuracy is clearly explained. [ABSTRACT FROM AUTHOR]

Loosening of rockmass during its excavation in an infrastructure project is carried by rock blasting. The blast-induced ground vibrations pose a major challenge to the blasting engineers, whose main objective is to control their potential to cause any damage to the buildings in the vicinity. The research reported in this paper explains how the error in the prediction of the Peak Particle Velocity (PPV) by the United States Bureau of Mines (USBM)-based approach can be minimised using machine learning techniques. The complex correlation between the blast parameter and the PPV value has been modelled using the least square boosted decision tree approach after the selection of the best suitable feature has been selected based on the correlation matrix. The proposed model automatically maps the input blast feature (SD) with the target PPV values by aggregating the decision of various weak learners. The generalization of the proposed model has been validated through a 5-fold cross-validation approach using a dataset comprising of two hundred blast records generated by monitoring the blasts at International airport site, Navi Mumbai, India. The assessment of the prognostic ability of the proposed model demonstrates that it has outperformed the USBM-based approach for PPV prediction. The results establish that the predictions by the pro- posed model are closer to the measured values than the other regression models. [ABSTRACT FROM AUTHOR]

This study aims to develop an assumption-free data-driven model to accurately forecast COVID-19 spread. Towards this end, we firstly employed Bayesian optimization to tune the Gaussian process regression (GPR) hyperparameters to develop an efficient GPR-based model for forecasting the recovered and confirmed COVID-19 cases in two highly impacted countries, India and Brazil. However, machine learning models do not consider the time dependency in the COVID-19 data series. Here, dynamic information has been taken into account to alleviate this limitation by introducing lagged measurements in constructing the investigated machine learning models. Additionally, we assessed the contribution of the incorporated features to the COVID-19 prediction using the Random Forest algorithm. Results reveal that significant improvement can be obtained using the proposed dynamic machine learning models. In addition, the results highlighted the superior performance of the dynamic GPR compared to the other models (i.e., Support vector regression, Boosted trees, Bagged trees, Decision tree, Random Forest, and XGBoost) by achieving an averaged mean absolute percentage error of around 0.1%. Finally, we provided the confidence level of the predicted results based on the dynamic GPR model and showed that the predictions are within the 95% confidence interval. This study presents a promising shallow and simple approach for predicting COVID-19 spread. [ABSTRACT FROM AUTHOR]

This study compares several advanced machine learning models to obtain the most accurate method for predicting the aeration efficiency (E20) through the Parshall flume. The required dataset is obtained from the laboratory tests using different flumes fabricated in National Institute Technology Kurukshetra, India. Besides, the potential of K Nearest Neighbor (KNN), Random Forest Regression (RFR), and Decision Tree Regression (DTR) models are evaluated to predict the aeration efficiency. In this way, several input combinations (e.g. M1-M15) are provided using the laboratory parameters (e.g. W/L, S/L, Fr, and Re). Different predictive models are obtained based on those input combinations and machine learning models proposed in the present study. The predictive models are assessed based on several performance metrics and visual indicators. Results show that the KNN-M11 model ( R M S E t e s t i n g = 0.002 , R t e s t i n g 2 = 0.929), which includes W/L, S/L, and Fr as predictive variables outperforms the other predictive models. Furthermore, an enhancement is observed in KNN model estimation accuracy compared to the previously developed empirical models. In general, the predictive model dominated in the present study provides adequate performance in predicting the aeration efficiency in the Parshall flume. [ABSTRACT FROM AUTHOR]

Aims: Cost-utility studies are crucial tools that help policy-makers promote appropriate resource allocation. The objective of this study was to evaluate the extent and quality of cost-utility analysis (CUA) in India through a systematic literature review. Methods: Comprehensive database search was conducted to identify the relevant literature published from November 2009 to November 2019. Gray literature and hand searches were also performed. Two researchers independently reviewed and assessed study quality using Consolidated Health Economic Evaluation Reporting Standards checklist. Results: Thirty-five studies were included in the final review. Thirteen studies used Markov model, five used decision tree model, four used a combination of decision tree and Markov model and one each used microsimulation and dynamic compartmental model. The primary therapeutic areas targeted in CUA were infectious diseases (n = 12), ophthalmology (n = 5), and endocrine disorders (n = 4). Five studies were carried out in Tamil Nadu, four in Goa, three in Punjab, two each in Delhi, Maharashtra, and Uttar Pradesh, and one each in West Bengal and Karnataka. Twenty-three, eight, and four studies were found to be of excellent, very good, and good quality, respectively. The average quality score of the studies was 19.21 out of 24. Conclusions: This systematic literature review identified the published CUA studies in India. The overall quality of the included studies was good; however, features such as subgroup analyses and explicit study perspective were missing in several evaluations. [ABSTRACT FROM AUTHOR]

Objective: To assess the cost‐effectiveness of decentralised diagnostic programme for hepatitis B virus (HBV) implemented in Tamil Nadu, South India with specific focus on a selected key population at increased risk of HBV. Methods: A combination of decision tree and Markov model was developed to compare cost‐effectiveness of the new and standard strategy. Cost and health outcomes were calculated based on the proportion of cohort in each respective health state. Total costs, quality‐adjusted life years (QALYs) and incremental cost‐effectiveness ratio (ICER) of the intervention and comparator strategies were calculated. The model parameter uncertainties were evaluated by sensitivity analysis. Results: Considering decentralised HBV diagnosis followed by early treatment and vaccination for negatives for a cohort of 1000 population resulted in 505 QALYs gained and incremental cost‐saving of 180749 ($2620). The decentralised diagnostic strategy could avert 294 deaths, gain 293 life years and reduce out‐of‐pocket expenditure of 3274 ($47) per person for HBV management. Conclusion: Decentralised HBV diagnosis followed by early treatment and vaccination for negatives in Tamil Nadu can save lives and reduce out‐of‐pocket expenditures compared to standard strategy. [ABSTRACT FROM AUTHOR]

In the drought prone district of Dholpur in Rajasthan, India, groundwater is a lifeline for its inhabitants. With population explosion and rapid urbanization, the groundwater is being critically over-exploited. Hence the current groundwater potential mapping study was undertaken to ascertain the areas that are more likely to yield a larger volume of groundwater against those areas that have poor groundwater potential and accordingly perpetuate the much needed damage control. Thematic layers for 14 groundwater influencing factors were considered for the study region, including elevation, slope, aspect, plan curvature, profile curvature, topographic wetness index (TWI), geology, soil, land use, normalized difference vegetation index (NDVI), surface temperature, precipitation, distance from roads, and distance from rivers. These were then subjected to an overlay operation, with the groundwater inventory which comprised of the locations of observational groundwater wells. The resulting geospatial database was then used to train two decision tree based ensemble models: gradient boosted decision trees (GBDT) and random forest (RF). The predictive performance of these models was then compared using various performance metrics such as area under curve (AUC) of receiver operating characteristics (ROC), sensitivity, accuracy, etc. It was found that GBDT (AUC: 0.79) outperformed RF (AUC: 0.71). The validated GBDT model was then used to construct the groundwater potential zonation map. The generated map showed that about 20.2% of the region has very high potential, while 22.6% has high potential to yield groundwater, and approximately 19.9–17.5% of the study region has very low to low groundwater potential. [ABSTRACT FROM AUTHOR]

Like several indigenous populations, Sauria Paharias, a vulnerable indigenous tribal group residing in a biodiverse environment of Jharkhand, India, have high levels of undernutrition. We assessed agroforestry and dietary diversity, food consumption especially indigenous food (IF) intake and nutritional status of Sauria Paharia women through a cross‐sectional study conducted in 18 villages of Godda district, Jharkhand. Household level information was elicited through household surveys including a dietary survey and a food frequency questionnaire. Twenty‐four‐hour dietary recalls (24 HDR) and anthropometric assessments were taken on one randomly selected woman per household. An index, Food Accessed Diversity Index (FADI) created to measure agroforestry diversity, showed a low mean score of 0.21 ± 0.15 and range: 0, 0.85. Fifty‐nine percent of women consumed any IF during 24 HDR. Median minimum dietary diversity score for women (MDD‐W) was 3 (acceptable score ≥5). More than 96% of women had intakes below estimated average requirements for all nutrients studied (energy; vitamins A, C, thiamine, riboflavin, niacin, pyridoxine; folate; iron; calcium and zinc) except protein; 41% women were underweight. IF consumption was independently associated with calcium and vitamin A intake. Decision trees developed for micronutrient consumption at different levels of MDD‐W score and IF consumption scenarios revealed 1.3 to 2.9 times higher consumption of micronutrients among women with MDD‐W ≥ 3 or 4. Strategies like agricultural extension programmes promoting indigenous varieties and nutrition education for increasing dietary diversity with IFs have potential to address undernutrition in Sauria Paharia women. [ABSTRACT FROM AUTHOR]

Background There is marked interindividual variability in metabolism and resulting toxicity and effectiveness of drugs used for tuberculosis treatment. For isoniazid, mutations in the N -acetyltransferase 2 (NAT2) gene explain >88% of pharmacokinetic variability. However, weight-based dosing remains the norm globally. The potential clinical impact and cost-effectiveness of pharmacogenomic-guided therapy (PGT) are unknown. Methods We constructed a decision tree model to project lifetime costs and benefits of isoniazid PGT for drug-susceptible tuberculosis in Brazil, South Africa, and India. PGT was modeled to reduce isoniazid toxicity among slow NAT2 acetylators and reduce treatment failure among rapid acetylators. The genotyping test was assumed to cost the same as the GeneXpert test. The main outcomes were costs (2018 US dollars), quality-adjusted life-years (QALYs), and incremental cost-effectiveness ratios. Results In Brazil, PGT gained 19 discounted life-years (23 QALYs) and cost $11 064 per 1000 patients, a value of $476 per QALY gained. In South Africa, PGT gained 15 life-years (19 QALYs) and cost $33 182 per 1000 patients, a value of $1780 per QALY gained. In India, PGT gained 20 life-years (24 QALYs) and cost $13 195 per 1000 patients, a value of $546 per QALY gained. One-way sensitivity analyses showed the cost-effectiveness to be robust to all input parameters. Probabilistic sensitivity analyses were below per capita gross domestic product in all 3 countries in 99% of simulations. Conclusions Isoniazid PGT improves health outcomes and would be cost-effective in the treatment of drug-susceptible tuberculosis in Brazil, South Africa, and India. [ABSTRACT FROM AUTHOR]

In this paper, we introduce two types of hybridization. The first contribution is the hybridization between the Viterbi algorithm and Baum Welch in order to predict crime locations. While the second contribution considers the optimization based on decision tree (DT) in combination with the Viterbi algorithm for criminal identification using Iraq and India crime dataset. This work is based on our previous work [1]. The main goal is to enhance the results of the model in both consuming times and to get a more accurate model. The obtained results proved the achievement of both goals in an efficient way. [ABSTRACT FROM AUTHOR]

Data analysis plays an important role in the health sector of any government in predicting the health conditions of their people and take measures to improve the health of their people. In this article, we present the problems faced by the department of health and family welfare in collecting and analyzing data from various directorates of the government of Tamilnadu in India. We then propose a framework that can collect data from various directorates of the government and analyze the data collected to predict the resources needed by the hospital to better serve their patients. Logistic regression and decision trees were used to analyze the data collected from various directorates. Further, the data are researched to identify potholes in the healthcare system to improve the existing system. In addition, we used five key performance indicator measures to evaluate the outcomes. [ABSTRACT FROM AUTHOR]

Indian coir industry is one among the small-scale cottage industries which contributes a great amount to the economy of India. India and Sri Lanka are the top two international producers, and they account for 90% of world coir manufacturing. The modern digital technologies can monitor the physical, chemical and microbiological properties in agriculture sector and can analyse the data in a faster manner. This paper highlights the large opportunities of deep learning in agriculture as an intelligent method for classifying and predicting the quality of the raw materials and thereby correlating into suitable products in the agricultural sector. Here we gathered the dataset from Central Coir Research Institute, Alappuzha, tested the chemical properties and applied several decision tree algorithms on the dataset and compared for analysis. This paper focuses specifically on a hybrid method with the usage of the back-propagation in deep learning and fuzzy logic decision tree. Experimental results reveal that the proposed method shows an accuracy of 98.75%. The comparison results with various decision tree-based classification algorithms, such as C4.5, ID3, CART, Naïve Bayes, fuzzy, MLP, verify the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Data mining offers strong techniques for different sectors involving education. In the education field the research is developing rapidly increasing due to huge number of student's information which can be used to invent valuable pattern pertaining learning behavior of students. The institutions of education can utilize educational data mining to examine the performance of students which can support the institution in recognizing the student's performance. In data mining classification is a familiar technique that has been implemented widely to find the performance of students. In this study a new prediction algorithm for evaluating student's performance in academia has been developed based on both classification and clustering techniques and been ested on a real time basis with student dataset of various academic disciplines of higher educational institutions in Kerala, India. The result proves that the hybrid algorithm combining clustering and classification approaches yields results that are far superior in terms of achieving accuracy in prediction of academic performance of the students. [ABSTRACT FROM AUTHOR]

India, being an abundant source of minerals, Mineral exploration, on a large scale, is promising to provide good impact for the future of the country. India has a rich source of coal, bauxite, limestone etc. With the advent of remote sensing technologies capturing broader area, exploration of minerals has now become an appreciable problem. Satellite Cluster Images spanning over wider areas can effectively serve as a tool for mapping minerals, qualitatively and quantitatively. In this paper, an algorithm, based on decision tree classification, is developed to map minerals, specifically, coal and limestone for a specific region. Multi-temporal cluster images are employed to map dynamic change detection resulting in greater accuracy. In this paper, multi-temporal cluster images (Landsat 8 OLI/TIRS) are analyzed to map coal, limestone and 'no-mineral' regions with the help of the algorithm developed using decision tree classification. Classification results are compared with ground truth data for assessing its accuracy. [ABSTRACT FROM AUTHOR]

Children in slums are at high risk of acute malnutrition and death. Cost-effectiveness of community-based management of severe acute malnutrition programmes has been demonstrated previously, but there is limited evidence in the context of urban slums where programme cost structure is likely to vary tremendously. This study assessed the cost-utility of adding a community based prevention and treatment for acute malnutrition intervention to Government of India Integrated Child Development Services (ICDS) standard care for children in Mumbai slums. The intervention is delivered by community health workers in collaboration with ICDS Anganwadi community health workers. The analysis used a decision tree model to compare the costs and effects of the two options: standard ICDS services with the intervention and prevention versus standard ICDS services alone. The model used outcome and cost data from the Society for Nutrition, Education & Health Action’s Child Health and Nutrition programme in Mumbai slums, which delivered services to 12,362 children over one year from 2013 to 2014. An activity-based cost model was used, with calculated costs based on programme financial records and key informant interviews. Cost data were coupled with programme effectiveness data to estimate disability adjusted life years (DALYs) averted. The community based prevention and treatment programme averted 15,016 DALYs (95% Uncertainty Interval [UI]: 12,246–17,843) at an estimated cost of $23 per DALY averted (95%UI:19–28) and was thus highly cost-effective. This study shows that ICDS Anganwadi community health workers can work efficiently with community health workers to increase the prevention and treatment coverage in slums in India and can lead to policy recommendations at the state, and potentially the national level, to promote such programmes in Indian slums as a cost-effective approach to tackling moderate and severe acute malnutrition. [ABSTRACT FROM AUTHOR]

Present study models the injury severity of acidents on non-urban sections of highways in India using multinomial logit model and two non-parametric models, namely, decision tree and random forest model. Dataset used consists of 2664 accidents with four severity levels and twenty five explanatory variables belonging to seven highways in state of Haryana, India. As there was an imbalance in crashes by injury level with the available dataset, two approaches for class balancing: synthetic minority oversampling (SMOTE) and randomise class balancing were used. Analysis of results suggests a significant improvement in predicting injury severity after class balancing. A comparison of results provided by used approaches suggests a superior performance by random forest classifier in terms of classification of accident severity levels. Another advantage of the use of random forest classifier is the availability of a ranked list of all input variables during classification of injury severity, which can be used to identify important variables required for classification of accident data in different severity levels. Analysis of results reveals that in order to reduce fatal and severe accidents, more attention needs to be provided to local and vulnerable road user traffic entering and crossing the highways in hazardous situations. The contribution of side swipe crashes and crashes with parked vehicles in severe outcomes was found to be another area of concern, requiring enforcement of lane discipline on highways and construction of lay-by bays as per the requirement of traffic. [ABSTRACT FROM AUTHOR]

One of the backbones of the Indian economy is agriculture, which is conditioned by the poor soil fertility. In this study we use chemical soil measurements to classify many relevant soil parameters: village-wise fertility indices of organic carbon (OC), phosphorus pentoxide ( P 2 O 5 ), manganese (Mn) and iron (Fe); soil pH and type; soil nutrients nitrous oxide ( N 2 O ), P 2 O 5 and potassium oxide ( K 2 O ), in order to recommend suitable amounts of fertilizers; and preferable crop. To classify these soil parameters allows to save time of specialized technicians developing expensive chemical analysis. These ten classification problems are solved using a collection of twenty very diverse classifiers, selected by their high performances, of families bagging, boosting, decision trees, nearest neighbors, neural networks, random forests (RF), rule based and support vector machines (SVM). The RF achieves the best performance for six of ten problems, overcoming 90% of the maximum performance in all the cases, followed by adaboost, SVM and Gaussian extreme learning machine. Although for some problems ( pH, N 2 O, P 2 O 5 and K 2 O ) the performance is moderate, some classifiers (e.g. for fertility indices of P 2 O 5 ,Mn and Fe) trained in one region revealed valid for other Indian regions. [ABSTRACT FROM AUTHOR]

Background Non-vaccinated children are a particularly vulnerable and understudied group. Machine learning algorithms, such as decision trees, might be useful for identifying subgroups with high prevalence of zero dose (neither BCG, polio, DPT nor measles vaccines received). Methods We developed Classification and Regression Tree models using data from DHS surveys of India 2015 and Chad 2014 in order to identify risk groups of zero dose. Results The first split variable for India was the child's place of delivery, followed by the mother's tetanus vaccination status for the higher-risk subgroup of children born in noninstitutional facilities. For Chad, administrative region was selected, and two high zero dose regions were defined. For those regions, children whose mother did not receive any dose of tetanus vaccine were also considered a higher-risk subgroup. Conclusions Two trees were created with only two splits each. Subgroups with zero dose prevalence higher than 40% were identified. Key messages Decision trees might be valuable tools for exploratory data analysis and risk groups identification in epidemiological research. 144 Figure 1 Open in new tab Download slide Decision trees and zero dose maps 144 Figure 1 Open in new tab Download slide Decision trees and zero dose maps [ABSTRACT FROM AUTHOR]

• A hybrid algorithm applying ST, HT and RBDT is designed for recognition of multiple PQ disturbances. • Designed HPQI and HPLI indices effectively detect and localize the CPQDs. • Proposed RBDT supported decision rules driven by features computed from HPQI and HPLI effectively classified the CPQDs. Classification accuracy higher than 98% is achieved. • Performance of algorithm is better compared to HT and RBDT based technique. • Algorithm is validated to recognize CPQDs incident on practical distribution network. Hence, it can be incorporated in online PQ monitoring devices. This paper proposes an algorithm based on the Stockwell transform (ST) and Hilbert transform (HT) to classify complex power quality disturbances (CPQDs). Two indices are used for detection and locating the CPQDs. The first index is the hybrid power quality index (HPQI), which is used to classify CPQDs. The Stockwell maximum values factor (SMVF), Stockwell summation values factor (SSVF), and Stockwell variance values factor (SVVF) are computed by utilizing the ST to process a voltage signal having a superimposed PQ disturbance. To compute the Hilbert index (HI), the voltage signal is also processed using the HT. The HPQI index is calculated by element-to-element multiplication of the SMVF, SSVF, SVVF, HI, and a weight factor (WF). The second index is the hybrid power quality disturbance time localization index (HPLI), which is used to locate the complex PQ events in a given time frame. For classification of investigated CPQDs, seven different features extracted from various indices defined using ST and HT are taken as input to the rule-based decision tree (RBDT). RBDT classifies the disturbances using simple decision rules, which has the merit of low computational time. The algorithm is able to correctly detect CPQDs with a 98% accuracy rate. The proposed algorithm outperforms the literature's algorithm, which is based on HT and RBDT. Furthermore, it is demonstrated that the algorithm effectively recognizes CPQDs that occur in real-time on a practical distribution network in Rajasthan State, India. [Display omitted] [ABSTRACT FROM AUTHOR]

The development of composite culture of carps in Purulia, an economically underdeveloped district of India, is confronted with several socio-economic, topographic, and climatic challenges. West Bengal Accelerated Development of Minor Irrigation Project (WBADMIP), an innovative integrated program of irrigation, agriculture, horticulture, and fisheries interventions through community institutions, was implemented in Purulia by the Government of West Bengal and was supported by the World Bank. Performances of the composite culture of carps in 172 water bodies by fisheries interest groups under the initiatives of WBADMIP were evaluated based on 18 different parameters using the k-mean algorithm. The results indicated that the water bodies could be categorized into three Clusters, based on inputs, expenditure, production, and income. Cluster 1, consisting of 85 water bodies, exhibited moderate performance with an average production of 2351 kg/ha, followed by Cluster 2 of 49 water bodies exhibiting poor performance with an average production of 1463 kg/ha, and Cluster 3 of 38 water bodies exhibiting best performance with an average production of 2488 kg/ha. Euclidean distance between Cluster centers 1 and 2 was 4.511 as compared to 3.737 and 5.102 between the Cluster centers 1 and 3 and 2 and 3, respectively. The water bodies with higher productivity were primarily associated with the WBADMIP policy that mobilized trained farmers who could adequately manage the application of feed, lime, inorganic fertilizers, and organic fertilizers. Using decision tree analysis, we predicted that the maintenance of water pH above 6.9 is crucial in obtaining optimum carps production from these water bodies. In contrast, the use of lime, inorganic fertilizers, organic fertilizers, and an adequate amount of feed need to be adjusted to maintain an economically viable production system. It is concluded that the farmers' motivation and professional training along with project support can substantially improve the productivity of the water bodies, support the nutrition of the farmers' community, and boost the local economy. • Purulia is a drought-prone underdeveloped district of India. • Aquaculture is a viable option to support the economy of the district. • Multiple criteria affect the success of aquaculture. • Clustering and decision tree analysis is used to identify critical factors. • The development of an institutional framework to coordinate farmers is vital. [ABSTRACT FROM AUTHOR]

Agriculture is one of the most important sectors in the Indian context. It is one of the highest employing sectors in the Indian scenario. Unlike other sectors agriculture is highly dependent on the quality and the quantity of both the external factors like rainfall, climate, pH of the soil, fertilizers and insecticides used, and internal factors like the quality of seeds. This paper predicts the production of crops as a function of rainfall for four Indian States. This knowledge can be implemented in generating a rough overview of how the production is based on rainfall and how much can a specific crop production for the amount of rainfall it receives. Two crops each belonging to four different states are chosen and the best regression model for the crop of the state is chosen. There is no research done solely on how rainfall affects crops of particular states. The proposed method of evaluation is better than other existing methods of evaluation as it evaluates all the regression techniques (Linear Regression, Polynomial Regression, Support Vector Regression, Decision Tree Regression, Random Forest, and XGBRegression) for two crops of four individual states. For balanced evaluation, two states of North India and two states of South India are selected. The regression techniques are evaluated based on their Mean Squared Error. • Crops of same species have different production rates within a same country. • Performance of each algorithm is different for crops within state. • Rainfall is the key determining factor for crop production in India. • Production of crop is affected by environment more than the genotype. • Predicting the production provides high impetus for developing economies. [ABSTRACT FROM AUTHOR]

We investigate the performance of existing state-of-the-art rule induction and tree algorithms, namely Single Conjunctive Rule Learner, Decision Table, M5 Model Tree, Decision Stump and REPTree. Downscaling models are developed using these algorithms to obtain projections of mean monthly precipitation to lake-basin scale in an arid region in India. The effectiveness of these algorithms is evaluated through application to downscale the predictand for the Lake Pichola region in Rajasthan state in India, which is considered to be a climatically sensitive region. The predictor variables are extracted from (1) the National Centre for Environmental Prediction (NCEP) reanalysis dataset for the period 1948-2000 and (2) the simulations from the third-generation Canadian Coupled Global Climate Model (CGCM3) for emission scenarios A1B, A2, B1 and COMMIT for the period 2001-2100. M5 Model Tree algorithm was found to yield better performance among all other learning techniques explored in the present study. The precipitation is projected to increase in future for A2 and AIB scenarios, whereas it is least for B1 and COMMIT scenarios using predictors. [ABSTRACT FROM AUTHOR]

Abstract: In this study we explored the potential of multi-temporal IRS P6 (Resourcesat) Advanced Wide Field Sensor (AWiFS) data for mapping of LULC for Delhi, India. The study presents the result of a decision tree classification of seasonal composite data (three seasons). The study has identified 13 classes with description of cropping pattern namely, double crops, kharif, rabi and zaid from 56m spatial resolution AWiFS data. Delhi has a diverse range of land use predominantly mosaic of built-up. More than half of the area is urban settlement. Results indicate that the temporal data set with a good definition of training sites can result in good overall accuracy (=91.81) as well as individual classification accuracies (producers accuracy ⩾76.92 and users accuracy ⩾60). It is evident that AWiFS data can be used to provide timely and detailed LULC maps with limited ancillary data. The AWiFS derived maps could be very useful as input to biogeochemical models that require timely estimation of LULC patterns. [ABSTRACT FROM AUTHOR]

Conventional statistical downscaling techniques for prediction of multi-site rainfall in a river basin fail to capture the correlation between multiple sites and thus are inadequate to model the variability of rainfall. The present study addresses this problem through representation of the pattern of multi-site rainfall using rainfall state in a river basin. A model based on K-means clustering technique coupled with a supervised data classification technique, namely Classification And Regression Tree (CART), is used for generation of rainfall states from large-scale atmospheric variables in a river basin. The K-means clustering is used to derive the daily rainfall state from the historical daily multi-site rainfall data. The optimum number of clusters in the observed rainfall data is obtained after application of various cluster validity measures to the clustered data. The CART model is then trained to establish relationship between the daily rainfall state of the river basin and the standardized, dimensionally-reduced National Centers for Environmental Prediction/National Center for Atmospheric Research (NCEP/NCAR) reanalysis climatic data set. The relationship thus developed is applied to the General Circulation Model (GCM)-simulated, standardized, bias free large-scale climate variables for prediction of rainfall states in future. Comparisons of the number of days falling under different rainfall states for the observed period and the future give the change expected in the river basin due to global warming. The methodology is tested for the Mahanadi river basin in India. [ABSTRACT FROM AUTHOR]

• About 21–36% wetland is identified as high to very highly vulnerable. • No significant increase of vulnerable area is recognized in last ten years. • Predictability is found high using machine learning over bivariate models. • REP tree is identified as the best suited as per accuracy assessment. • Water presence frequency and depth are found as the most sensitive factors of vulnerability. The present study aims to measure wetland habitat vulnerability (WHV) in moribund deltaic part of India using ten conditioning parameters e.g., WPF, water depth, change in WPF, change in depth, APF, seasonality, fragmentation, distance from river, distance from road, and settlement area. For developing suitable models integrating the data layers, four bivariate models namely, logistic regression (LR) frequency ratio (FR), Shannon entropy (SE), and weights of evidence (WoE), and four machine learning (ML) algorithms namely, artificial neural network (ANN), J48 decision trees (DTs), random forest (RF), and reduced error pruning (REP) tree have been used. Results reveal that the 20 km2 (11.13%) to 35.70 km2 (19.88%) and 18.43 km2 (10.27%) to 29.01 km2 (16.16%) of area to total wetland area has emerged as high and very high habitat vulnerable zones in phase II, whereas, 17.60 km2 (11.72%) to 31.10 km2 (20.75%) and 16.23 km2 (10.82%) to 28.61 km2 (19.07%) of area found as high to very high vulnerable in phase III in case of both bivariate and ML models. Accuracy assessment of using AUC (ROC), Kappa, and overall accuracy have confirmed the acceptability of all the models but result of machine learning based model is more precise than bivariate models. Frequency ratio from bivariate models and REP tree from machine learning models are found to be the most acceptable. Sensitivity analysis shows that WPF factor followed by depth of wetland are the most important contributing factors. So, application of machine learning model for vulnerability study is recommended. [ABSTRACT FROM AUTHOR]

This article investigates the classification of normal and COPD subjects on the basis of respiratory sound analysis using machine learning techniques. Thirty COPD and 25 healthy subject data are recorded. Total of 39 lung sound features and 3 spirometry features are extracted and evaluated. Various parametric and nonparametric tests are conducted to evaluate the relevance of extracted features. Classifiers such as support vector machine (SVM), k-nearest neighbor (KNN), logistic regression (LR), decision tree and discriminant analysis (DA) are used to categorize normal and COPD breath sounds. Classification based on spirometry parameters as well as respiratory sound parameters are assessed. Maximum classification accuracy of 83.6% is achieved by the SVM classifier while using the most relevant lung sound parameters i.e. median frequency and linear predictive coefficients. Further, SVM classifier and LR classifier achieved classification accuracy of 100% when relevant lung sound parameters, i.e. median frequency and linear predictive coefficient are combined with the spirometry parameters, i.e. forced vital capacity (FVC) and forced expiratory volume in 1 s (FEV1). It is concluded that combining lung sound based features with spirometry data can improve the accuracy of COPD diagnosis and hence the clinician's performance in routine clinical practice. The proposed approach is of great significance in a clinical scenario wherein it can be used to assist clinicians for automated COPD diagnosis. A complete handheld medical system can be developed in the future incorporating lung sounds for COPD diagnosis using machine learning techniques. [ABSTRACT FROM AUTHOR]

An attempt is made in this paper to report how a supervised methodology has been adopted for the task of Word Sense Disambiguation (WSD) in Bengali with necessary modifications. At the initial stage, four commonly used supervised methods, Decision Tree (DT), Support Vector Machine (SVM), Artificial Neural Network (ANN) and Naïve Bayes (NB), are developed at the baseline. These algorithms are applied individually on a data set of 13 most frequently used Bengali ambiguous words. On experimental basis, the baseline strategy is modified with two extensions: (a) inclusion of lemmatization process into the system and (b) bootstrapping of the operational process. As a result, the levels of accuracy of the baseline methods are slightly improved, which is a positive signal for the whole process of disambiguation as it opens scope for further modification of the existing method for better result. In this experiment, the data sets are prepared from the Bengali corpus, developed in the Technology Development for Indian Languages (TDIL) project of the Government of India and from the Bengali WordNet, which is developed at the Indian Statistical Institute, Kolkata. The paper reports the challenges and pitfalls of the work that have been closely observed during the experiment. [ABSTRACT FROM AUTHOR]

Background: A variety of mobile-based health technologies (mHealth) have been developed for use by community health workers to augment their performance. One such mHealth intervention—ReMiND program, was implemented in a poor performing district of India. Despite some research on the extent of its effectiveness, there is significant dearth of evidence on cost-effectiveness of such mHealth interventions. In this paper we evaluated the incremental cost per disability adjusted life year (DALY) averted as a result of ReMiND intervention as compared to routine maternal and child health programs without ReMiND. Methods: A decision tree was parameterized on MS-Excel spreadsheet to estimate the change in DALYs and cost as a result of implementing ReMiND intervention compared with routine care, from both health system and societal perspective. A time horizon of 10 years starting from base year of 2011 was considered appropriate to cover all costs and effects comprehensively. All costs, including those during start-up and implementation phase, besides other costs on the health system or households were estimated. Consequences were measured as part of an impact assessment study which used a quasi-experimental design. Proximal outputs in terms of changes in service coverage were modelled to estimate maternal and infant illnesses and deaths averted, and DALYs averted in Uttar Pradesh state of India. Probabilistic sensitivity analysis was undertaken to account for parameter uncertainties. Results: Cumulatively, from year 2011 to 2020, implementation of ReMiND intervention in UP would result in a reduction of 312 maternal and 149,468 neonatal deaths. This implies that ReMiND program led to a reduction of 0.2% maternal and 5.3% neonatal deaths. Overall, ReMiND is a cost saving intervention from societal perspective. From health system perspective, ReMiND incurs an incremental cost of INR 12,993 (USD 205) per DALY averted and INR 371,577 (USD 5865) per death averted. Conclusions: Overall, findings of our study suggest strongly that the mHealth intervention as part of ReMiND program is cost saving from a societal perspective and should be considered for replication elsewhere in other states. [ABSTRACT FROM AUTHOR]

Background: This paper explores the heterogeneities in antenatal care (ANC) utilization in India's most populated state, Uttar Pradesh. Taking an intersectionality lens, multiple individual- and district-level factors are used to identify segments of any antenatal care usage in Uttar Pradesh Objective: This paper seeks to understand the multilevel contexts of ANC utilization. The planning and programming challenge is that such knowledge of contextual specificity is rarely known upfront at the initial stages of planning or implementing an intervention. Exploratory data analysis might be needed to identify such contextual specificity. Methods: Tree-structured regression methods are used to identify segments and interactions between factors. The results from the tree-structured regression were complemented with multilevel models that controlled for the clustering of individuals within districts. Results: Heterogeneities in utilization of any ANC were observed. The multiple segments of ANC utilization that were developed went from a low utilization of 23.7% for respondents who were not literate and did not have home ownership to a high of 82.4% for respondents who were literate and at the highest level of wealth. Key variables that helped define the segments of ANC utilization include: woman's literacy, ownership of home, wealth index, and district-level sex ratio. Based on the multilevel model of any ANC utilization, cross-level interactions also were obtained between sex ratio and ownership of home as well as between sex ratio and literacy. Increases in sex ratio increased the influence of ownership of home on any ANC, while increases in sex ratio reduced the impact of woman's literacy on receiving any ANC. Conclusion: We argue that a focus on heterogeneous segments of utilization can help build knowledge of the mechanisms that underlie inequities in maternal health utilization. Such knowledge of heterogeneity needs to be incorporated in contextualizing interventions to meet a variety of recipients' needs. [ABSTRACT FROM AUTHOR]