Search

The semi-distributed approaches proposed in the past for simulating rainfall–runoff (RR) still require extensive hydro-meteorological data for their complex calibration process. Therefore, research efforts are needed to develop novel and innovative semi-distributed RR models that require a minimum amount of data and effort. In this study, four semi-distributed RR models are proposed using a simple lumped model in a distributed sense by gradually enforcing spatial distribution in terms of hydro-meteorological and physiographical features in a basin. Only rainfall, runoff, and temperature data from three contrasting basins were employed in developing the proposed models. The results show that semi-distributed models performed better than the lumped model, and the accuracy increased with gradual enforcement of the spatial variations in various data. Further, the results suggest that incorporation of the basin's physiographic features is the most important aspect in developing efficient semi-distributed RR models followed by hydrological and climatic information in a basin. [ABSTRACT FROM AUTHOR]

Network Intrusion detection performances are highly affected by imbalance data problems due to the presence of less number of attack information in the dataset. Deep learning models are applied in existing methods to improve the efficiency that has limitations of overfitting problems. The Generative Adversarial Network (GAN) -- Bayesian optimization Multi-class Support Vector Machine (BMSVM) is proposed to overcome imbalance and overfitting problems in intrusion detection systems. The Min-Max Normalization method is applied to normalize the input data to reduce the differences in features. GAN model is applied to generate minority class to balance the data instances to train the model. The proposed GAN-BMSVM model is compared with the classical sampling method, optimization, and classifier in the intrusion detection model in terms of Accuracy, Detection Rate (DR), and False Alarm Rate (FAR). The classical sampling methods are Near-miss, SMOTE and Autoencoder; traditional classifiers are KNN, RF, SVM, DNN and LSTM, and classical optimizations are PSO and WOA. The existing researches such as HCRNN, HLD, DONN, FL-NIDS and CNN-LSTM are used to evaluate the efficiency of GAN-BMSVM model. The GAN-BMSVM model has achieved 99.58% and 85.38 % accuracy for NSL-KDD and UNSW-NB15 dataset respectively, which is higher than the existing CNN-LSTM model. [ABSTRACT FROM AUTHOR]

Artificial neural network (ANN) and regression models are developed for the estimation of concrete slump using concrete constituent data. The concrete mix constituent and slump data from laboratory tests have been employed to develop all models. The results obtained in this study demonstrate the superiority of the ANN models. It was found that combining one or more concrete mix constituents and treating them as an independent input variable is not advantageous when using regression but can be very useful when using ANNs for modeling concrete slump. Sensitivity analyses based on the ANN models were carried out to evaluate the impact of different concrete mix constituents on the slump values. It was found that the slump attains a minimum value at the critical levels of mortar and coarse aggregates, and tends to increase with paste content and decrease with sand content in the concrete mix. DOI: 10.1061/(ASCE)0899-1561(2008)20:9(628) CE Database subject headings: Artificial networks; Neural networks; Concrete; Aggregates; Estimation; Sensitivity analysis; Regression analysis.

The temporal and spatial characterization of unknown groundwater pollution sources remains an important problem in effective aquifer remediation and assessment of associated health risks. The characterization of contaminated source involves identifying spatially and temporally varying source locations, injection rates, and release periods. The proposed methodology exploits the universal function approximation capability of a feed forward multilayer artificial neural network (ANN) to identify the unknown pollution sources. The ANN is trained to identify source characteristics based on simulated contaminant concentration measurement data at specified observation locations in the aquifer. These concentrations are simulated for a large set of randomly generated pollution source fluxes. The back-propagation algorithm is used for training the ANN, with each corresponding set of source fluxes and resulting concentration measurement constituting a pattern for training the ANN. Performance of this methodology is evaluated for various data availability, measurement error, and source location scenarios. The developed ANNs are capable of identifying unknown groundwater pollution sources at multiple locations using erroneous measurement data. CE Database subject headings: Ground-water pollution; Pollution sources; Neural networks; Data analysis.

In the past, studies involving optimal design of composite channels have employed Horton's equivalent roughness coefficient, which uses a lumped approach in assuming constant velocity across a composite channel cross section. In this paper, a new nonlinear optimization program (NLOP) is proposed based on a distributed approach that is equivalent to Lotter's observations, which allows spatial variations in velocity across a composite channel cross section. The proposed NLOP, which consists of an objective function of minimizing total construction cost per unit length of a channel, is solved using genetic algorithm (GA). Several scenarios are evaluated, including no restrictions, restricted top width, and restricted channel side slopes, to account for certain site conditions. In addition, the proposed NLOP is modified to include constraints on maximum permissible velocities corresponding to different lining materials of the composite channel cross section, probably for the first time. The proposed methodology is applied to trapezoidal and triangular channel cross sections but can be easily extended to other shapes or compound channels. Optimal design graphs are presented to determine the channel dimensions of a composite trapezoidal channel cross section. The results obtained in this study indicate that cost savings up to 35% can be achieved for the unconstrained velocity case and up to 55% for the limiting velocity case when the proposed NLOP is solved using GA as compared with the existing NLOP solved using either the classical optimization solution technique or GA. CE Database subject headings: Channel design; Optimization; Algorithms; Composite structures.

In cervical spine, the anatomical course of medial branch in significantly different and thermal radiofrequency ablation (RFA) of the cervical median branch though proven effective is technically challenging. Cooled RFA (CRFA) is a newer revolutionary technique that creates a larger spherical lesion and thus compensates for the anatomical variability of the medial branches in the cervical spine. Our case report is aimed to discuss the technique of CRFA for the treatment of cervical facetogenic pain. Right C2–C5 medial branch CRFA was done under fluoroscopic guidance under monitored anesthesia care using 17G with 75 mm length and 2 mm active tip cooled radiofrequency cannula. After 8 weeks of follow-up, the patient reported 80% pain relief. [ABSTRACT FROM AUTHOR]

While India battles COVID-19, hand washing is the survival strategy. The research objective is to critically examine water rights in India. The research asks whether there are rights to water in India, and if so, what kind. The thesis proposition is that India has derivative rights to water that are neither fundamental nor human rights. The research methodology is by narrative analysis. States prefer an economic good theory, rom the Dublin Statement. Fundamental rights to ground water in India are unlikely to be shifted. The India Constitution now provides additional rights. Human rights for water are only derivative and undeclared. [ABSTRACT FROM AUTHOR]

Fast Moving Consumer Goods (FMCG), are products that are sold quickly at relatively low cost. Though the absolute profit made on FMCG products is relatively small, they generally sell in large quantities, so the cumulative profit on such products can be large. They try to achieve this by segmenting the whole market into segment markets which has a number of customers with common needs and buying motives, and then developing solutions that appealed particularly strongly to those segments. The best segments must have Potential, Lifespan, Accessibility, and Profitability. The key is identifying which segments provide value in terms of potential, lifespan, accessibility and profitability; because a sales strategy's effectiveness increases according to our capacity to size segments, identify them, and dissect them. How to segment a market is always an issue in the FMCG marketing? One of the bases of segmentation is to base segment on the geographical location of the customer. However, there are number of factors that affect the customer choice of location, that describe why the customer has chosen a particular location for his purchases. This research paper endeavour to describe the relative importance of the factors affecting the geographical segmentation in FMCGs. [ABSTRACT FROM AUTHOR]

The massive unstructured and semi structured heterogeneous data generated from devices, household appliances and from day to day activities namely sensors used to gather climate information, posts to social media sites, digital pictures and videos, purchase transaction records, and cell phone GPS signals to name a few collected together is what we call as big data. Machine learning's supervised and unsupervised learning techniques can be used to process this large amount of data using the various classifiers to derive useful insight from the data and to predict the future patterns and trends. Supervised learning techniques uses the concept of train and test data i.e., some portion of data is first trained with expected results for a given input and then the remaining chunk of data is used to test the algorithm for prediction accuracy. J48, IBK & Naïve Bayes are few of the supervised learning classifiers used in this comparative analysis. The classifiers are used on the multivariate big data set selected under different models of training and test data to obtain the best performing classifier among them on basis of correctly classified instances (one of the performance measuring criteria). On the basis of the experiments conducted, J48 is the classifier which is generating the maximum number of correctly classified instances. Hence it can be considered as the best classifier to process the data of given kind. [ABSTRACT FROM AUTHOR]

Software maintenance is one of the important stages of the software development life cycle (SDLC) which needs to be performed carefully to reduce the cost and the effort associated with the software. Software maintainability is the ability to perform the changes in the software without the recurring bugs. If we are able to predict which classes of the object oriented (OO) software need maintenance, it will be very helpful to turn down the cost and the manpower required by the software. In this article, dynamic metrics of the software are used as the internal attributes and the number of changes made from one version of software to another, is taken as the external attribute for the prediction of software maintainability. Dynamic metrics capture more information as they are collected during the execution of the software, hence they are more accurate. For the prediction purpose, machine learning algorithms have been used. Out of the algorithms used, decision table algorithm performs the best in terms of MAE and RMSE. [ABSTRACT FROM AUTHOR]

Image classification is one of the techniques for abnormal state semantic analysis. There are numerous procedures for semantic analysis for instance - image data mining, learning revelation and shrewd data analysis. An image by and large contains numerous semantic objects, so it's not satisfactory to apply image classification as per the worldwide highlights. Subsequently, image division ought to be performed before image classification. Be that as it may, exact semantic division is still out of certain semantic protest regions in a single image, as one semantic question with unmistakable regions would be fragmented into a few superfluous regions by current most image segmentation algorithms in light of low level shading or surface highlights, i.e. the various hierarchical clustering based approach (HCBA), the Eigen regions based system (EBS) and the novel regions based strategies (NRBM). This outcomes in the loss of productivity for these locale based image classification plans, as a bigger number of understood regions will be accomplished. In the interim, these individual regions are for the most part significant and surprising outcome would be caused by these classification systems, albeit some salient regions are extricated to diminish the unpredictability of district based image classifications. Moreover, these plans are the subjective classification schemes, and couldn't show the data about other certain objects. In this venture, we propose salient regions based fluffy classification (SrFC) techniques. Right off the bat we give a salient area extraction calculation in view of prevailing hues and Gabor surface highlights. At that point, as per the proposed classification conspire the salient regions from all the hopeful images are grouped into disjoint classes. These classifications are additionally utilized with respect to those applicant images, and for every hopeful image its number of salient regions having a place with a similar classification would be utilized to assess the level of the images having a place with the class. With correlation with the HCBA plot for singular salient regions, the classification exactness is enhanced by around 16% by the proposed SrFC scheme. [ABSTRACT FROM AUTHOR]