arXiv subjects

Subrata Chakraborty

Publications and source records attributed to Subrata Chakraborty.

At least 37 records · Page 2Linked to original sources

arXiv2021

Modeling proportion of success in high school leaving examination- A comparative study of Inflated Unit Lindley and Inflated Beta distribution

In this article, we first introduced the inflated unit Lindley distribution considering zero or/and one inflation scenario and studied its basic distributional and structural properties. Both the distributions are shown to be members of exponential family with full rank. Different parameter estimation methods are discussed and supporting simulation studies to check their efficacy are also presented. Proportion of students passing the high school leaving examination for the schools across the state of Manipur in India for the year 2020 are then modeled using the proposed distributions and compared with the inflated beta distribution to justify its benefits.

stat.ME↗

arXiv2021

MAVIDH Score: A COVID-19 Severity Scoring using Chest X-Ray Pathology Features

The application of computer vision for COVID-19 diagnosis is complex and challenging, given the risks associated with patient misclassifications. Arguably, the primary value of medical imaging for COVID-19 lies rather on patient prognosis. Radiological images can guide physicians assessing the severity of the disease, and a series of images from the same patient at different stages can help to gauge disease progression. Hence, a simple method based on lung-pathology interpretable features for scoring disease severity from Chest X-rays is proposed here. As the primary contribution, this method correlates well to patient severity in different stages of disease progression with competitive results compared to other existing, more complex methods. An original data selection approach is also proposed, allowing the simple model to learn the severity-related features. It is hypothesized that the resulting competitive performance presented here is related to the method being feature-based rather than reliant on lung involvement or opacity as others in the literature. A second contribution comes from the validation of the results, conceptualized as the scoring of patients groups from different stages of the disease. Besides performing such validation on an independent data set, the results were also compared with other proposed scoring methods in the literature. The results show that there is a significant correlation between the scoring system (MAVIDH) and patient outcome, which could potentially help physicians rating and following disease progression in COVID-19 patients.

eess.IV↗

arXiv2021

Potential Features of ICU Admission in X-ray Images of COVID-19 Patients

X-ray images may present non-trivial features with predictive information of patients that develop severe symptoms of COVID-19. If true, this hypothesis may have practical value in allocating resources to particular patients while using a relatively inexpensive imaging technique. The difficulty of testing such a hypothesis comes from the need for large sets of labelled data, which need to be well-annotated and should contemplate the post-imaging severity outcome. This paper presents an original methodology for extracting semantic features that correlate to severity from a data set with patient ICU admission labels through interpretable models. The methodology employs a neural network trained to recognise lung pathologies to extract the semantic features, which are then analysed with low-complexity models to limit overfitting while increasing interpretability. This analysis points out that only a few features explain most of the variance between patients that developed severe symptoms. When applied to an unrelated larger data set with pathology-related clinical notes, the method has shown to be capable of selecting images for the learned features, which could translate some information about their common locations in the lung. Besides attesting separability on patients that eventually develop severe symptoms, the proposed methods represent a statistical approach highlighting the importance of features related to ICU admission that may have been only qualitatively reported. While handling limited data sets, notable methodological aspects are adopted, such as presenting a state-of-the-art lung segmentation network and the use of low-complexity models to avoid overfitting. The code for methodology and experiments is also available.

eess.IV↗

arXiv2021

A new method for constructing continuous distributions on the unit interval

A novel approach towards construction of absolutely continuous distributions over the unit interval is proposed. Considering two absolutely continuous random variables with positive support, this method conditions on their convolution to generate a new random variable in the unit interval. This approach is demonstrated using some popular choices of the positive random variables such as the exponential, Lindley, gamma. Some existing distributions like the uniform and the beta are formulated with this method. Several new structures of density functions having potential for future application in real life problems are also provided. One of the new distributions having one parameter is considered for parameter estimation and real life modelling application and shown to provide better fit than the popular one parameter Topp-Leone model.

math.ST↗

arXiv2020

A new generalization of the geometric distribution using Azzalini's mechanism: properties and application

The skewing mechanism of Azzalini for continuous distributions is used for the first time to derive a new generalization of the geometric distribution. Various structural properties of the proposed distribution are investigated. Characterizations, including a new result for the geometric distribution, in terms of the proposed model are established. Extensive simulation experiment is done to evaluate performance of the maximum likelihood estimation method. Likelihood ratio test for the necessity of additional skewing parameter is derived and corresponding simulation based power study is also reported. Two real life count datasets are analyzed with the proposed model and compared with some recently introduced two-parameter count models. The findings clearly indicate the superiority of the proposed model over the existing ones in modelling real life count data.

math.ST↗

arXiv2020

Changing Clusters of Indian States with respect to number of Cases of COVID-19 using incrementalKMN Method

The novel Coronavirus (COVID-19) incidence in India is currently experiencing exponential rise but with apparent spatial variation in growth rate and doubling time rate. We classify the states into five clusters with low to the high-risk category and study how the different states moved from one cluster to the other since the onset of the first case on $30^{th}$ January 2020 till the end of unlock 1 that is $30^{th}$ June 2020. We have implemented a new clustering technique called the incrementalKMN (Prasad, R. K., Sarmah, R., Chakraborty, S.(2019))

q-bio.PE↗

arXiv2020

On a family that unifies Generalized Marshall-Olkin and Poisson-G family of distribution

Unifying the generalized Marshall-Olkin (GMO) and Poisson-G (P-G) a new family of distribution is proposed. Density and the survival function are expressed as infinite mixtures of P-G family. The quantile function, asymptotes, shapes, stochastic ordering, moment generating function, order statistics, probability weighted moments and Rényi entropy are derived. Maximum likelihood estimation with large sample properties is presented. A Monte Carlo simulation is used to examine the pattern of the bias and the mean square error of the maximum likelihood estimators. An illustration of comparison with some of the important sub models of the family in modeling a real data reveals the utility of the proposed family.

math.ST↗

arXiv2020

Beta Poisson-G Family of Distributions: Its Properties and Application with Failure Time Data

A new generalization of the family of Poisson-G is called beta Poisson-G family of distribution. Useful expansions of the probability density function and the cumulative distribution function of the proposed family are derived and seen as infinite mixtures of the Poisson-G distribution. Moment generating function, power moments, entropy, quantile function, skewness and kurtosis are investigated. Numerical computation of moments, skewness, kurtosis and entropy are tabulated for select parameter values. Furthermore, estimation by methods of maximum likelihood is discussed. A simulation study is carried at under varying sample size to assess the performance of this model. Finally suitability check of the proposed model in comparison to its recently introduced models is carried out by considering two real life data sets modeling.

math.ST↗

arXiv2020

The Poisson Transmuted-G Family of Distributions: Its Properties and Applications

In this paper introduces a new family of continuous distributions namely the Poison transmuted-G family of distribution is proposed by inducing two addition parameter on the base line G distribution. Some of its mathematical properties including explicit expressions for the moments generating function, order statistics, Probability weighted moments, stress-strength reliability, residual life, reversed residual life, Rényi entropy and mean deviation are derived. Some special models of the new family are listed. Estimation of the model parameters by the maximum likelihood method is discussed. The advantage of the proposed family in data fitting is illustrated by means of two applications to failure time data set.

math.ST↗

arXiv2020

A Practical Blockchain Framework using Image Hashing for Image Authentication

Blockchain is a relatively new technology that can be seen as a decentralised database. Blockchain systems heavily rely on cryptographic hash functions to store their data, which makes it difficult to tamper with any data stored in the system. A topic that was researched along with blockchain is image authentication. Image authentication focuses on investigating and maintaining the integrity of images. As a blockchain system can be useful for maintaining data integrity, image authentication has the potential to be enhanced by blockchain. There are many techniques that can be used to authenticate images; the technique investigated by this work is image hashing. Image hashing is a technique used to calculate how similar two different images are. This is done by converting the images into hashes and then comparing them using a distance formula. To investigate the topic, an experiment involving a simulated blockchain was created. The blockchain acted as a database for images. This blockchain was made up of devices which contained their own unique image hashing algorithms. The blockchain was tested by creating modified copies of the images contained in the database, and then submitting them to the blockchain to see if it will return the original image. Through this experiment it was discovered that it is plausible to create an image authentication system using blockchain and image hashing. However, the design proposed by this work requires refinement, as it appears to struggle in some situations. This work shows that blockchain can be a suitable approach for authenticating images, particularly via image hashing. Other observations include that using multiple image hash algorithms at the same time can increase performance in some cases, as well as that each type of test done to the blockchain has its own unique pattern to its data.

cs.CV↗

arXiv2019

Charge transport through spin-polarized tunnel junction between two spin-split superconductors

We investigate transport properties of junctions between two spin-split superconductors linked by a spin-polarized tunneling barrier. The spin-splitting fields in the superconductors (S) are induced by adjacent ferromagnetic insulating (FI) layers with arbitrary magnetization. The aim of this study is twofold: On the one hand, we present a theoretical framework based on the quasiclassical Green's functions to calculate the Josephson and quasiparticle current through the junctions in terms of the different parameters characterizing it. Our theory predicts qualitative new results for the tunneling differential conductance, $dI/dV$, when the spin-splitting fields of the two superconductors are non-collinear. We also discuss how junctions based on FI/S can be used to realize anomalous Josephson junctions with a constant geometric phase shift in the current-phase relation. As a result, they may exhibit spontaneous triplet supercurrents in the absence of a phase difference between the S electrodes. On the other hand, we show results of planar tunneling spectroscopy of a EuS/Al/Al$_2$O$_3$/EuS/Al junction and use our theoretical model to reproduce the obtained $dI/dV$ curves. Comparison between theory and experiment reveals information about the intrinsic parameters of the junction, such as the size of the superconducting order parameter, spin-splitting fields and spin relaxation, and also about properties of the two EuS films, as their morphology, domain structure, and magnetic anisotropy.

cond-mat.supr-con↗

arXiv2019

The Generalized-Alpha-Beta-Skew-Normal Distribution: Properties and Applications

In this paper we have introduced a generalized version of alpha beta skew normal distribution in the same line of Sharafi et al. (2017) and investigated some of its basic properties. The extensions of the proposed distribution have also been studied. The appropriateness of the proposed distribution has been tested by comparing the values of Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) with the values of some other known related distributions for better model selection. Likelihood ratio test has been used for discriminating between nested models.

math.ST↗

arXiv2019

Selection of link function in binary regression: A case-study with world happiness report on immigration

Selection of appropriate link function for binary regression remains an important issue for data analysis and its influence on related inference. We prescribe a new data-driven methodology to search for the same, considering some popular classification assessment metrics. A case-study with World Happiness report,2018 with special reference to immigration is presented for demonstrating utility of the prescribed routine.

stat.AP↗

arXiv2019

The Balakrishnan Alpha Skew Laplace Distribution: Properties and Its Applications

In this study by considering Balakrishnan mechanism a new form of alpha skew distribution is proposed and properties are investigated. The suitability of the proposed distribution has tested at the end with appropriate data fitting experiment and then comparing the values of Akaike Information Criterion (AIC), and Bayesian Information Criterion (BIC) with the values of some other related distributions. Likelihood ratio test is carried out for nested models, that is, for Laplace and the proposed distributions. Keywords: Skew Distribution, Alpha-Skew Distribution, Bimodal Distribution, Alpha-Skew-Normal Distribution, Alpha-Skew-Laplace Distribution, Balakrishnan-Alpha-Skew-Normal Distribution

math.ST↗

arXiv2019

Further study on inferential aspects of log-Lindley distribution with an application of stress-strength reliability in insurance

The log-Lindley distribution was recently introduced in the literature as a viable alternative to the Beta distribution. This distribution has a simple structure and possesses useful theoretical properties relevant in insurance. Classical estimation methods have been well studied. We introduce estimation of parameters from Bayesian point of view for this distribution. Explicit structure of stress-strength reliability and its inference under both classical and Bayesian set-up is addressed. Extensive simulation studies show marked improvement with Bayesian approach over classical given reasonable prior information. An application of a useful metric of discrepancy derived from stress-strength reliability is considered and computed for two categories of firm with respect to a certain financial indicator.

math.ST↗

arXiv2019

R = P(Y < X) for unit-Lindley distribution: inference with an application in public health

The unit-Lindley distribution was recently introduced in the literature as a viable alternative to the Beta and the Kumaraswamy distributions with support in (0; 1). This distribution enjoys many virtuous properties over the named distributions. In this article, we address the issue of parameter estimation from a Bayesian perspective and study relative performance of different estimators through extensive simulation studies. Significant emphasis is given to the estimation of stress-strength reliability employing classical as well as Bayesian approach. A non-trivial useful application in the public health domain is presented proposing a simple metric of discrepancy.

stat.ME↗

arXiv2019

Thermalization of hot electrons via interfacial electron-magnon interaction

Recent work on layered structures of superconductors (S) or normal metals (N) in contact with ferromagnetic insulators (FI) has shown how the properties of the previous can be strongly affected by the magnetic proximity effect due to the static FI magnetization. Here we show that such structures can also exhibit a new electron thermalization mechanism due to the coupling of electrons with the dynamic magnetization, i.e., magnons in FI. We here study the heat flow between the two systems and find that in thin films the heat conductance due to the interfacial electron-magnon collisions can dominate over the well-known electron-phonon coupling below a certain characteristic temperature that can be straightforwardly reached with present-day experiments. We also study the role of the magnon band gap and the induced spin-splitting field induced in S on the resulting heat conductance and show that heat balance experiments can reveal information about such quantities in a way quite different from typical magnon spectroscopy experiments.

cond-mat.mes-hall↗

arXiv2018

Thermoelectric radiation detector based on a superconductor-ferromagnet junction: calorimetric regime

We study the use of a thermoelectric junction as a thermal radiation detector in the calorimetric regime, where single radiation bursts can be separated in time domain. We focus especially on the case of a large thermoelectric figure of merit $ZT$ affecting significantly for example the relevant thermal time scales. This work is motivated by the use of hybrid superconductor/ferromagnet systems in creating an unprecedentedly high low-temperature $ZT$ even exceeding unity. Besides constructing a very general noise model which takes into account cross correlations between charge and heat noise, we show how the detector signal can be efficiently multiplexed by the use of resonant LC circuits giving a fingerprint to each pixel. We show that for realistic detectors operating at temperatures around 100 to 200 mK, the energy resolution can be as low as 1 meV. This allows for a broadband single-photon resolution at photon frequencies of the order or below 1 THz.

cond-mat.mes-hall↗