This article focuses on solving parametric transmission problems in one and two spatial dimensions. These problems belong to a class of partial differential equations that arise in the modeling of physical systems with heterogeneous materials. They often exhibit discontinuities across interfaces and singularities at points where interfaces intersect. To address these problems, we propose a new deep learning approach named {\it{Least-Squares-Based Regularity-Conforming Neural Network (LS-ReCoNN)}}. This approach proposes a loss function that is shown to be a consistent upper bound for the energy-norm error. The method represents the solution as the sum of a principal component and a singular component. The principal component is decomposed into smooth and gradient-jump parts, which capture both the regular solution behavior and reduced regularity across interfaces in one- and two-dimensional problems. The singular component is introduced to model junction singularities and it is approximated using basis functions computed from a one-dimensional finite element eigenvalue problem. For the principal component, a separated representation is employed, consisting of parameter-dependent co
The National Academic Depository of India is a distinctive, novel and progressive step visualized by Ministry of Human Resources Development, Govt. of India towards maintaining a database to hold the academic awards issued by Educational Institutions in an electronic and digital form. NAD promises to abolish the difficulties / inefficiencies of collecting, maintaining, and presenting physical paper certificates that can be easily copied / created and the verification processes which are costly, time consuming and disorganized. The depository can eradicate the need to store academic awards in physical form. It can verify the awards issued by different Institutions to the students in an easy way. The secure digital depository is a good proposal to do away with fake and forged certificates. The concept of academic depository is identical to the concept of financial securities. The pilot project is successfully completed with the help of Central Board of Secondary Education and some universities. In order to become fully functional, the depository has to conquer a few challenges with respect to academic diversities in terms of duration of courses and equivalence. National Academic Depo
We study classes of objects whose combinatorics are closely related to those of posets. The framework of operads and operad algebras allows us to make this relationship precise and provides tools for a deeper understanding of their combinatorial structure. In this note, we present a nontrivial example of a suboperad of the operad of posets, called Wixárika posets, together with its associated algebras. This example is sufficiently rich to exhibit key structural features of the theory, while remaining accessible and avoiding unnecessary technicalities.
While strides have been made in deep learning based Bengali Optical Character Recognition (OCR) in the past decade, the absence of large Document Layout Analysis (DLA) datasets has hindered the application of OCR in document transcription, e.g., transcribing historical documents and newspapers. Moreover, rule-based DLA systems that are currently being employed in practice are not robust to domain variations and out-of-distribution layouts. To this end, we present the first multidomain large Bengali Document Layout Analysis Dataset: BaDLAD. This dataset contains 33,695 human annotated document samples from six domains - i) books and magazines, ii) public domain govt. documents, iii) liberation war documents, iv) newspapers, v) historical newspapers, and vi) property deeds, with 710K polygon annotations for four unit types: text-box, paragraph, image, and table. Through preliminary experiments benchmarking the performance of existing state-of-the-art deep learning architectures for English DLA, we demonstrate the efficacy of our dataset in training deep learning based Bengali document digitization models.
In this work, we present our deployment-ready Speech-to-Speech Machine Translation (SSMT) system for English-Hindi, English-Marathi, and Hindi-Marathi language pairs. We develop the SSMT system by cascading Automatic Speech Recognition (ASR), Disfluency Correction (DC), Machine Translation (MT), and Text-to-Speech Synthesis (TTS) models. We discuss the challenges faced during the research and development stage and the scalable deployment of the SSMT system as a publicly accessible web service. On the MT part of the pipeline too, we create a Text-to-Text Machine Translation (TTMT) service in all six translation directions involving English, Hindi, and Marathi. To mitigate data scarcity, we develop a LaBSE-based corpus filtering tool to select high-quality parallel sentences from a noisy pseudo-parallel corpus for training the TTMT system. All the data used for training the SSMT and TTMT systems and the best models are being made publicly available. Users of our system are (a) Govt. of India in the context of its new education policy (NEP), (b) tourists who criss-cross the multilingual landscape of India, (c) Indian Judiciary where a leading cause of the pendency of cases (to the ord
Stock market prediction has been an active area of research for a considerable period. Arrival of computing, followed by Machine Learning has upgraded the speed of research as well as opened new avenues. As part of this research study, we aimed to predict the future stock movement of shares using the historical prices aided with availability of sentiment data. Two models were used as part of the exercise, LSTM was the first model with historical prices as the independent variable. Sentiment Analysis captured using Intensity Analyzer was used as the major parameter for Random Forest Model used for the second part, some macro parameters like Gold, Oil prices, USD exchange rate and Indian Govt. Securities yields were also added to the model for improved accuracy of the model. As the end product, prices of 4 stocks viz. Reliance, HDFC Bank, TCS and SBI were predicted using the aforementioned two models. The results were evaluated using RMSE metric.
Finding the semantically accurate answer is one of the key challenges in advanced searching. In contrast to keyword-based searching, the meaning of a question or query is important here and answers are ranked according to relevance. It is very natural that there is almost no common word between the question sentence and the answer sentence. In this paper, an approach is described to find out the semantically relevant answers in the Bengali dataset. In the first part of the algorithm, a set of statistical parameters like frequency, index, part-of-speech (POS), etc. is matched between a question and the probable answers. In the second phase, entropy and similarity are calculated in different modules. Finally, a sense score is generated to rank the answers. The algorithm is tested on a repository containing a total of 275000 sentences. This Bengali repository is a product of Technology Development for Indian Languages (TDIL) project sponsored by Govt. of India and provided by the Language Research Unit of Indian Statistical Institute, Kolkata. The shallow parser, developed by the LTRC group of IIIT Hyderabad is used for POS tagging. The actual answer is ranked as 1st in 82.3% cases. T
Recent advances in grazing incidence X-ray optics using synchrotron radiation sources have stimulated the need for basic research in high quality mirror materials for novel applications. In this paper we communicate the results of the first measurements of glazing angle X-ray reflectivity (XRR) of speculum metal mirrors using synchrotron radiation sources. Our results agree with similar measurements made in polished speculum gratings by Arthur Compton and his collaborators during 1923 using ordinary X-ray tubes. Our experimental investigations are based on synchrotron radiation research facilities (Indus 2) maintained by Govt of India in Indore . Variations in the XRR with grazing angles of incidence and X-ray energy for cast, thin film and electron irradiated samples of speculum metal will be discussed in this paper.
Maps are used to describe far-off places . It is an aid for navigation and military strategies. Mapping of the lands are important and the mapping work is based on (i). Natural resource management & development (ii). Information technology ,(iii). Environmental development ,(iv). Facility management and (v). e-governance. The Landuse / Landcover system espoused by almost all Organisations and scientists, engineers and remote sensing community who are involved in mapping of earth surface features, is a system which is derived from the united States Geological Survey (USGS) LULC classification system. The application of RS and GIS involves influential of homogeneous zones, drift analysis of land use integration of new area changes or change detection etc.,National Remote Sensing Agency(NRSA) Govt. of India has devised a generalized LULC classification system respect to the Indian conditions based on the various categories of Earth surface features , resolution of available satellite data, capabilities of sensors and present and future applications. The profusion information of the earth surface offered by the high resolution satellite images for remote sensing applications. Using
The ongoing novel coronavirus epidemic has been announced a pandemic by the World Health Organization on March 11, 2020, and the Govt. of India has declared a nationwide lockdown from March 25, 2020, to prevent community transmission of COVID-19. Due to absence of specific antivirals or vaccine, mathematical modeling play an important role to better understand the disease dynamics and designing strategies to control rapidly spreading infectious diseases. In our study, we developed a new compartmental model that explains the transmission dynamics of COVID-19. We calibrated our proposed model with daily COVID-19 data for the four Indian provinces, namely Jharkhand, Gujarat, Andhra Pradesh, and Chandigarh. We study the qualitative properties of the model including feasible equilibria and their stability with respect to the basic reproduction number $\mathcal{R}_0$. The disease-free equilibrium becomes stable and the endemic equilibrium becomes unstable when the recovery rate of infected individuals increased but if the disease transmission rate remains higher then the endemic equilibrium always remain stable. For the estimated model parameters, $\mathcal{R}_0 >1$ for all the four p
In this research work, we have proposed an algorithm based on supervised learning methodology to extract the root forms of the Bengali verbs using the grammatical rules proposed by Panini [1] in Ashtadhyayi. This methodology can be applied for the languages which are derived from Sanskrit. The proposed system has been developed based on tense, person and morphological inflections of the verbs to find their root forms. The work has been executed in two phases: first, the surface level forms or inflected forms of the verbs have been classified into a certain number of groups of similar tense and person. For this task, a standard pattern, available in Bengali language has been used. Next, a set of rules have been applied to extract the root form from the surface level forms of a verb. The system has been tested on 10000 verbs collected from the Bengali text corpus developed in the TDIL project of the Govt. of India. The accuracy of the output has been achieved 98% which is verified by a linguistic expert. Root verb identification is a key step in semantic searching, multi-sentence search query processing, understanding the meaning of a language, disambiguation of word sense, classific
Based on the sense definition of words available in the Bengali WordNet, an attempt is made to classify the Bengali sentences automatically into different groups in accordance with their underlying senses. The input sentences are collected from 50 different categories of the Bengali text corpus developed in the TDIL project of the Govt. of India, while information about the different senses of particular ambiguous lexical item is collected from Bengali WordNet. In an experimental basis we have used Naive Bayes probabilistic model as a useful classifier of sentences. We have applied the algorithm over 1747 sentences that contain a particular Bengali lexical item which, because of its ambiguous nature, is able to trigger different senses that render sentences in different meanings. In our experiment we have achieved around 84% accurate result on the sense classification over the total input sentences. We have analyzed those residual sentences that did not comply with our experiment and did affect the results to note that in many cases, wrong syntactic structures and less semantic information are the main hurdles in semantic classification of sentences. The applicational relevance of
We demonstrate that machine learning enables the capability to infer an individual's propensity to vote from their past actions and attributes. This is useful for microtargeting voter outreach, voter education and get-out-the-vote (GOVT) campaigns. Political scientists developed increasingly sophisticated techniques for estimating election outcomes since the late 1940s. Two prior studies similarly used machine learning to predict individual future voting behavior. We built a machine learning environment using TensorFlow, obtained voting data from 2004 to 2018, and then ran three experiments. We show positive results with a Matthews correlation coefficient of 0.39.
In the absence of neither an effective treatment or vaccine and with an incomplete understanding of the epidemiological cycle, Govt. has implemented a nationwide lockdown to reduce COVID-19 transmission in India. To study the effect of social distancing measure, we considered a new mathematical model on COVID-19 that incorporates lockdown effect. By validating our model to the data on notified cases from five different states and overall India, we estimated several epidemiologically important parameters as well as the basic reproduction number ($R_{0}$). Combining the mechanistic mathematical model with different statistical forecast models, we projected notified cases in the six locations for the period May 17, 2020, till May 31, 2020. A global sensitivity analysis is carried out to determine the correlation of two epidemiologically measurable parameters on the lockdown effect and also on $R_{0}$. Our result suggests that lockdown will be effective in those locations where a higher percentage of symptomatic infection exists in the population. Furthermore, a large scale COVID-19 mass testing is required to reduce community infection. Ensemble model forecast suggested a high rise in
Researchers have proposed that black holes stop evaporating at the last moment, leaving behind tiny remnants that preserve all the information they contain。 The same seven-dimensional geometry behind this idea could also help explain why elementary particles have mass
NASA’s Lucy spacecraft discovered that asteroid Donaldjohanson is a wobbling, peanut-shaped relic born from a violent collision and slowly reshaped by the subtle force of sunlight。 It also carries traces of ancient water, making it an important clue to the solar system’s mysterious past
Scientists have taken an important step toward building quantum detectors that could reveal some of the universe’s biggest secrets。 Using a prototype device with two clouds of ultracold atoms, researchers showed that a clever noise-canceling technique can recover hidden signals even when individual measurements appear completely overwhelmed by inte
Search has for a long time been an important tool for users to retrieve information. Syntactic search is matching documents or objects containing specific keywords like user-history, location, preference etc. to improve the results. However, it is often possible that the query and the best answer have no term or very less number of terms in common and syntactic search can not perform properly in such cases. Semantic search, on the other hand, resolves these issues but suffers from lack of annotation, absence of WordNet in case of low resource languages. In this work, we have demonstrated an end to end procedure to improve the performance of semantic search using semi-supervised and unsupervised learning algorithms. An available Bengali repository was chosen to have seven types of semantic properties primarily to develop the system. Performance has been tested using Support Vector Machine, Naive Bayes, Decision Tree and Artificial Neural Network (ANN). Our system has achieved the efficiency to predict the correct semantics using knowledge base over the time of learning. A repository containing around a million sentences, a product of TDIL project of Govt. of India, was used to test
A new sunlight-powered material can convert visible light into higher-energy UV light, overcoming a challenge that has frustrated scientists for years。 The breakthrough could enable cleaner air purification, solar-driven chemistry, and advanced manufacturing technologies using nothing more than natural sunlight