Vellore Institute of Technology (VIT) is a private university located in&nbsp;Tamil Nadu, India. Founded in 1984, as Vellore Engineering College, the institution offers 20 undergraduate, 34 postgraduate, four integrated and four research programs. It has campuses in Vellore, Amravati, Bhopal and Chennai.

VIT is one of the top ranked private universities in India according to NIRF, THE and QS Rankings.&nbsp;Govt. of India has recognized&nbsp;VIT, Vellore as an&nbsp;Institution of Eminence. This has allowed VIT to take independent quality initiatives and move up in world ranking.

&nbsp;

&nbsp;

VIT University

Department of Database Systems

Diabetes is the key critical issue needs to be concerned for various problems in our body. Increase in glucose and fructose content in our body results in diabetes mellitus. When a body generates higher insulin level than the required, it results in increased urination and excessive thirstiness which in turn results in kidney failure and other cardio-related issues. Many research agencies invested their funds on defining the predictive methodology and finding the root cause of those results in mellitus. Mellitus results in the highest mortality rate compared to any other disease reported by the health organizations across the globe. In this, the predictive methodologies, various classification techniques are discussed, and the results are analyzed. The classification methodology could be on medications, food habits, personal behaviors, age factors and so on. The datasets are processed and analyzed with the neural network algorithms, and the results are compared with one another. The datasets are taken from the National Family Health Survey results published during the period of 2016–2017. The result implies that men between ages 15–49 among 1 billion people have reported with diabetes mellitus. Diagnose and forecast on this disease are done by recognizing the pattern formation and grouping the similar structures. Various algorithmic techniques like M-layer perceptron, nearest neighbor, vector machines, data regressions, binary regression and their accuracy of forecast, speed and sensitivity are calculated, analyzed and compared to define the accurate prediction methodology over a short span of time. The forecast methodologies are focussed to provide solutions to avoid the intensive care system provided proper medications with a long duration when it is been predicted to be a risk factor. A statistical method of analyzing is performed for the comparative analysis. The learning and training methodologies are discussed in this system. Accuracy, specificity, sensitivity are the key parameters to define the best forecast methodology. Classification on association, regression techniques and neural algorithmic techniques is analyzed and compared to refine the best predictive forecast methodology by processing 30 samples across the states of India with focus on determining the type of mellitus along with the accuracy on definition. The forecast data utilized to define the type of mellitus and the prediction on critical measures over a period of time. © Springer Nature Singapore Pte Ltd 2021.

School of Computer Science and Engineering

Vellore Campus

Department of Commerce

School of Social Sciences & Languages

World University root node

Lecture Notes in Networks and Systems

Supervised learning method and neural network algorithm for the analysis of diabetic mellius and its comparitive analysis

Weed is one of the main reason for getting less production in agriculture field. At present, farmers are using herbicides to control the weed but it's having negative impact on crop production. To increase the crop production, farmers want to reduce the usage of herbicides. One of the machine learning algorithm, convolution neural network used to classify the weed and crop with high accuracy. CNN gives the best performance in some critical situations also, such as collected images of different lighting conditions, identification of plant species, overlapped crop with weed and designing an autonomous patch sprayers. This review article presents a brief overview of some significant research efforts in weed detection system by using image processing techniques and convolution neural network in machine learning algorithms. Not only CNN, this article reviewed different successful machine learning techniques of supervised and unsupervised learning approaches such as support vector machines, random forest and artificial neural network. This paper aims at providing different challenges and imperatives of various convolution neural network algorithms used in weed detection. Finally, compared different convolution neural network architectures and finds the best CNN architecture used for weed identification system with respect to accuracy. © 2021 IEEE.

Proceedings - International Conference on Artificial Intelligence and Smart Systems, ICAIS 2021

A Detailed Review on Challenges and Imperatives of Various CNN Algorithms in Weed Detection

Deep learning has been widely used in numerous real-world engineering applications and for classification problems. Real-world data is present with neutrality and indeterminacy, which neutrosophic theory captures clearly. Though both are currently developing research areas, there has been little study on their interlinking. We have proposed a novel framework to implement neutrosophy in deep learning models. Instead of just predicting a single class as output, we have quantified the sentiments using three membership functions to understand them better. Our proposed model consists of two blocks, feature extraction, and feature classification. Having a separate feature extraction block enables us to use any model as a feature extractor. We experimented with BiLSTM using GloVe (Global Vectors for word representation), BERT (Bidirectional Encoder Representations from Transformers), ALBERT (A Lite BERT), RoBERTa (Robustly optimized BERT approach), MPNet, and stacked ensemble models. Feature classification performs prediction and dimensionality reduction of features. Experimental analysis was done on the SemEval 2017 Task 4 dataset (Subtask A). We used the intermediate layer features to define membership functions of Single Valued Neutrosophic Sets (SVNS). We used these membership functions for prediction as well. We have compared our models with the top five teams of the task and recent state-of-the-art systems. Our proposed stacked ensemble model achieved the best recall (0.733) score. © 2021 Elsevier B.V.

Knowledge-Based Systems

Comparison of neutrosophic approach to various deep learning models for sentiment analysis[Formula presented]

In this paper, we predict the deformation behavior of alloy 690 by explicitly modeling finite thickness grain boundaries within the crystal plasticity framework. This is realized by using a blend of quasi-static tensile experiments, microstructural characterization using scanning electron microscopy, atomistic simulations and crystal plasticity finite element analysis. We begin with material characterization to probe the type of grain boundaries present in the Ni-based alloy 690. Inspired by physically justified multiscale modeling schemes, we then use atomistic simulations within the framework of embedded-atom method to quantify activation parameters for dislocation nucleation from those grain boundaries in Ni bicrystals. The kinetic activation parameters are then passed on to a transition state theory based crystal plasticity model. At this scale, the grain boundaries are explicitly modeled using finite elements and prescribed distinct constitutive parameters obtained from lower scale atomistic simulations. The predictive capabilities of the adopted methodology are demonstrated by capturing the uniaxial tensile behavior of alloy 690 at different temperatures (250C–6000C) (stress-strain curves as well as texture evolution). Our implementation of grain boundaries in crystal plasticity is built upon a standard representative volume element used in most polycrystalline simulations. It is envisioned that this comprehensible approach will provide a compelling platform for unifying many other interface related deformation processes at the mesoscale. Eventually, the demonstrated methodology can expedite in apprehending the structure-property relationships through apt use of physically justified multiscale modeling schemes. © 2021 Acta Materialia Inc.

Materialia

Atomistically informed crystal plasticity analysis of deformation behavior of alloy 690 including grain boundary effects

The long-term corrosion behavior of IN-RAFM steel in molten Pb-Li upto 10, 000 h of exposure was studied in a rotating disc corrosion test facility at 823 K. As-received IN-RAFMS material possessed an air-formed surface layer containing oxides of Fe, Ce and W. Precipitation of chromium carbides at the grain/lath boundaries of as-received IN-RAFMS led to grain boundary attack by liquid Pb-Li during the initial “Incubation Period” which depicted slow dissolution of surface oxides and lesser corrosion rate. Complete dissolution of oxide layer altered the corrosion mechanism from grain boundary attack to matrix dissolution thereby increasing the corrosion rate. © 2021 Elsevier Ltd

Corrosion Science

Effect of oxide layer and the duration of exposure on the liquid metal corrosion mechanism of RAFM steel in molten Pb-Li

The proposition of a new algorithm facilitates the predictability of weak/strong monsoons that lead to drought/flood events, respectively, in the Indian summer monsoon rainfall (ISMR). The proposed method estimates skewed Gaussian kernel distribution in the extreme values extracted from the rainfall series and confidence levels of drought and flood years are obtained using bootstrap. Using the selected Coupled Model Intercomparison Phase 5 (CMIP5) simulations under representative concentration pathways (RCP) 8.5 scenario, the proposed method detects that extreme droughts (at 99{\%} confidence level) in India are likely to occur in 2024 and 2027 in the early 21st century. Similarly, models project that 2031, 2032 and 2033 will be the most prominent flood years. It is projected that the probability of drought occurrence is likely to increase by 16{\%}. In contrast, it is expected to diminish flood events by 11{\%} in the future under projected global warming. Notably, our analysis reveals that 23.4{\%} of grids covering {\~{}}30{\%} of the Indian region are likely to experience increased frequency and intensity of droughts during 2020–2029, mainly covering the Northeast, Central and Southern India. Furthermore, during this period, the Northeast and some parts in the North would experience floods over 29.6{\%} (which covers {\~{}} 39{\%}) of the total grids. The proposed algorithm may be used for drought and flood monitoring over any geographical terrain.

Theoretical and Applied Climatology

A novel method to detect drought and flood years in Indian rainfall associated with weak and strong monsoon

This paper focuses to solve two important problems faced in this digital era. The first problem is handling large amount of digital data. Digital sectors are finding difficulty in handling repetitive works and standardization of process. Our University is a eco-friendly, digital university each and every work of our students are maintained digitally. So for a course instructor to open, view, correct and finding plagiarism in the assignment or homework submitted by students is a tough task. So in this paper this tough task is simplified by applying Robotic Process Automation (RPA). The process of downloading and reading the students’ digital document is automated. The second problem is the plagiarism. There has been a massive threat in the academic and technological integrity of innovation in the name of paraphrasing. In a survey conducted by Donald McCabe of Rutgers University over 70 percent of students admit to paraphrasing and copying of others work. They have also fabricated bibliography. This problem can be addressed using the state-of-the-art document similarity deep learning neural networks. In this paper, we propose the design and implementation of a Siamese Neural Network (SNN), Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) to identify paraphrasing, using only features of similarity between words where the dataset is evolutionary and the model keeps learning with incremental sources of the document’s plagiarism. This paper will also provide the Tensorflow based implementation of such a model. The proposed system will predict the similarity between documents and predict the plagiarism percentage.

Department of Analytics

Solid State Technology

Plagiarism Detection using Attention based Text Similarity

Polycystic ovary syndrome (PCOS) disorder is identified by the presence of a number of follicles present in the ovary of female reproductive system. Ultrasound imaging of the ovary contains essential information about the size, number of follicles and its position. In real time, the detection of PCOS is a difficult task for radiologists due to the various sizes of follicles and is highly connected with blood vessels and tissues. This often results in error diagnosis. For preprocessing various standard filtering techniques are applied on ovary image. Based on the performance, appropriate filter is chosen to remove the noise from the image. This paper presents an effectual active contour with modified Otsu threshold value to automated discovery of follicles from the ultrasound images. The performances of the proposed method illustrate the betterments of the proposed approach over other techniques. © 2019, Springer Science+Business Media, LLC, part of Springer Nature.