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

Prediction and forecasting of crop yields based on various climatic conditions are an intensive process. In the proposed work adaptive boosting prediction model is presented, using the datasets of barley crops along with the climatic features of the states located in Colorado, Wyoming, Idaho and Montana. The study discusses the experiments on weak regression technique like the linear regression, SVR Linear regression and proposes a strong prediction method in terms of regression like the Adaptive Boosting (Adaboost) technology in order to increase the accuracy in the prediction techniques. The Adaboost method combines several weak regressors and presents a weighted sum representing a strong final output. It has been observed from the experiments that, the decision of weak regressors varies due to frequent, inherent attributes of climatic conditions for crop production. The various numerical simulations, results and statistical measures demonstrate the efficiency of the strong Adaboost based regressors when compared to weak linear regressors. © 2020 Ecological Society of India. All rights reserved.

World University root node

Department of Analytics

School of Computer Science and Engineering

Vellore Campus

Indian Journal of Ecology

Barley crop yield forecasting using ensemble method-adaptive boosting of weak regressors

Diabetes mellitus is a metabolic illness that causes high blood sugar, which is widely known as diabetes. Insulin is a hormone produced by an organ situated behind the abdomen called the pancreas. This insulin agent moves glucose from your blood into the cells for energy and storage. With diabetic disorder, the body either will not create enough insulin or can’t effectively use the insulin it does create. Untreated high blood glucose or sugar from diabetic disorder will harm the nerves, eyes, kidneys, and different organs of the body. There are different data mining software tools to predict and analyze diabetes. Many attempts have been made by researchers to improve the efficiency of various models. The proposed method is Dimensionality reduction and clustering technique. It gives the highest accuracy for the larger dataset for both balanced and imbalanced datasets. In this paper, large and small datasets have been taken for clustering using K-means approach, Farthest first method, Density based technique, Filtered clustering method and X-means approach. K-means, density based and X-means gives the highest accuracy of 75.64%. For the larger balanced dataset when compared with the smaller balanced dataset.

Department of Software and Systems Engineering

School of Information Technology and Engineering

Department of Computer Applications

Proceeding of the International Conference on Computer Networks, Big Data and IoT (ICCBI - 2019)

Detection of Type 2 Diabetes Using Clustering Methods – Balanced and Imbalanced Pima Indian Extended Dataset

The Travelling salesman problem also popularly known as the TSP, which is the most classical combinatorial optimization problem. It is the most diligently read and an NP hard problem in the field of optimization. When the less number of cities is present, TSP is solved very easily but as the number of cities increases it gets more and more harder to figure out. This is due to a large amount of computation time is required. So in order to solve such large sized problems which contain millions of cities to traverse, various soft computing techniques can be used. In this paper, we discuss the use of different soft computing techniques like Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Ant Colony Optimization (ACO) and etc. to solve TSP.

VIT Unitversity

International Journal of Recent Technology and Engineering Special Issue

Performance Analysis of Genetic Algorithm, Particle Swarm Optimization and Ant Colony Optimization for solving the Travelling Salesman Problem

Applications on sentiment analysis via diverse context in analyzing the individual opinion on various issues such as political events, e-governance, and product reviews. Decision-making through the sentiment analysis improves the understanding of the public opinion. Opinion mining can be achieved by retrieving data through social network, microblogs, blogs, and search engines. Twitter tweets are an invaluable source of knowing the individual opinion from lots and lots of unique personality. However, the unstructured data to the huge volume and unwanted punctuation used in context and the emoticons used in the context is the main task to analyze the efficiency data and with greater accuracy. Most of the existing computational methods/algorithms identify sentiment on unstructured data via algorithms on machine learning like (BOW) bags of word approach. Here is the work of both the supervised and unsupervised approaches on various training datasets used. Automatic generation of the sentiment for the tweets extracted is provided by the unsupervised approach. And various algorithms in machine learning are used to determine the sentiment analysis they are as Maximum entropy (ME), Multinomial Naïve Bayes (MNB) and support vector machines (SVM) and are used to identify sentiment from the tweets. Here in this work I have achieved an accuracy of 87% unsupervised 97% in supervised approach. The ngram, unigram, bigram, and parts-of-speech (POS) were combined together to identify the hidden emotion and sentiment in the context that mentioned in the tweet that are all in an unstructured format. The lexicon-based approach 75.20% is achieved, based on the sentiment prediction the opinion is given. © Springer Nature Singapore Pte Ltd. 2019.

Department of Information Technology

Advances in Intelligent Systems and Computing Information Systems Design and Intelligent Applications

Provision of Efficient Sentiment Analysis for Unstructured Data

A brain can easily able to recognize handwritten words which computers are still struggling to figure out what they are. This amazing power of the brain requires to be transferred to a computer for automatic digitalization of handwritten texts and manuscript. The work proposed in this paper tried to make a model which is able to recognize and predict different Hindi character which can be further used to identify and digitalize the whole page or text. Our model uses different approaches like a base-line neural network, convolutional neural network, and a simple logistic regression model. © 2019 IEEE.

2019 International Conference on Intelligent Computing and Control Systems, ICCS 2019

Performance analysis of the classifiers for optical character recognition

Decades of research and experiences on organizing the large databases and contemporary world’s resilient interests in gigantic data, information transferred various indexing techniques to a new level. From all-encompassing experiments, the PFB+-tree has exhibited its excellent prospective for big data in-memory management system. The PFB+-tree does quick indexing structure by utilizing range searches, point searches and multi-level key ranges are quickened by prior pruning of searches to get distrait data. The depth-first or breath-first searching methods are used for range searches. One consortium of various searches can be set upwards with only single pass on indexing structure to reduce the time complexity. The LSB+-tree can speed up queries directed to the most accessed ranges of indices. Extended experiments are presented to show and high efficiency, particularly for testing its adaptability and performance for big data storage management.

Indian Journal of Public Health Research & Development

PFB+_ Tree For Big Data Memory Management System

The scope of urbanization variation due to landscape alterations, profoundly affect the ecological features and decision-making process for built-up standards. In various parts of the globe, urbanization studies have identified a significant correlation between substantial human benefits on quality-of life, quantifiable resource utilization and local dimatic parameters. Remote sensing environment has developed techniques to estimate change in spatio-temporal attributes in a georeferenced imagery. The Intricate multi-signature classes and massive data interpretation for minute urban change detection for Kancheepuram, Tamilnadu. Supervised Land-use-and-Land-Cover (LULC) classification using Neural Network(NN), Minimum Distance, Support Vector Machine(SVM) &amp; Maximum Likelihood technique to estimate change In urban class is performed. Unsupervised classification, Inbuilt programmed distance learning algorithm, with ISO and K-means, is performed on preprocessed and enhanced PCA image. Support vector machine and ISO classification techniques with enhanced imagery, on the prior basis shows more accuracy (80-85%) amongst mentioned techniques. Thus, following SVM to categorize attributes to classes and performing urban change detection. The discrepancy between new land and barren land via SVM practice remain uncertain. © 2018 Ecological Society of India. All Rights Reserved.

Identify suitable classification algorithm for water pixel count extraction - A case study of puzhal lake

Live migration an advancement with which the whole running virtual machine (VM) is relocated beginning with one physical machine then onto the next. Migration at the level of the whole VM suggests that in-memory state can be moved in a reliable and effective design. Migrating operating system instances across distinct physical hosts is a beneficial tool for clusters and administrators. Along these outlines, a significant issue in live migration is the complete migration time and the downtime. To enrich the concert of live migration, an optimized iterative pre-copy procedure is used to decrease the dirty rate of virtual machines. In pre-copy approach that is essentially utilized as a part of live migration, total migration time, which influence on the performance of VM, is delayed by iterative copy operations and the noteworthy measure of transferring data during the entire migration process. In this paper, we presented a system that incorporates pre-processing phase in traditional pre-copy based live migration for decreasing the amount of transferred data. In pre-processing stage, we recommend the prediction working set algorithm to show that complete migration time as well as downtime is controlled by specific memory utilization patterns. Applying the proposed lazy counting based splay tree algorithm, the system can diminish the amount transferred memory page. The lazy counting-based splay algorithm along with traditional pre-copy approach that contains a pre-processing phase for diminishing the amount of transferring memory page and total migration time. © 2006-2017 Asian Research Publishing Network (ARPN).

ARPN Journal of Engineering and Applied Sciences

Enhancing the concert of live migration of virtual machine's with lazy counting-based splay tree algorithm

The project aims to provide an efficient, low-cost automated energy management system for houses. It also provides a facility for surveillance of the house. The system has been built after evaluating the utility features of surveillance and energy management systems available at present and is an attempt to improve these features. In addition to providing a cost-effective solution for energy management in the household, it also provides features to cater to natural disasters like fire. The system is built on an Arduino UNO microcontroller board and uses embedded C as the programming language which provides a facility of easier coding for new features. © 2017 IEEE.

2017 International Conference on Communication and Signal Processing (ICCSP)

Home security and energy efficient home automation system using arduino

A new modularization technique is used in Aspect-oriented software development for the separation of widely used functions such as logging, caching, synchronization, and exception handling from the core business logic functions. Aspects are identified using the mathematical modeling tool, Colored Petri nets. During the software development process, dependencies may arise as a result of using operators such as Before, After, Around, and Replace; such dependencies are consumed by our framework. Using the specification of aspects, we generate a composition rule for every match point, which directs the composition process at the initial requirements phase of software development. The proposed FTS approach, incorporating a feedback edge set, topological ordering, and second valid ordering, is efficient in resolving conflicts and dependencies among the aspects. To analyze the second valid ordering, grey relational analysis is used to rank the aspects, while analysis of variance method is used for the verification thereof. The proposed approach is illustrated by a case study. © 2014, King Fahd University of Petroleum and Minerals.