Showing papers by "KCG College of Technology published in 2021"

Topological Characterization and Graph Entropies of Tessellations of Kekulene Structures: Existence of Isentropic Structures and Applications to Thermochemistry, Nuclear Magnetic Resonance, and Electron Spin Resonance.

Abstract: Tessellations of kekulenes and cycloarenes are of considerable interest as nanomolecular belts in trapping and transportation of heavy metal ions and chloride ions, as they possess optimal electronic features and pore sizes. A class of cycloarenes called kekulenes have been the focus of several experimental and theoretical studies from the stand point of aromaticity, superaromaticity, chirality, and novel electrical and magnetic properties. In the present study, we investigate the entropies and topological characterization of different tessellations of kekulenes through topological computations of superaromatic structures with pores. We introduce the self-powered vertex degree-based topological indices and then derive the graph entropy measures for three different tessellations (zigzag, armchair, and rectangular) via various molecular descriptors that we derive here. Several applications to computing the molecular properties are pointed out. We demonstrate the existence of isentropic and yet nonisomorphic tessellations of kekulenes for the first time. The two tessellations are predicted to be quite close in energy with comparable energy gaps. Graph theory-based PPP methods with parameters derived from higher levels of theory are proposed to be promising tools for the predictions of relative stabilities of kekulene tessellations. We show that the developed techniques can be applied in the general context of artificial intelligence for the machine generation of nuclear magnetic resonance and electron spin resonance spectroscopic patterns as well as in robust computations of thermochemistry of a large combinatorial libraries of tessellations of kekulenes through the generation of bond-equivalence classes.

Statistical–model based voice activity identification for human-elephant conflict mitigation

Abstract: Human-Elephant conflict has become very common in the forest borders causing phenomenal increase in death of elephants as well as loss of human life. Elephants face a shortage of resources such as food and water, resulting in Human–Elephant Conflict. The presence of elephants can be predicted using elephant voice detection. We propose a machine learning based system for detecting elephant voice and predicting the presence of elephants in the forest border areas. Existing methods for elephant voice detection in literature require large feature data dimensions. In the proposed system, feature extraction methods combined with Principal Component Analysis (PCA) that greatly reduces feature data dimension is proposed to improve the performance metrics of the recorded elephant voice samples. A Support vector machine (SVM) classifier is used for the predictive model in this work. The proposed system is validated using the cross validation method and the performance metrics such as Accuracy, Sensitivity, Specificity, Precision, F1 Score and Computation time are evaluated. It is observed that with the proposed approach the average accuracy is 93.32% and feature data dimension is 1422 compared to an average accuracy of 83.5% obtained and feature data dimension of 18,882 with methods using Mel-Frequency Cepstral Co-efficient (MFCC)

Design and implementation of hybrid integration of cognitive learning and chaotic countermeasures for side channel attacks

Abstract: Security in embedded systems is considered to be more important and needs to be a diagnosis for every minute. Also with the advent of the Internet of Things (IoT), security in the embedded system has reached its new peak of dimension. A Mathematically secure algorithm was formulated and runs on the cryptographic chips which are embedded in the systems, but secret keys can be at risk and even information can be retrieved by the prominent side-channel attacks. Fixed encryption keys, non-intelligent detection of side-channel attacks are some of the real-time challenges in an existing system of encryption. Following the limitations of existing systems, this research article focuses on the integration of powerful machine learning algorithms by retrieving the secret key information with countermeasures methodology using the chaotic logistic maps and includes the following contributions: (a) Preparation of Data Sets from the Power consumption traces captured from ARTIX-7 FPGA boards while running the Elliptical Curve Cryptography(ECC) on it (b) Implementation of High Speed and High Accurate Single feed-forward learning machines for the detection and classification of side-channel attacks (c) Design of Chaotic Countermeasures using 3-Dlogistic maps for attacked bits. The test_bed has been developed using the integration of FPGA along with Cortex-A57 architectures for experimentation of the proposed work and various evaluation parameters such as Accuracy, F-calls, Precision rates, sensitivity, and correlation co-efficient, entropy were calculated and analyzed. Moreover, the parameters of the proposed system which has been analyzed prove to outperform the other existing algorithms in terms of performance and detection.

Low rate DDoS mitigation using real-time multi threshold traffic monitoring system

Abstract: The low rate distributed denial of service (DDoS) attack has been identified as most vulnerable to the network services which has been studied recently. The approaches consider only the high rate DoS attacks and ignore rest in low rate. The existing techniques suffer with poor detection of low rate attacks as they consider only limited features of network traffic. Variety of techniques mitigate such threats using different parameters like amount of data in service packet as payload, number of intermediate nodes, and so on. The previous techniques struggle to detect and mitigate them in efficient way. Towards improving the detection and mitigation performance of low rate threats, the author presents a novel real time traffic monitoring algorithm which uses multi threshold traffic analysis. By considering the payload, hop count, latency, packet counts, the method analyzes the real time traffic. Using the features obtained from the traffic, the method computes the low rate threat measure. Based on computed threat measure, the packets trustworthy have been validated. The method produces higher detection rate in low rate DDoS attack detection and produces efficient results.

Flow Mixing and Screech Characteristics of Elliptical Supersonic Jet

Abstract: Abstract The acoustic and flow characteristics of a jet with elliptical throat is studied at different levels of nozzle expansion ratio. In this study, we have taken two types of CD nozzle configuration (circular and elliptical throat) and it is studied for various NPR ratios of 2, 3, 4 and 6. In addition, the acoustic characteristic of the jet flow is also measured for respective NPRs. Measurements of acoustic data are done using microphones placed at 30, 60 and 90 degrees to imprison the effects of screech tone. At NPR 2, 3 and 4, the jet with elliptical throat witnesses superior mixing and shorter core length compared to the circular throat. Its surprising to see both the configurations provides the identical oscillation at NPR 2, 3 and 4, however the efficiency of jet mixing is larger in elliptical throat jet. As the nozzle pressure ratio increased from 2 to 3 and 3 to 4, the potential core length of the jet reduces marginally about 5 to 10 % for every NPR until nozzle pressure ratio of 5. At NPR 2 and 3, the centerline pitot pressure profile shows, the decay of jet from the elliptical throat is healthier than a circular jet. At various levels of nozzle expansion, shock cell shows an appreciable change with an increase in NPR. Introduction of the elliptical throat on circular modifies the structure of shock cell which significantly changes the magnitude of screech tone due to the weakening of shock waves.

Stock Price Prediction Based on LSTM Deep Learning Model

Abstract: Predicting the stock market is either the easiest or the toughest task in the field of computations. There are many factors related to prediction, physical factors vs. physiological, rational and irrational , capitalist sentiment, market , etc. All these aspects combine to make stock costs volatile and are extremely tough to predict with high accuracy. The prices of a stock market depend very much on demand and supply. High demand stocks will increase in price while heavy selling stocks will decrease. Fluctuations in stock prices affect investor perception and thus there is a need to predict future share prices and to predict stock market prices to make more acquaint and precise investment decisions. We examine data analysis in this domain as a game-changer. This paper proposes that historical value bears the impact of all other market events and can be used to predict future movement. Machine Learning techniques can detect paradigms and insights that can be used to construct surprisingly correct predictions. We propose the LSTM (Long Short Term Memory) model to examine the future price of a stock. This paper is to predict stock market prices to make more acquaint and precise investment decisions.

LMI approach-based sampled-data control for uncertain systems with actuator saturation: application to multi-machine power system

Abstract: This paper is devoted to the topic of robust stabilization for uncertain multi-machine power systems (MMPSs) using input delay-based sampled-data control. The study explores the sampled-data control for a nonlinear MMPS with parametric uncertainties exacerbated with sector saturating actuators. A saturated controller is considered for the system to recover the loss of stability in the continuous time domain. An approach, comprising linear matrix inequality technique and average dwell time method, is exploited, employing proper Lyapunov–Krasovskii functional, to show that the proposed saturated sampled-data control renders exponential stability. More precisely, the existence condition of sampled-data control law is developed in form of linear matrix inequalities. In order to simplify the derivation in main results, Schur complement and Wirtinger inequalities are used. Through the simulation tests on a two-machine infinite bus system model, the effectiveness and robustness of the proposed controller over the time delays and parameter uncertainties are verified.

Fuzzy artificial bee colony-based CNN-LSTM and semantic feature for fake product review classification

Abstract: In recent years, online reviews are considered as the most significant resource for consumers to make a decision regarding the purchase of a particular product. The reviews can either encourage or relegate a product; therefore posting fake reviews has turned into a money‐spinning business in the modern period. The detection of fake reviews has become a center of attraction for various business people. This research study aims in detecting fake product reviews using four significant phases namely the data pre‐processing, feature extraction, feature selection, and classification. The features obtained in the pre‐processing phase are extracted and selected using chi‐squared technique to obtain a delegate subset among all data and to reduce the complication issues. Then a CNNLSTM‐FABC approach classifies and detects the review as fake or real. Finally, the performance evaluation and the comparative analysis are carried out to determine the effectiveness of the proposed approach. The results reveal that the proposed approach performs well irrespective of the product type and sentiment polarity.

Blockchain-Based Agriculture Assistance

Abstract: The working system of agricultural management is very important in today’s world. However, growers, farmers, and sellers are physically dispersed to choose their pesticides and to manage both data and information. As a result, the production of crops and trust between the consumer and producer decreases regularly. In this paper, we propose an overall structure of agricultural maintenance system to work efficiently and properly based on blockchain technology in order to solve the problems facing by the farmers to choose the best pesticides on their own. The information recorded in these management operations includes farmer’s problems and other suggestions. Using blockchain techniques and methods to the field of agricultural not only improves schema, domain, and application of blockchain but also supports the farmers to choose the suitable pesticides and helps in increasing the trusted network among different stakeholders around agricultural environment.

Miniaturized Dual-Band Metamaterial-Loaded Antenna for Heterogeneous Vehicular Communication Networks

Abstract: The design of a dual-band monopole antenna exploiting the concepts of metamaterial loading is presented in this paper. A left-handed metamaterial (LHM) is designed to reduce the size of a simple mo...

A Drug Recommendation System for Multi-disease in Health Care Using Machine Learning

Abstract: The remarkable technological advancements in the health care industry have improved recently for the betterment of patients’ life and providing better clinical decisions. Applications of machine learning and data mining can change the available data to valuable information that can be used for recommending appropriate drugs by analyzing symptoms of the disease. In this work, a machine learning approach for multi-disease with drug recommendation is proposed to provide accurate drug recommendations for the patients suffering from various diseases. This approach generates appropriate recommendations for the patients suffering from cardiac, common cold, fever, obesity, optical, and ortho. Supervised machine learning approaches such as Support Vector Machine (SVM), Random Forest, Decision Tree, and K-nearest neighbors were used for generating recommendations for patients. The experimentation and evaluation of the study was carried out on a sample dataset created only for testing purpose and is not obtained from any source (medical practitioner). This experimental evaluation shows that the Random Forest classifier approach yields a very good recommendation accuracy of 96.87% than the other classifiers under comparison. Thus, the proposed approach is considered as a promising tool for reliable recommendations to the patients in the health care industry.

Modified Hardware Security Algorithms for Process Industries Using Internet of Things

Abstract: Embedded devices used in process industries are vulnerable to a variety of attacks due to its large number of deployments to reduce the measurement and process complexity. The security threats incr...

Impact of cutting parameters on machining of Ti-6Al-4V alloy: an experimental and FEM approach

Abstract: Titanium alloys are used as an aerospace material due to their inherent properties such as high strength to weight ratio, corrosion, and fracture resistance. However, the low conductivity and reactivity towards plastic deformation causes these materials to be difficult to cut category. The prediction of various parameters like chip formation and actual cutting forces are important factors for better machinability which involves lot of resources. To overcome such issues, this work proposes three-dimensional FE approach to simulate the machinability behavior of Ti-6Al-4V especially on conventional turning. The impact of cutting speed and feed rate on the cutting force, thrust force, feed force and surface roughness were analyzed experimentally for various conditions. The predicted machining forces showed strong correlation with the experimental results and the effective von mises stress were examined.

Synthesis of CuAl2O4 Nanoparticle and Its Conversion to CuO Nanorods

Abstract: The molten salt approach was used to convert CuAl2O4 nanoparticles to CuO nanorods in this study. Molten hydroxide (NaOH) synthesis was chosen over molten salts (NaCl/KCl) for removing aluminium oxide from copper aluminate at low temperatures. The molten salt process is environmentally beneficial. Polymeric precursors were used to make nanosized copper aluminates. Alginic acid polymer is used to gel aqueous solutions of copper acetate and aluminium nitrate, yielding precursor after further heating. The precursor provides 14 nm nanosized copper aluminates after being heated at 900°C for 5 hours. XRD, FTIR, SEM, and TEM were used to characterize the nanosized copper aluminate powder. Solid state mixing and solution technique were used to investigate molten hydroxide treatment of spinel CuAl2O4. The products of the reaction were identified using XRD. FTIR and SEM are also used to analyze the sample. Using UV-DRS absorbance spectrum analysis, the optical characteristics of CuAl2O4 and CuO nanorods were examined. Using the Tauc plot method, the band gaps of CuAl2O4 and CuO were calculated to be 4.3 and 3.93 eV.

Twitter-Based Disaster Management System Using Data Mining

Abstract: Social media is an essential part of life for most people around. No wonder even during emergencies like flood or cyclone, more and more people look up to Twitter, Facebook, WhatsApp groups, etc., for immediate assistance. This helps to get data from even remote places and from small groups which will be difficult to reach. This sheer amount of data generated during a short span of time is also the challenge in this approach. Even when there are resources available for help, many requests could go unnoticed. This paper addresses above-mentioned problem by collecting the generated requests for help and resource availability and plot the location in the map. Request data shall be analysed using three machine learning algorithms called linear ridge regression, SGD classifier and Naive Bayes algorithm for the initial filtering and will be passed through natural language processing to match needs and offers within a given geographic boundary. The system is working with 96% accuracy for linear ridge regression and Naive Bayes classifier and 95% accuracy for SGD classifier. The report shall be published to provide a centralized status of requests. This brings more efficient management of disaster situations.

Identification of Plant Diseases Using Image Processing and Image Recognition

Abstract: In the field of agriculture, image processing is a constantly evolving field of research and progress. Currently, several plant disease identification studies are underway. Identifying plant diseases can not only help farmers increase yields, but also promote a variety of agricultural practices. This paper proposes an algorithmic program for the diseases detection and categorization with the assistance of machine learning mechanisms and image recognition tools. First detect and record the contaminated area and then perform image pre-processing. Then collect the fragments, identify the infected area, and perform feature extraction on it. This article discusses the methods of using leaf photography to detect plant diseases. In addition, this article also introduces some feature segmentation and extraction algorithms for plant disease detection.

On the Expressive Power of Scientific Manuscripts

Abstract: Every research manuscript is appreciated in the form of citations. Citations are expected to carry the essence of the underlying base paper by some rhetorical means. However, this is not true in reality. Citation manipulations are equally possible which shall be identified using research semantics. This paper discusses machine learning based approaches for analyzing research citations with the aim of finding quality research citations. On analyzing the semantics of the research manuscript and the respective citations, this paper proposes various metrics for citation quality analysis including deep cite, raw expressive power, expressive power and normalized expressive power.