National Institute of Technology, Kurukshetra

Software development is based on explicit technical fundamentals and techniques. There are diverse phases executed to predict the defects in software, such as employing the data for input, pre-processing it, extracting the attributes, and classifying the defect. The given paper introduces an ensemble framework that comprises algorithms, namely Gaussian Naive Bayes (GNB), Bernoulli Naïve Bayes (BNB), Random Forest (RF), and Support Vector Machine (SVM), for predicting the software defects. This ensemble approach consists of Principal Component Analysis (PCA) or orthogonal linear transformation (OLT) with class balancing for feature selection. Python is executed for simulating the proposed model. In addition to this, the performance of the proposed work is validated and compared with existing recent studies based on evaluation metrics such as accuracy, precision and recall. The results show that the proposed framework outperforms the existing recent studies in terms of performance.

Code clones in software system are identical or similar pieces of code. The code is repeatedly generated by the copy and paste program. As a result, every duplicate contains a defect that was detected in one unit and the existing techniques are unable to achieve high accuracy for the code clone detection. In this research work, a hybrid deep learning model is proposed which comprises four phases namely pre-processing, feature set generation, feature set optimization and clone detection. We have utilized particle swarm optimization (PSO) and genetic algorithm (GA) for optimization along with convolutional neural network (CNN) and long short-term memory (LSTM) for clone detection. The proposed model is implemented in python and tested on several datasets in terms of accuracy (%), precision (%) and recall (%). In addition to this, the proposed model is compared with existing recent studies in terms of performance and the results show that the proposed hybrid model attains the highest accuracy (94.67%), highest precision (93.12%) and highest recall (93.13%) in case of big clone bench (BCB) dataset. Similarly, our model attains the highest accuracy (93.90%), highest precision (93.50%) and highest recall (93.52%) in case of Google code jam dataset while in case of Java dataset, accuracy, precision and recall are 93.78%, 92.67% and 92.66% respectively.

Skin cancer is one of the most common cancers worldwide where early detection is crucial for effectively diagnosing and treatment. Traditional diagnostic methods largely depend on dermoscopic or skin lesion image analysis. However, they are limited by the clinician’s expertise and primarily rely on image data. While most of the studies focus on improving the diagnostic accuracy using the skin lesion images, this study addresses the limitations of using just the image data and proposes a novel approach that enhances skin cancer detection by combining skin lesion images with patients’ clinical notes. The methodology proposes a multi-modal approach that combines the images data with doctor’s clinical notes to enhance early diagnosis and accurate detection of skin cancer. Clinical notes are synthetically generated using two distinct multimodal large language models (LLMs), corresponding to each patient’s skin cancer images, encapsulating both visual and textual data representative of real-world scenarios. The generated clinical notes are validated through a dual-method approach involving cross-evaluation and consensus scoring, utilizing metrics such as the BLEU score, ROUGE score, Overlap Coefficient, and Jaccard index. Furthermore, this paper compares the performance of a model built using skin lesion images with four other models built using skin lesion images and synthetically generated clinical notes. The experimental results demonstrates a significant improvement in all classification metrics compared to single-modality models, achieving an accuracy of 99.51%, precision of 96.19%, recall of 97.95%and f1-score of 97.03% with the multimodal ALBEF model that uses GPT-4-turbo for clinical notes generation. The results show a significant improvement in performance with the multi modal approach for medical diagnosis. Also, this research sets a foundational framework that can be leveraged for many potential healthcare diagnoses applications.

Protecting private video content against access and interception is crucial in the current digital age. In order to overcome this difficulty, the work in this paper provides a secure encryption method that uses chaotic maps to encrypt multiple video streams simultaneously. Our scheme takes videos of different sizes as input, and then merges them into a single video by combining subsequent frames of all the videos horizontally. The encryption strategy comprises of two- chaotic map based two-level permutation and diffusion strategy that is applied to each merged frame. Two of our earlier proposed chaotic maps, Cascaded Coupled Logistic-Sine–Cosine (CCLSC) map and Sine–TangentSine (STS) map, have been used to implement two-level permutation and diffusion. In two-level permutation step, first row wise permutation is carried using STC chaotic map, and then column wise permutation is performed using CCLSC map. Likewise, both maps are used to implement the diffusion step of the proposed multiple video encryption scheme. The encrypted merged video is then either stored or transmitted over an unreliable network from the sender to the receiver site. Since the videos are encrypted and encapsulated as a single unit, they cannot be intercepted or accessed by eavesdroppers, thus ensures secure communication. Only the authorized individuals can decrypt the merged encrypted video at the receiver end and retrieve the different original videos by splitting the merged video into its constituent videos. Experimental evaluation of the work is performed using different standard encryption and decryption parameters. The information entropy of the encrypted image was 7.9998 and correlation coefficient was − 0.00065, − 0.00008, − 0.00053 in horizontal, vertical and diagonal directions respectively. The obtained value of other parameters also demonstrates the effectiveness and security of the proposed encryption scheme. Also, comparative analysis with existing encryption techniques showcases the superiority of our method.

The primary objective of the work is to elucidate the impact of inclined viscous dissipative magnetisation and ohmic heating on the steady-state flow of Casson nanofluid across a stretched sheet inside a porous medium including a heat source or sink. In this study, we consider convective heating in relation to temperature and particle slip conditions to determine velocity and concentration. The thermal radiation and Soret effects with chemical reaction are also considered. The energy and concentration equations incorporate Brownian motion and thermophoretic phenomena to accurately depict the behaviour of nanofluids within the boundary layer domain. We have used a typical Casson fluid model to differentiate the flow properties of non-Newtonian fluids from Newtonian fluids. We construct a coupled, nonlinear system of partial differential equations to describe the behaviour of heat and mass transfers. Similarity transformations are employed in sequence to lower the dimensional difficulty of the resulting differential equations. Here we use the bvp4c Matlab solver to find out the solutions for velocity, temperature, and concentration. The main outcome of this study is that the velocity of the Casson nanofluid reduces significantly whilst the thermal field rises as the inclined magnetisation increases. Additionally, increasing Eckert number result in increased thermal transport. An increase in the slip effect leads to a reduction in both concentration and velocity distribution. On the other hand, a higher value of porosity reduces velocity distribution. Finally, we verify the generated bvp4c solutions by comparing them to the existing solutions, finding remarkable agreement.

Model Order Reduction (MOR) techniques play a crucial role in reducing the computational complexity of high-dimensional mathematical models, enabling efficient simulations and analysis. In recent years, Artificial Intelligence (AI) has emerged as a powerful tool in various domains, including MOR. This survey paper provides an overview of AI-based MOR techniques, exploring how AI methods are being integrated into traditional MOR approaches. Different AI algorithms, such as machine learning, deep learning, and evolutionary computing, and their applications in MOR are discussed in this paper. The advantages, challenges, and future directions of AI-based MOR techniques are also highlighted.

During situations involving dangerous activities, such as armed robbery in public areas, surveillance systems often exhibit delays or inefficiencies in their prompt responding. To obviate the necessity for human involvement, there is a requirement for technology capable of autonomously detecting harmful objects, like firearms in surveillance footages. This article presents WeaponVision AI, an advanced software system that has the ability to accurately identify weapons in live feeds, recorded videos, and images. Moreover, this software has the capability to detect guns even under weak lighting circumstances. The deep learning architecture based on modified YOLOv7 was trained on a vast dataset assortment of 79,558 images of weapons, in developing the WeaponVision AI. The model exhibited satisfactory results following the training phase, achieving significant outcome metrics: a precision rate of 91.75% and a mean average precision of 92.15%. The efficacy of WeaponVision AI is showcased through its ability to accurately identify weapons across diverse environmental and visual conditions.

Distribution system networks (DSN) are presently gaining more concern in terms of security and stability due to the penetration of distributed energy resources (DERs) and the integration of microgrids. The active DSNs may have their operating point near the maximum power transfer capability due to the optimal asset utilization, which jeopardizes the stability of the network. The article presents a novel approach for optimizing the placement of solar PV, D-STATCOMs and energy storage in distribution networks. The method minimizes the L-Index for voltage stability, reduces voltage deviations, and lowers annual operating costs, ensuring enhanced performance and cost-efficiency for modern power systems. The article presents a novel approach of progressive L-Index to determine the suitable sites for optimal location of the DERs using Particle Swarm Optimization (PSO). It also utilizes the mixed-integer nonlinear programming using GAMS to compute the appropriate size of DERs, D-STATCOM and the energy storage. The studies have been performed on the standard IEEE-69 bus test DSN. The results exhibit that the proposed scheme enhances the overall voltage profile of the DSN, and the VSM is upgraded by 41.60 % with DER installation, as compared without DG.

The present study investigates the effect of the shape factor on the numerical simulation of thermally stratified three-dimensional rotational Darcy-Forchheimer flow. Specifically, the study focuses on the behaviour of a hydromagnetic Williamson hybrid nanofluid composed of Ag-MoS4 under the influence of suction/injection, considering the non-linear radiation and heat source/sink effects. The shape factor, which represents the geometric shape of the nanoparticles, plays a significant role in enhancing thermophysical properties. Further, the proposed models for viscosity (Gharesim) and thermal conductivity (Hamilton-Crosser) boost the thermal properties. The numerical simulations are carried out for the appropriate governing transformed equations for the flow phenomena, followed by similarity transformation rules adopted herein. In particular, Matlab inbuilt bvp5C code is deployed for the graphical solution and numerical calculations of the transformed model. The characteristic of each constraint is presented through graphs, and the simulated results for the rate coefficients are also depicted in tabular form. The results reveal that the shape factor significantly affects the flow behaviour, with different shape factors leading to distinct flow patterns. However, the important flow patterns are obtained as the non-Newtonian effect caused by the Williamson parameter retards the fluid velocity at points within the flow domain for its greater values; moreover, increasing thermal radiation enhances the fluid temperature with the inclusion of the proposed conductivity model. Graphical Abstract Investigating the influence of the numerous shaped nanoparticles on the flow behaviour mainly provides significant enhancement in thermophysical properties; the proposed Hamilton-Crosser model conductivity is particularly important in enhancing the thermal properties. Hybrid nanofluids combine different nanoparticles, such as Ag-MoS4, offering enhanced thermal conductivity and convective heat transfer properties. Analysing the behaviour of specific hybrid nanofluid under different flow conditions contributes to understanding its potential applications in various thermal systems. Investigating the numerical simulation of thermally stratified thermal flow in the context of Darcy-Forchheimer flow provides the fluid’s heat transfer characteristics and behaviour under complex flow conditions.

Wireless sensor and actor networks (WSANs) are gaining substantial recognition because of their utility in inhospitable environments where humans have restricted accessibility. The extensive applications of WSANs in different domains require effectiveness, reliability, and some degree of robustness. These networks may experience frequent node failures due to their deployment in rough environments, such as energy depletion and onboard electronics failure of network nodes. These failures result in areas with no coverage, which may further deteriorate the standard of data accumulated. However, the network partitioning into disjoint fragments is the most severe repercussion that arises due to these failures. The network partitioning results in various negative impacts like obstruction in data exchange and restricted coordination among the nodes. Therefore, detecting partitioning in the network and restoring connectivity are important. This paper reviews the reported network partition detection and recovery techniques. The limitations of these recovery techniques are highlighted, along with the advantages of incorporating unmanned aerial vehicles (UAVs) in various ground wireless networks. UAVs are evolving to become a critical element of future wireless network technologies. This paper incorporates the analysis of UAV‐assisted networks consisting of technical issues, challenges, and requirements related to the UAVs' employment in wireless networks. Thereafter, the reported UAV‐assisted partition recovery techniques are discussed along with the application scenarios for incorporating UAVs in network partitioning detection and recovery problem. The paper also highlights the challenges for associating UAVs in the network recovery process.

Human Activity Recognition (HAR) has been identified as a hotspot for finding the daily living activity of human beings. The researchers have proposed a substantial number of models for HAR. In addition, both time and performance domains greatly emphasize the experimental outcomes. The research challenges occurred whenever the HAR model produced the optimal results under these domains and found compelling temporal sensor activity data features. To address these research challenges, we have proposed a Performance and Time-efficient Recurrent (PTeR) model for learning and capturing the frequent occurrence of activity patterns. Further, the proposed model contains the recurrent networks for handling the temporal sensor data. The experimental findings regarding performance measurement and time reduction have been evaluated using the FLAAP dataset. We have used the F1 score (%) and precision (%) for the performance measurement. Moreover, total training time (sec.), saved training time (sec.), and their percentage employed for computing the time consumption. The impact of computational time reduction and performance improvement are evaluated. According to the experimental findings, the PTeR (RNN) has surpassed Random Forest (RF), CNN + RNN, CNN + LSTM, and has attained the F1 score of 95.04% and a precision of 89.95%. Further, the PTeR (GRU) saved 50.15% of the overall training time and used 221.38 s less than the CNN + RNN and CNN + LSTM models. Finally, the recognition rates of the PTeR model outperformed the comparative models developed using convolutional-aided and traditional learning algorithms.

In the present work, we consider Stancu variant of a parametrically generalized Baskakov operator. We discuss the error of appoximation of these operators by means of unified Ditzian Totik modulus of smoothness and various moduli of continuity like modulus of continuity of second order and weighted modulus of continuity. Also, we study the statistical convergence of newly defined operators. Some numerical examples illustrating the error functions for varying Stancu parameters and the approximation by proposed operators are also given using MATLAB programming.

Non-orthogonal multiple access (NOMA) is a notable technology for enhancing spectrum usage in wireless communication. On the other hand, cognitive radio (CR) networks are also renowned technology for increasing spectrum efficiency. However, fifth-generation wireless networks cannot provide a dynamic wireless environment. This barrier is overcome by sixth-generation (6G) wireless networks. In 6G, a dynamic wireless environment can be achieved by an intelligent reflecting surface (IRS). IRS is an eminent technology that enhances the overall quality of experience in wireless systems. This paper presents users’ performance analysis in IRS-aided NOMA-based 6G CR networks to capitalize on these technologies. The most popular five machine learning (ML)-based classifiers have been considered to sense the feature of the spectrum and evaluate the performance of the IRS-aided NOMA-based 6G CR network for the probability of detection, throughput, and energy efficiency. The simulation results have been validated for the proposed network with and without ML-based classifiers. Further, the performance of the proposed network has been tested for the different ratios of sensing time to total time, probability of false alarms, and different signal sizes of the CR network. The time complexity of the proposed network has been evaluated and found that the network has satisfactory inference time. The simulation results also suggest that the proposed network may fulfill the spectrum, energy, and reliability requirements of the 6G wireless networks.

We have demonstrated the superior electrochemical properties of copper oxide (CuO) with a monoclinic structure and nanoflower morphology, synthesized using the sol–gel and wet chemical routes, with copper foil as the substrate. The prepared nanostructures have been characterized by x-ray diffraction (XRD), scanning electron microscopy (SEM), and Fourier-transform infrared (FTIR) spectroscopy. The XRD results confirm the monoclinic structure of the CuO nanostructures and the SEM analysis reveals nanoflower formation via the wet chemical route, while the sol–gel route produces bulk morphology. We have found that the unique combination of a monoclinic structure and nanoflower morphology is advantageous for use as a promising electrode material in supercapacitor applications. The electrochemical properties in three electrode system using the obtained nanostructures have been investigated through cyclic voltammetry (CV), galvanostatic charge–discharge (GCD), and electrochemical impedance spectroscopy (EIS) in 6.0 M KOH electrolyte. The CuO nanostructured electrode material prepared via the wet chemical route (Sample 3) demonstrates a maximum specific capacitance of 453 F/g at a scan rate of 2 mV/s. These results indicate that CuO is a potential electrode material for supercapacitor applications.

Autonomous underwater vehicles (AUVs) are highly nonlinear, coupled, uncertain, and time‐varying mechatronic systems that inevitably suffer from uncertainties and environmental disturbances. This study presents an intelligent hybrid fractional‐order fast terminal sliding mode controller that utilizes the positive aspects of a model‐free control approach, designed to enhance the tracking control of AUVs. Using a nonlinear fractional‐order fast terminal sliding manifold, the proposed control approach integrates intelligent hybrid sliding mode control with fractional calculus to guarantee finite‐time convergence of system states and provide explicit settling time estimates. The nonlinear dynamics of the AUVs is modeled using radial basis function neural networks, while bound on uncertainties, external disturbances, and the reconstruction errors are accommodated by the adaptive compensator. By using a fast terminal‐type sliding mode reaching law, the controller exhibits enhanced transient response, resulting in robustness and finite‐time convergence of tracking errors. Using fractional‐order Barbalat's lemma and the Lyapunov technique, the stability of the control scheme is validated. The effectiveness of the proposed control scheme is validated by a numerical simulation study, which also shows enhanced trajectory tracking performance for AUVs over existing control schemes. This hybrid technique addresses the complicated nature of AUV dynamics in unpredictable circumstances by utilizing the advantages of model‐free intelligent control and fractional calculus.

The objective of this study is to numerically investigate and compare the characteristics of two distinct hybrid nanofluids EG-MoS 2 -SiO 2 and H 2 O-Cu-Al 2 O 3 flowing steadily over a channel created by two non-parallel absorbent porous walls. Further, the considered fluid flow is under the influence of exponential space-based heat source, viscous dissipation, Joule heating, radiation, and external magnetic field. The nonlinear partial differential equations and subjecting boundary conditions of the flow are transformed into a system of nonlinear ordinary differential equations through suitable similarity transformations and are solved by combining the shooting technique with the traditional Runge–Kutta method. The consequential results are produced utilizing MATLAB software. The comparisons of the velocity profiles, temperature profiles, surface drag, and Nusselt number for both hybrid nanofluids are illustrated as graphs. The findings indicate that H 2 O-Cu-Al 2 O 3 exhibits better velocity profiles, EG-MoS 2 -SiO 2 displays enhanced temperature profiles, while H 2 O-Cu-Al 2 O 3 has improved skin friction and Nusselt number. It is noteworthy to mention that numerous industries, including manufacturing, power generation, chemical processes, microelectronics, and transportation, depend on improving heat transfer coefficients. Hence, the significance and novelty of this work lie in the evaluation of optimization and sensitivity analysis of the EG-MoS 2 -SiO 2 hybrid nanofluid to enhance heat transmission using Response Surface Methodology.

To identify infections caused by the COVID-19 virus, specifically those leading to Pneumonia. The datasets used included images of infected individuals, X-rays of the Chest, and standard non-COVID-19 X-ray images of the chest. In our research, X-ray chest images were utilized to detect Pneumonia caused by COVID-19. The dataset was collected from the medical database of Johns Hopkins University (USA). To extract detailed features and characteristics from the input images and aid in detecting COVID-19 induced pneumonia cases, a lightweight Stacked Shallow Convolutional Neural Network (CNN) was implemented in the proposed work. An overall sample size of 2292 input images was considered, with 542 COVID-19 infected images and 1266 non-COVID-19 images used for training. Similarly, for testing, 167 COVID-19 infected images and 317 non-COVID-19 infected images were used. Our proposed model was validated against an established Stacked Shallow CNN to analyze its accuracy. It demonstrated that the proposed framework achieved an accuracy of 98.76%, with each class accuracy of 98.42% for diagnosing standard cases and 99.40% for diagnosing infected COVID-19 cases. In light of the ongoing research in enhancing the Deep Neural Network model (DNN) to accomplish better accuracy, the CNN architecture presented in this paper introduces a Stacked Shallow Learning approach. This leads to a significant enhancement in both accuracy and efficiency, particularly in computation and response time. Our work outlines a lightweight architecture that is instrumental in expediting the diagnosis of COVID-19 by streamlining response times effectively. Our proposed model also has limitations, such that it cannot hold a large-scale unbalanced dataset and has less scope for pre-processing the data set. Furthermore, the suggested model cannot capture the lung region impacted by the COVID-19 infection.