Browse Items (14421 total)
Sort by:
-
Effect of Mentha arvensis enriched diet to promote the growth and immune response of Clarias batrachus against Aeromonas hydrophila challenge
The study was conducted to investigate the effects of fish fed diet Mentha arvensis extract on growth performance, non-specific immunity and expression of some immune-related genes and resistance to Aeromonas hydrophila in Clarias batrachus. Five diets were formulated with 0, 1, 2, 3, and 4% of M. arvensis leaf extract. The results indicated that, compared to the control groups, 2-4% dietary inclusion increased growth and feed consumption. In the dietary inclusion of 3-4% M. arvensis extract groups were increased relative on weight gain, specific growth rate, RBC, WBC, total hemocyte counts, total protein, globulin than control. Fed diet supplements with 3% mint-extract increased the total protein, WBC and globulin and phagocytic indexes and lysozyme activity increased at the 2, 3 and 4% of mint groups relative to the control. The PCR analysis showed that TNF, IL-1, MyD88, and TLRs were increased in the 2-4% fed diet M. arvensis extract groups than the control. These results suggest that 3% of M. arvensis extract significantly influences the immunomodulatory activity and immune-specific genes of C. batrachus. 2024 The Authors -
Multitask EfficientNet affective computing for student engagement detection
In the realm of education, feedback emerges as a pivotal component, serving to foster engagement and interaction while also facilitating the refinement of teaching methods to capture and maintain student attention. Traditional classroom assessment methods often struggle to accurately gauge the degree of comprehension among students during lectures, relying on manual comment collection that inherently carries the risk of inaccuracies. In response to this challenge, a novel system has been proposed, harnessing the power of Facial Emotion Recognition (FER) technology to capture student feedback. Within this framework, students are given a unique avenue to convey their emotions and reactions, employing facial expressions and gestures as the means to communicate. This innovative approach enables the analysis of students emotional responses and thereby provides invaluable insights into their comprehension levels, as well as the overall quality and engagement experienced during lectures. The approach takes shape through the utilization of Computer Vision techniques, with a particular focus on an unobtrusive methodology for assessing students overall engagement. Overcoming limitations of traditional assessment, our approach integrates compound scaling, employing the proposed Multitask EfficientNetB0 model recognized for its proved accuracy in emotion recognition (95.7%) and behavior analysis (96.3%) across diverse datasets (DAiSEE, iSED, iSAFFE). The behavioral classification system categorizes students into Engaged and Disengaged classes within a multi-class framework, providing nuanced insights into comprehension and Student engagement. Assessment metrics, including ROC Curves, Precision, Recall, and F1-Score, ensure a thorough evaluation. Our systems adaptability is demonstrated across varied educational environments, showcasing real-world efficacy in classrooms, laboratories, and seminar halls. The inclusion of MTCNN enhances face detection capabilities, facilitating robust analysis in dynamic scenarios. Expanding its applicability, the model has been put to the test in a range of educational settings, including classrooms, laboratory environments, and seminar halls, offering dual-capability analysis of both emotions and behavior. This comprehensive approach yields nuanced insights into student engagement and interaction, and its performance has been validated through real-world deployment within classrooms and seminars The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024. -
Multimodal emotional analysis through hierarchical video summarization and face tracking
The era of video data has fascinated users into creating, processing, and manipulating videos for various applications. Voluminous video data requires higher computation power and processing time. In this work, a model is developed that can precisely acquire keyframes through hierarchical summarization and use the keyframes to detect faces and assess the emotional intent of the user. The key-frames are used to detect faces using recursive Viola-Jones algorithm and an emotional analysis for the faces extracted is conducted using an underlying architecture developed based on Deep Neural Networks (DNN). This work has significantly contributed in improving the accuracy of face detection and emotional analysis in non-redundant frames. The number of frames selected after summarization was less than 30% using the local minima extraction. The recursive routine introduced for face detection reduced false positives in all the video frames to lesser than 2%. The accuracy of emotional prediction on the faces acquired through the summarized frames, on Indian faces achieved a 90%. The computational requirement scaled down to 40% due to the hierarchical summarization that removed redundant frames and recursive face detection removed false localization of faces. The proposed model intends to emphasize the importance of keyframe detection and use them for facial emotional recognition. 2021, The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature. -
EMONET: A Cross Database Progressive Deep Network for Facial Expression Recognition
Recognizing facial features to detect emotions has always been an interesting topic for research in the field of Computer vision and cognitive emotional analysis. In this research a model to detect and classify emotions is explored, using Deep Convolutional Neural Networks (DCNN). This model intends to classify the primary emotions (Anger, Disgust, Fear, Happy, Sad, Surprise and Neutral) using progressive learning model for a Facial Expression Recognition (FER) System. The proposed model (EmoNet) is developed based on a linear growing-shrinking filter method that shows prominent extraction of robust features for learning and interprets emotional classification for an improved accuracy. EmoNet incorporates Progressive- Resizing (PR) of images to accommodate improved learning traits from emotional datasets by adding more image data for training and Validation which helped in improving the model's accuracy by 5%. Cross validations were carried out on the model, this enabled the model to be ready for testing on new data. EmoNet results signifies improved performance with respect to accuracy, precision and recall due to the incorporation of progressive learning Framework, Tuning Hyper parameters of the network, Image Augmentation and moderating generalization and Bias on the images. These parameters are compared with the existing models of Emotional analysis with the various datasets that are prominently available for research. The Methods, Image Data and the Fine-tuned model combinedly contributed in achieving 83.6%, 78.4%, 98.1% and 99.5% on FER2013, IMFDB, CK+ and JAFFE respectively. EmoNet has worked on four different datasets and achieved an overall accuracy of 90%. 2020. All Rights Reserved. -
Engagement Detection through Facial Emotional Recognition Using a Shallow Residual Convolutional Neural Networks
Online teaching and learning has recently turned out to be the order of the day, where majority of the learners undergo courses and trainings over the new environment. Learning through these platforms have created a requirement to understand if the learner is interested or not. Detecting engagement of the learners have sought increased attention to create learner centric models that can enhance the teaching and learning experience. The learner will over a period of time in the platform, tend to expose various emotions like engaged, bored, frustrated, confused, angry and other cues that can be classified as engaged or disengaged. This paper proposes in creating a Convolutional Neural Network (CNN) and enabling it with residual connections that can enhance the learning rate of the network and improve the classification on three Indian datasets that predominantly work on classroom engagement models. The proposed network performs well due to introduction of Residual learning that carries additional learning from the previous batch of layers into the next batch, Optimized Hyper Parametric (OHP) setting, increased dimensions of images for higher data abstraction and reduction of vanishing gradient problems resulting in managing overfitting issues. The Residual network introduced, consists of a shallow depth of 50 layers which has significantly produced an accuracy of 91.3% on ISED & iSAFE data while it achieves a 93.4% accuracy on the Daisee dataset. The average accuracy achieved by the classification network is 0.825 according to Cohens Kappa measure. 2020, Intelligent Engineering & System. All rights reserved. -
Unobtrusive Engagement Detection through Semantic Pose Estimation and Lightweight ResNet for an Online Class Environment
Analysing student engagement in a class through unobtrusive methods enhances the learning and teaching experience. During these pandemic times, where the classes are conducted online, it is imperative to efficiently estimate the engagement levels of individual students. Helping teachers to annotate and understand the significant learning rate of the students is critical and vital. To facilitate the analysis of estimating the engagement levels among students, this paper proposes a dual channel model to precisely detect the attention level of individual students in a classroom. Considering the possible inaccuracy of emotion recognition, a dual channel is configured with a Lightweight ResNet model for macro-level attention estimation and a 3d pose estimation using Euler angles for Pitch, yaw and roll that is trained, validated and tested on the Daisee database. The Emotional detection extracts the context of Engaged, frustrated, confused and disgust as higher levels of classroom attention cognition while the facial pose coordinates provide the real-time movement of the faces in the video to provide a series of engaged and disengaged coordinates. The Lightweight ResNet Model achieves a 95.5% accuracy and the Pose estimation test is able to distinguish the test videos at 92% as Engaged and Bored on the Daisee Dataset. The Overall Accuracies using the Dual channel was curated to 87%. 2023 Scrivener Publishing LLC. -
Security and privacy aspects in intelligence systems through blockchain and explainable AI
Explainable AI (XAI) is a method of creating artificial intelligence (AI) systems that are transparent and understandable to humans. By allowing people to understand how the system arrived at its conclusions or suggestions, XAI systems strive to make AI more accountable, trustworthy, and ethical. Responsibility, trust, ethics, regulation, and innovation are some of the societal ramifications of XAI. By making AI systems more transparent, XAI fosters accountability. This means that consumers will be able to understand how the system made its decisions and hold it accountable if something goes wrong. By making the decision-making process more transparent, XAI fosters trust between people and AI systems. This boosts user trust in the system and encourages wider adoption of AI technologies. It also contributes to the ethical design of AI systems by making the decision-making process public in order to uncover and mitigate biases and other ethical issues that may occur in AI systems. It aids regulators and policymakers in understanding and regulating AI systems. XAI gives insight into how AI systems operate, which can assist regulators in developing laws that promote ethical and responsible AI use. Because XAI can help developers better and innovate new systems by making it easier for them to design new AI systems and by providing insights into how AI systems work. The proposed chapter will focus on important aspects of algorithmic bias and changing notions of privacy in XAI, which will necessitate the need for AI systems that can adapt accountability, trust, ethics, and compliance with regulations, as well as produce better innovation that can benefit humanity. More openness, greater control over personal data, new types of data privacy, and newer privacy networks are all required. To address algorithmic bias in XAI, it is critical to build the system so that it is aware of the possibility of bias and actively mitigates it. This can involve employing diverse and representative data, inspecting the system for unwanted features, offering detailed explanations, and incorporating a wide range of stakeholders in the system's development and deployment. The envisaged report provides a framework that combines XAI and blockchain to provide a secure and transparent way to store and track the provenance of data used by XAI systems, validate the performance of AI models stored on the blockchain on decentralized systems so that the models are stored and executed on a distributed network of nodes rather than a centralized server, and create a token-based economy that encourages data sharing and AI development. Tokens can be used to compensate individuals and organizations who contribute data or algorithms to the blockchain or who employ AI models stored on the blockchain. Overall, the combination of XAI and blockchain can lead to more trustworthy, transparent, and decentralized AI systems. This approach can have a significant impact on various industries such as finance, healthcare, and supply chain management by increasing efficiency, reducing costs, and improving data privacy and security. 2024 Elsevier Inc. All rights reserved. -
Multitask EfficientNet affective computing for student engagement detection
In the realm of education, feedback emerges as a pivotal component, serving to foster engagement and interaction while also facilitating the refinement of teaching methods to capture and maintain student attention. Traditional classroom assessment methods often struggle to accurately gauge the degree of comprehension among students during lectures, relying on manual comment collection that inherently carries the risk of inaccuracies. In response to this challenge, a novel system has been proposed, harnessing the power of Facial Emotion Recognition (FER) technology to capture student feedback. Within this framework, students are given a unique avenue to convey their emotions and reactions, employing facial expressions and gestures as the means to communicate. This innovative approach enables the analysis of students emotional responses and thereby provides invaluable insights into their comprehension levels, as well as the overall quality and engagement experienced during lectures. The approach takes shape through the utilization of Computer Vision techniques, with a particular focus on an unobtrusive methodology for assessing students overall engagement. Overcoming limitations of traditional assessment, our approach integrates compound scaling, employing the proposed Multitask EfficientNetB0 model recognized for its proved accuracy in emotion recognition (95.7%) and behavior analysis (96.3%) across diverse datasets (DAiSEE, iSED, iSAFFE). The behavioral classification system categorizes students into Engaged and Disengaged classes within a multi-class framework, providing nuanced insights into comprehension and Student engagement. Assessment metrics, including ROC Curves, Precision, Recall, and F1-Score, ensure a thorough evaluation. Our systems adaptability is demonstrated across varied educational environments, showcasing real-world efficacy in classrooms, laboratories, and seminar halls. The inclusion of MTCNN enhances face detection capabilities, facilitating robust analysis in dynamic scenarios. Expanding its applicability, the model has been put to the test in a range of educational settings, including classrooms, laboratory environments, and seminar halls, offering dual-capability analysis of both emotions and behavior. This comprehensive approach yields nuanced insights into student engagement and interaction, and its performance has been validated through real-world deployment within classrooms and seminars The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024. -
Forecasting Flight Delays with a Multilayered Memory Fusion Network
One of the biggest worldwide sectors is aviation, hence delays in flight services not only perturb customers but also result in large losses for airlines. Forecasting these delays is still difficult because of the erratic character of elements like climate. Accurate projections are challenging even using accepted analytical methods. This work employs sophisticated deep learning methods to enhance the forecast of aircraft delays - more especially, those resulting from weather-related causes.We investigate their effect on aircraft delays using datasets from both the United States and India, including meteorological fluctuations. Built on a Multilayered Memory Fusion Network, the model captures intricate temporal patterns in the data by merging Bidirectional LSTM (Bi-LSTM) and Long Short-Term Memory (LSTM). This network generates more accurate forecasts and is meant to effectively manage several factors. For the United States dataset, the proposed network attained a Mean Absolute Error (MAE) score of 72.41 and Root Mean Square Error (RMSE) scores of 118.87 and 11.83 and 21.82 for India respectively. Our deep learning methodology clearly predicts flight delays as these performance measures are far better than those attained by conventional machine learning techniques, including linear regression. By using these cutting-edge algorithms, the research provides a more accurate way to forecast flight delays, hence perhaps lowering passenger discontent and airline financial losses. 2025 The Author(s). -
Quantum approaches to sustainable resource management in supply chains
Quantum computing leverages the principles of quantum mechanics to process information in ways that classical computers cannot. This capability is particularly advantageous for solving complex optimization problems that are common in supply chain management. Quantum algorithms, such as the quantum approximate optimization algorithm (QAOA) and quantum annealing, have shown promise in efficiently solving these problems by exploring numerous potential solutions simultaneously and identifying optimal strategies. The purpose of this chapter is to investigate the rapidly developing topic of quantum computing and its potential applications in managing sustainable resources within supply chains. Traditional resource allocation methods often struggle to maximize efficiency while minimizing environmental impact. However, new developments in quantum computing have opened up potentially fruitful pathways for addressing these issues. This study aims to explore how quantum computing can revolutionize through an examination of quantum algorithms, optimization approaches, and case studies. 2024 by IGI Global. All rights reserved. -
Attention-Powered Deep Learning for Employee Analytics: A Multi-Model Approach
In the ever-evolving field of human resources analytics, there is the integration of the latest techniques of machine learning that can strongly enhance decision-making. This paper introduces a revolutionary architecture for multi-model neural networks that integrate disparate networks in analyzing the background, development, performance, and engagement of an employee for all key elements of this employee. Each of the processes with attention fine-tunes the importance of features and therefore largely improves the concentration and interpretability of results. These networks are thus ensured of thorough analysis in the form of in-depth evaluation, which enables classification to be discrete and into clear performance categories. Preparation of raw data was also done with much care; we used the Employee/HR Dataset from Kaggle in order to process this raw data before its use in deep learning application. Our proposed architecture outperformed by accurately classifying the employee performance categories, with result showing a high classification accuracy of 86.49% on the test set. This study, therefore, establishes that customized neural network architectures are applicable in supporting organizations in realizing their data driven culture and in making human resource operations more efficient. 2026, Springer Science and Business Media Deutschland GmbH. All rights reserved. -
Deep learning-based diabetic retinopathy detection with advanced image segmentation and transfer learning techniques
Diabetic retinopathy (DR), a dangerous side effect of diabetes, can result in permanent blindness. This work presents a state-of-the-art deep learning-based system that uses retinal images to detect and classify DR early on. Utilizing transfer learning and pre-trained models, the system combines Django, Numpy, and Keras to improve diagnostic precision. It accurately detects DR-affected areas and delivers real-time graphical outputs for prompt medical interpretation and decision-making using the ResNet and Mask RCNN architectures. Simple picture uploads are made possible by the user-friendly interface, which lets Numpy handle data processing and preparation. To improve accuracy and reduce the amount of new data required, the system uses transfer learning and pre-trained datasets. The system's robustness and efficacy are highlighted by its evaluation, which shows its high accuracy with an overall accuracy of 95.55%, precision, recall, and F1-scores above 0.95. The suggested approach provides an affordable, effective, and scalable means of detecting DR early on; it is especially helpful in healthcare settings with limited resources. The technology has the potential to greatly enhance patient outcomes and lessen the toll that diabetic retinopathy has on both individuals and healthcare systems by enabling prompt diagnosis and treatment. 2026 Author(s). -
Effect of Gravity Modulation on the Onset of Ferroconvection in a Densely Packed Porous Layer /
IOSR Journal of Applied Physics, Vol.3, Issue 3, pp.30-40, ISSN No: 2278-4861.
The stability of a horizontal porous layer of a ferromagnetic fluid heated from below is studied when
the fluid layer is subject to a time-periodic body force.Modified Darcy law is used to describe the fluid motion.
The effect of gravity modulation is treated by a perturbation expansion in powers of the amplitude of
modulation. The stability of the system,characterized by a correction Rayleigh number,is determined as a
function of the frequency of modulation, magnetic parameters, and Vadasz number. -
Analysis and Forecasting of Crude Oil Price Based on Univariate and Multivariate Time Series Approaches
This paper discusses the notion of multivariate and univariate analysis for the prediction of crude oil price in India. The study also looks at the long-term relationship between the crude oil prices and its petroleum products price such as diesel, gasoline, and natural gas in India. Both univariate and multivariate time series analyses are used to predict the relationship between crude oil price and other petroleum products. The Johansen cointegration test, EngleGranger test, vector error correction (VEC) model, and vector auto regressive (VAR) model are used in this study to assess the long- and short-run dynamics between crude oil prices and other petroleum products. Prediction of crude oil price has also been modeled with respect to the univariate time series models such as autoregressive integrated moving average (ARIMA) model, Holt exponential smoothing, and generalized autoregressive conditional heteroskedasticity (GARCH). The cointegration test indicated that diesel prices and crude oil prices have a long-run link. The Granger causality test revealed a bidirectional relationship between the price of diesel and the price of gasoline, as well as a unidirectional association between the price of diesel and the price of crude oil. Based on in-sample forecasts, accuracy metrics such as root mean square logarithmic error (RMSLE), mean absolute percentage error (MAPE), and mean absolute square error (MASE) were derived, and it was discovered that VECM and ARIMA models can efficiently predict crude oil prices. 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
The Evolution of Forecasting Techniques: Traditional Versus Machine Learning Methods
Forecasting is used effectively and efficiently to support decision-making for the future. Over time, several methods have been created to conduct forecasting. Finding a forecasting technique with the ability to provide the best estimate of the system being modeled has always been a challenge. The selection and comparison criteria for forecasting methodologies can be organized in a variety of ways. Accurate forecasting has a great demand for various fields like weather prediction, economic condition, business forecasting, demand and supply forecasts and many more. When deciding whether to utilize a certain model to predict future events, accuracy is very important. In every field, machine learning (ML) algorithms are being used to forecast future events. These algorithms can handle more complex data and make predictions that are more accurate. Based on the least values of forecasting errors, forecasters create a model to determine the best strategy for prediction. For centuries, forecasting has been used to assist individuals in making future-related decisions. In the past, forecasts were based on intuition and experience, but as technology has advanced, so have forecasting methods. Currently, advanced ML models and methods for data analysis are used to provide forecasts. To forecast the future, these models incorporate a range of inputs, including historical data, present trends, and economic indicators. Forecasting is a vital tool for businesses to employ when making future plans. It is used in a wide range of industries, from finance to weather prediction. 2024 Sachi Nandan Mohanty, Preethi Nanjundan and Tejaswini Kar. -
Global Analysis of Quantum Technology Discourse
he study provides a thorough exploration of the global quantum technology landscape, offering valuable insights for researchers, policymakers, and industry stakeholders. It employs advanced analytical methods such as Latent Dirichlet Allocation (LDA) and Non-Negative Matrix Factorization (NMF) for topic modeling. The research focuses on understanding discussion intensity, geographical distribution, co-mentioning patterns among countries, prevalent topics, and keyword-based trends. Utilizing diverse datasets, the study employs heatmaps, network analysis, and thematic analysis to categorize textual data. Evaluation metrics like Topic Coherence and Network Centrality Measures contribute to a robust methodology.Key findings include dominant discussions on quantum computing and investment strategies, with focused attention on governmental roles in R&D and specific quantum computer research. Notably, there is a niche focus on quantum algorithmic risks in Australia. Document characteristics vary, with some blending multiple themes and others centered around a single topic. LDA topic modeling and network analysis identify key countries, showcasing global hotspots and potential collaborations in quantum technology discussions. 2024 IEEE. -
Comparative study of recommender systems
Recommendation System is a quickly progressing study area. Many new approaches are offered so far. In this particular paper we have researched on various applications of recommender system and various techniques used in recommender system like collaborative filtering, content-based filtering and hybrid filtering. Collaborative filtering is amongst the common methods utilized in recommending process. So comparative study on various collaborative filtering is done and the results are plotted graphically. 2016 IEEE. -
Travails of New Mothers Returning to Work in Corporate India: A Phenomenological Study
A womans life is a myriad of experiences and none, perhaps, leaves a more lasting impression on her than motherhood. The child-birth event along with all its highs and lows not only has a deep psychological impact on her as a person but also impacts her career in many ways. Using interpretive phenomenological analysis, we have studied the lived experience of women who returned to work in corporate settings after maternity leave. Our study found that not only do they go through an emotional upheaval during this phase, but they also see a marked shift in the way they approach their careers. A womans natural instinct to mother her child comes in conflict with another natural (and equally important) desire to succeed in the workplace. Most women in our study experienced a stalling/break in their careers after childbirth and wished they had a mentor to assist them in transitioning back to office life. Besides trying to evaluate if childbirth was perceived as a threat or potential impediment to a high-flying career, we also explored how women were treated in their work environments, and whether their coworkers helped the women to cope during this phase. While the women in our study wanted to achieve success and satisfaction both within their families and careers, they found it most challenging to do so. 2022 Journal of International Womens Studies. -
DISTANCE SPECTRUM OF TWO FAMILIES OF GRAPHS
Let H1 and H2 be two copies of the complete graph Kn, n ? 3 with vertex sets V(H1) = {v1,v2...,vn} and V(H2) = {u1,u2,...,un}. Graph ?(n,p), 1 ? p ? n-1, is obtained from the union of graphs H1 and H2 by adding edges {uivi)|i ? {1, 2...,p}}. Graph ?(n) is obtained from the union of graphs H1 and H2 by joining each vertex vi of H1 to every vertex in {u1, u2, ..., un} \ {ui}, i = 1, 2, ..., n. The adjacency spectrum of ?(n, p) and ?(n) were determined in [9]. An open problem posed in [7] was to find families of graphs of diameter greater than two, for which the adjacency and distance spectrum are both integral. To answer the open problem, the distance spectrum of the above family of graphs is calculated, and new distance equienergetic graphs are constructed in this paper. 2024 Jangjeon Research Institute for Mathematical Sciences and Physics. All rights reserved. -
Spectrum of corona products based on splitting graphs
Let G be a simple undirected graph. Three new corona products of graphs based on splitting graph of G are defined. The adjacency spectra of the three new graphs based on splitting graph of G are determined. The number of spanning trees and the Kirchoff index of the new graphs are determined using their nonzero Laplacian eigenvalues. 2023 World Scientific Publishing Company.