Browse Items (14421 total)
Sort by:
-
Detection and identification of un-uniformed shape text from blurred video frames
The identification and recognition of text from video frames have received a lot of attention recently, that makes many computer vision-based applications conceivable. In this study, we modify the picture mask and the original identification of the mask region convolution neural network and permit detection in three levels, including holistic, sequence, and at the level of pixels. To identify the texts and determine the text forms, semantics at the pixel and holistic levels can be used. With masking and detection, existences of the character and the word are separated and recognised. In addition, text detection using the results of 2-D feature space instance segmentation is done. Moreover, we explore text recognition using an attention-based optical character recognition (OCR) method with mask region convolution neural networks (R-CNN) to address and detect the problem of smaller and blurrier texts at the sequential level. Using attribute maps of the word occurrences in sequence to seq, the OCR method calculates the character sequence. At last, a fine-grained learning strategy is proposed to constructs models at word level using the annotated datasets, resulting in the training of a more precise and reliable model. The well-known benchmark datasets ICDAR 2013 and ICDAR 2015 are used to test our suggested methodology. 2024, Institute of Advanced Engineering and Science. All rights reserved. -
Hybrid Deep Learning Framework for Continuous Blood Glucose Monitoring and Gestational Diabetes Risk Prediction
Gestational diabetes mellitus (GDM) affects almost 10%-12% of pregnancies worldwide, threatening maternal and fetal life. Continuous glucose monitoring (CGM) forms the backbone of managing GDM, and the current methodologies largely disregard physiological and behavioral factors, thereby greatly reducing accuracy and clinical interpretability. Methods: A hybrid deep learning framework was developed by fusing CGM with multi-sensing modality data, including heart rate, activity levels, sleep patterns, and dietary intake. For data preprocessing, Kalman filtering was applied for temporal alignment, adaptive normalization provided outlier handling and imputation, while the CNN-BiLSTM backbone with attention was harnessed for feature extraction. A Multi-Task Attention Fusion Network (MTAFN) was used to predict glucose values and classify GDM risk simultaneously, while SHAP and dynamic smoothing contributed to interpretability sets. Results: The framework was validated on an extended OhioT1DM dataset with adaptations for pregnancy. It reached a glucose prediction RMSE of 9.8 mg/dL and a GDM risk classification accuracy of 93%. Compared to competitive approaches, the present solution attained a 25% better accuracy on interpretability and an improvement in sensitivity and specificity of about 4-6% across various physiological conditions. Discussion: The use of multi-sensing data increased prediction robustness by capturing complex physiological dependencies. The SHAP-based interpretability justified the predictions through a physiological lens. With an attention mechanism for feature weighting, it was possible to identify crucial variables like meal intake and nighttime variability in the workflow sets. Conclusion: The hybrid framework proposed here is reliable for clinically interpretable continuous glucose monitoring and GDM risk predictions. Its application with high reliability can lead to integrating it within clinical protocols for real-time maternal care sets. 2026, The Author(s). This is an open access article distributed under the terms of the Creative Commons Attribution 4.0 International Public License (CC-BY 4.0), a copy of which is available at: https://creativecommons.org/licenses/by/4.0/legalcode. This license permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. -
Parental Attachment, Perceived Parental-Partner Similarity, and Relationship Satisfaction among Indian Emerging Adults
Theories of mate selection debate about whether people tend to choose partners based on similarities to their parents. The present study aimed to address whether a similarity in how people perceive their parents and their partners is associated with the relationship between parental attachment and relationship satisfaction by adopting a template-matching framework. Participants were urban, emerging adults in India (n = 263, 137 male and 126 female) who were measured for how they perceive the traits of a parental figure, traits of a partner, attachment to the parent, and relationship satisfaction with the partner. Data analysis was conducted using correlations, linear regressions, and moderation analyses. Findings show that perceived neuroticism of parents was associated with perceived neuroticism of the partner. Additionally, perceptions of neuroticism of parents predicted neuroticism in partners. Perceived agreeableness, neuroticism, and openness to experience moderated the relationship between parental attachment and relationship satisfaction. A gender difference with a small effect size in perceptions of similarity was observed for openness to experience and agreeableness. Finally, perceived agreeableness also moderated the relationship between parental attachment and relationship satisfaction for men and women separately. However, for men, perceived neuroticism also significantly moderated this relationship. The findings imply that, to an extent, the more emerging adults perceive similarities of certain traits in their parents and partners, the higher likelihood that their attachment to their parent predicts relationship satisfaction with their partner. Limitations and future directions have been discussed. The Author(s) under exclusive licence to National Academy of Psychology (NAOP) India 2024. -
Understanding blame attributions in rape among legal professionals
Rape in the Indian context, is a prominent issue, greatly influenced by socio-cultural values and beliefs. Victim blaming and the concept of an ideal victim is a social evil that makes life difficult for rape survivors. What would happen if officials responsible for providing justice possess this tendency? The study aimed at understanding this question through a qualitative study on eight legal professionals including two magistrates and six advocates. The data obtained was analysed using thematic network analysis as well as content analysis. It was observed that victim blaming was present in the responses given, but blame was directed onto other factors as well. Victim blaming varied with victim characteristics and blame was greater in case of acquaintance rape rather than stranger rape. Culture based stereotypes, sex roles and rape myths were observed and seemed to affect the way they made decisions. A more extensive study in future including a broader sample and professionals from different administrative realms can help understand the issue better. 2019 International Journal of Criminal Justice Sciences (IJCJS). -
Impact of Voluntary Disclosure on Valuation of Firms: Evidence from Indian Companies
This article investigates the effect of voluntary corporate disclosures on the firm value from the market value perspective. Financial reporting includes disclosures as prescribed by regulators, but few companies go beyond mandatory requirements and provide additional information voluntarily. This study empirically tests the extent of such voluntary disclosures using Corporate Voluntary Disclosure Index containing 81 items of both financial and non-financial information and panel data regression to test the hypotheses. The sample for this study is the non-financial companies in the BSE 100 Index and the period is five financial years from 20102011 to 20142015. This study finds a positive association between voluntary disclosures and firm value as measured by Tobins Q. Especially the market gives a higher valuation for companies disclosing optional information on social and environmental, corporate governance and financial information. This finding has a significant implication for emerging economies like India and it supports various disclosure theories such as agency, stakeholders and positive accounting theories. 2020 Management Development Institute. -
Do social and environmental disclosures increase firm value evidence from indian companies
There is a clear shift in the way the companies report their performance through the communications with their stakeholders. Moving from mere profit, the companies are increasingly showing their non-financial performance in terms of sustainability and social responsibility. Companies not only want to just spend on sustainability, but also like to project their activities to gain image among the stakeholders; more often with a separate set of report called corporate sustainability report, which is based on the triple bottom-line (profit, people, and planet). This study focused on understanding the corporate social and environmental reporting trends of Indian non-financial companies and the impact on market valuation. The sample constituted of companies in the BSE-100 index and data for 5 financial years - from FY2010 to FY2014 - were used. This period was chosen as it witnessed several regulatory changes in the triple bottom line reporting in the form of new Companies Act, 2013 and Clause 55 of the listing agreement. Paired 'f' test and panel data regression model were used for analyzing the data. This study found that the level of social and environmental disclosures has significantly improved post business responsibility reporting and positively significantly influenced market valuation. -
Strategies for facilitating listening skills among foreign language learners in US Universities
Developing from the thesis that understanding is the key to any and all meaningful conversation/s, this study focuses on the facilitation of listening skills among foreign language learners. It is conducted with the objective to find out the most effective ways in which an instructor can enable the development of listening skills among the learners of a foreign language. This paper reports the findings of an empirical study which followed a cross-sectional research design and employed a survey method to elicit the data. Twenty-seven Foreign Language Instructors/ Associate Instructors teaching around thirteen different foreign languages across sixteen different universities in the United States of America participated and reported to a survey on effective pre-listening, listening, and post-listening tasks, activities, and strategies which they found to be the most powerful in their respective classrooms. Thirteen of the Seventeen strategies and or/ tasks which were provided in the Strategies for Facilitating Listening (SFL) questionnaire were rated to be highly effective in the facilitation of the development of listening skills among the learners. The paper after discussing the efficacies of the strategies and tasks at hand ends by analyzing the pedagogical implications of the findings. 2020 The authors and IJLTER.ORG. All rights reserved. -
The mobility paradigm in higher education: a phenomenological study on the shift in learning space
The study, through the framework of mobility and space, explores the phenomenon of multiple shifts in learning spaces induced by COVID-19. The Interpretative Phenomenological Approach (IPA) is adopted to document the experiences and perceptions of learners caught within these spatial shiftsphysical, online, and hybrid. Online interviews were conducted with six first-year undergraduate and three first-year postgraduate students enrolled at the department of English and Cultural Studies in a Southern Indian University. Some of the dominant patterns emerging from the accounts of the participants are (1) the changing perception of conducive learning space, (2) the changing perceptions and roles of various classroom actors, and (3) the evolving nature of the learners and the learning process. The study utilizes the framework of mobility to locate the stage of embodied skill acquisition of the participants within the online learning space and illuminates the possibilities offered by this paradigm within the context of higher education. Some of the insights gained through the study include a changing perception of the conventional built classroom space, a notable preference towards a complete online or offline mode as opposed to the hybrid mode, and a transition towards self-directed learning. The study argues that these implications are highly pertinent and can significantly shape the way pedagogues and researchers engage with the various modes of learningphysical, online, and hybridand the future of higher education that is shaped by technology-enabled learning. 2021, The Author(s). -
Why learning space matters: a script approach to the phenomena of learning in the emergency remote learning scenario
The study focuses on how the notion of learning space is perceived and experienced by learners in the Emergency Remote Learning (ERL) scenario. In doing so, the lived experiences of remote learners who were abruptly shifted to a completely online learning space due to the pandemic COVID-19 in the Indian higher education system are documented. Online interviews were conducted with eight undergraduate and four postgraduate students of English and Cultural Studies, enrolled at a Southern Indian university, and their responses were explicated using the Interpretive Phenomenological Analysis (IPA) approach. The interviews revealed that the phenomenon of ERL is shaped by dissonance informed by the absence of a familiar learning space. Often the patterns of this dissonance were marked by (1) the perception of learning and learning space, (2) the lack of intimacy in learning and learning space, (3) the negotiations made for learning and the space of learning in ERL, and (4) the challenges to cope with the responsibilities of the ERL scenario. Further, the script approach was applied to analyse the data and the analysis revealed an expansion of the existing internal scripts that were based on previous learning experiences of the learners. The study thus establishes the centrality of space in the process of learning and points out how the lack of a familiar learning space is linked to the absence of internal scripts that considerably impact learning. The study concludes by discussing the possibilities of application of script approach to effectively incorporate the aspect of learning space in new pedagogies and learning models as Blended Learning (BL) and Online Learning (OL) become the new normal worldwide. 2021, Beijing Normal University. -
Understanding feminism on online platforms: Exploration and analysis of two online platforms
This article explores how feminism is practised and communicated on digital plat-forms. Feminism in India and Khabar Lahariya are two online platforms studied with interviews of respondents to understand how the online spaces are used for knowledge sharing that take feminist perspective. New media opened up spaces for people to communicate from any part of the world, create media content and circulate it. Visibility, privacy, accessibility and risks are negotiated by the report-ers and content creators to produce alternative cultural production from an inter-sectional feminist standpoint. 2022 Intellect Ltd Article. -
The Impact and Inheritance of Operating Leverage: A Study with Two Pharmaceutical Companies
International Journal of Management Research and Technology, Vol-7 (2), pp. 145-154. ISSN-0974-3502 -
LRE-MMF: A novel multi-modal fusion algorithm for detecting neurodegeneration in Parkinson's disease among the geriatric population
Parkinson's disease (PD) is a prevalent neurological disorder characterized by progressive dopaminergic neuron loss, leading to both motor and non-motor symptoms. Early and accurate diagnosis is challenging due to the subtle and variable nature of early symptoms. This study aims to address these diagnostic challenges by proposing a novel method, Localized Region Extraction and Multi-Modal Fusion (LRE-MMF), designed to enhance diagnostic accuracy through the integration of structural MRI (sMRI) and resting-state functional MRI (rs-fMRI) data. The LRE-MMF method utilizes the complementary strengths of sMRI and rs-fMRI: sMRI provides detailed anatomical information, while rs-fMRI captures functional connectivity patterns. We applied this approach to a dataset consisting of 20 PD patients and 20 healthy controls (HC), all scanned with a 3 T MRI. The primary objective was to determine whether the integration of sMRI and rs-fMRI through the LRE-MMF method improves the classification accuracy between PD and HC subjects. LRE-MMF involves the division of imaging data into localized regions, followed by feature extraction and dimensionality reduction using Principal Component Analysis (PCA). The resulting features were fused and processed through a neural network to learn high-level representations. The model achieved an accuracy of 75 %, with a precision of 0.8125, recall of 0.65, and an AUC of 0.8875. The validation accuracy curves indicated good generalization, with significant brain regions identified, including the caudate, putamen, thalamus, supplementary motor area, and precuneus, as per the AAL atlas. These results demonstrate the potential of the LRE-MMF method for improving early diagnosis and understanding of PD by effectively utilizing both sMRI and rs-fMRI data. This approach could contribute to the development of more accurate diagnostic tools. 2024 The Authors -
Machine Learning-based Currency Information Retrieval for Aiding the Visually Impaired People
Paper currency is one of the most in-demand and long-established payment modes across the globe. People suffering from visual disabilities often face difficulties while handling paper currencies. Over the years, assisting technology has been rekindling itself to serve the aged and disabled person more aptly. Image processing methods and other sophisticated technologies, like Artificial Intelligence, Deep Learning, etc., can be employed to identify banknotes and fetch other valuable pieces of information from them. This paper proposes a framework that focuses on an integrated approach to retrieving data from the paper currency's uploaded image. The current version of the framework focuses on identifying the authenticity of the paper currency and classifying it according to its value. This work is an initiative to help visually impaired people to use paper currencies without assistance from other individuals and support them in living independently. 2021 IEEE. -
Classification of Soil Images using Convolution Neural Networks
Classification of soil is crucial for the agricultural domain as it is an essential task in geology and engineering domains. Various procedures are proposed to classify soil types in the literature, but many of them consumed much time or required specially designed equipments/applications. Classification of soil involves the accounting of various factors due to its diversified nature. It can be observed that several critical domain-oriented decisions often depend on the type of soil like farmers might be benefitted from knowing the kind of soil to choose crops accordingly for cultivation. We have employed different Convolution Neural Network (CNN) architectures to identify the soil type accurately in real-time. This paper describes the comparative evaluation in terms of performances of various CNN architectures, namely, ResNet50, VGG19, MobileNetV2, VGG16, NASNetMobile, and InceptionV3. These CNN models are used to classify four types of soils: Clay, Black, Alluvial, and Red. The performance of the ResNet50 model is the best with a training accuracy and training loss of 99.47% and 0.0252, respectively compared to other competing models considered in this paper. 2021 IEEE. -
An AI Approach to Pose-based Sports Activity Classification
Artificial intelligence systems have permeated into all spheres of our life-impacting everything from our food habits to our sleep patterns. One untouched area where such intelligent systems are still in their infancy is sports. There has not been enough indulgence of AI techniques in sports, and most of the works are carried on manually by coaching staff and human appointees. We believe that intelligent systems can make coaching staff's work easier and produce findings that the human eye can often overlook. Here, we have proposed an intelligent system to analyze the beautiful game of tennis. With the use of computer vision architecture Detectron2 and activity-based pose estimation and subsequent classification, it can identify an action from a tennis shot (activity). It can produce a performance score for the player based on pose and movement like forehand and backhand. It can also be used to understand and evaluate the strengths and weaknesses of the player. The proposed approach provides a piece of valuable information for a player's performance and activity detection to be used for better coaching. The study achieves a classification accuracy of 98.60% and outperforms other SOTA CNN models. 2021 IEEE -
Enhancement of Image Classification Using Transfer Learning and GAN-Based Synthetic Data Augmentation
Plastic bottle recycling has a crucial role in environmental degradation and protection. Position and background should be the same to classify plastic bottles on a conveyor belt. The manual detection of plastic bottles is time consuming and leads to human error. Hence, the automatic classification of plastic bottles using deep learning techniques can assist with the more accurate results and reduce cost. To achieve a considerably good result using the DL model, we need a large volume of data to train. We propose a GAN-based model to generate synthetic images similar to the original. To improve the image synthesis quality with less training time and decrease the chances of mode collapse, we propose a modified lightweight-GAN model, which consists of a generator and a discriminator with an auto-encoding feature to capture essential parts of the input image and to encourage the generator to produce a wide range of real data. Then a newly designed weighted average ensemble model based on two pre-trained models, inceptionV3 and xception, to classify transparent plastic bottles obtains an improved classification accuracy of 99.06%. 2022 by the authors. Licensee MDPI, Basel, Switzerland. -
Unleashing human potential: Integrating cognitive behavioral neuroscience into HR strategies
The world of work is transforming, driven by insights from the frontiers of science. Human resource (HR) practices are no longer limited to traditional methods and increasingly incorporate knowledge from disciplines like Cognitive Behavioral Neuroscience (CBN). By understanding how our brains work, we can design HR practices that enhance employee well-being, engagement, and, ultimately, performance. Drawing from neuroscientific research on decision-making, communication, stress, learning, motivation, and workplace design, this chapter delves into the intersection of CBN and HR, offering evidence-based practices that support a thriving workforce. This interdisciplinary approach holds promise for maximizing human potential in the context of the modern workplace. 2024 by IGI Global. All rights reserved. -
Nutrition Analysis: A Data-Driven Approach for Optimizing Individual Dietary Choices
Maintaining good health, avoiding illnesses, and controlling ailments like diabetes, heart disease, and obesity all depend on proper diet. With the use of nutrition analysis, people can better understand their dietary requirements and choose foods that will support a healthy lifestyle. The goal of this studys data-driven approach to nutrition analysis is to maximize each persons dietary selections. Individualized recommendations are made for balanced nutrition by utilizing top-of-the-machine learning techniques to examine food patterns, nutrient consumption, and health effects. Food items are categorized based on their nutritional characteristics, and potential health effects are predicted using algorithms like Gradient Boosting, Multi-Layer Perceptron (MLP), Random Forest, Support Vector Classifier (SVC), Gaussian Nae Bayes (GNB), Decision Tree, Stochastic Gradient Descent (SGD), Linear Discriminant Analysis (LDA), and K-Nearest Neighbors (KNN). These models evaluated the relationship between dietary practices and nutritional needs. The final outcome is a comprehensive system that enables people to make knowledgeable food choices and optimize their nutrition in a way that promotes their overall health. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
Advanced Botnet Detection Using Hybrid Machine Learning Models
The improvement of computer network systems, cyberattacks that take advantage of system flaws have increased, resulting in significant monetary losses, business interruptions, harm to one's reputation, and legal repercussions. This research examines nine attack types, those are Fuzzers, Shellcode, Generic, Worms, Analysis, Normal, DoS, Exploits, Backdoor, and Reconnaissance. Botnet assaults are attacks in which a single operator controls several networked devices. The research study examines several models, such as Random Forest, XGBoost Classifier, Logistic Regression, and Decision Tree, to improve detecting skills. By utilizing the advantages of both methods, the suggested ERFwXGBoost (Enhanced Random Forest with XGBoost) model, which blends Random Forest and XGBoost, exhibits remarkable performance. Notably, first they analyze the accuracy, then precision value is also measured, third will measure recall value, and then finally F1 score of the ERFwXGBoost model are all impressively 0.98. In addition to outperforming individual models, our hybrid technique offers a reliable and effective way to detect different kinds of botnet attacks. The research emphasizes how well these models work together to boost overall system security against advanced cyber threats and greatly increase detection accuracy. 2025 IEEE.