Browsing by Type "Conference paper"

CVDs) are the important reason of international mortality, highlighting the life-threatening need for timely heart disease recognition. Electrocardiogram (ECG) signals have long been utilized for diagnosing cardiac abnormalities, but the application of deep learning models. This study systematically evaluates various deep learning architectures, including VGG-16, MobileNet, Custom CNN, and Vision Transformer (ViT), for heart disease prediction using ECG image data. Models were trained on a labeled dataset containing ECG images. The performance of each method was assessed. Among the models, the Vision Transformer outperformed others, achieving the highest accuracy ($94.6%) and consistently high precision ($94.1%) and recall ($93.8%). The results highlight the efficiency of the ViT in capturing both local and global patterns in ECG images, offering a promising tool for scalable heart disease prediction. Additionally, a cloud-based deployment framework was developed to facilitate real-time diagnostic support in clinical settings. © 2025 IEEE.

Background: Around thirty-five percent of the population in India eat according to a vegetarian diet (Kumar and Prakash 2017). A vegetarian diet is considered healthier than a non-vegetarian diet as it has less saturated fat. However, unhealthy practices like eating packaged foods, drinking alcohol, or smoking have contributed to a growing obesity pandemic. According to current estimates, globally around 2.8 million deaths occur due to people being overweight or obese each year (WHO 2016). Anemia has been shown to be prevalent among approximately one-third of the world's population reported by WHO (Botswana Compendium, 2017). Objectives: This research aims to compare hematological, body mass index and lipid profile parameters in accordance with vegetarianism and non-vegetarianism. Its other aim is to look for effective dietary strategies for fighting lifestyle diseases. Methods: The study included a cohort of thirty vegetarian men and thirty age-matched non-vegetarian men aged 35 to 50 years who were seeking health assessments at the Master Health Check-up clinic. Demographical features were recorded along with blood tests were performed including haematology, body mass index (BMI) and lipid profile evaluations for all subj ects. Blood samples were taken after 10 hours of overnight fasting for analysing serum lipids parameters. Results: The mean±SD values of Hemoglobin, Hematocrit (PCV), Mean Corpuscular Volume (MCV), Mean Corpuscular Hemoglobin (MCH), and Mean Corpuscular Hemoglobin Concentration (MCHC) were increased in the non-vegetarian group compared to the vegetarians. There were no significant differences in Total Red Blood Cell (RBC count), Red cell Distribution Width (RDW), Platelet count, and Erythrocyte sedimentation Rate between non-vegetarians and vegetarians. Lipid profile showed no significant differences in serum total cholesterol, triglycerides, LDL (Low-density lipoprotein) between non-vegetarians and vegetarians except for a relative increase in these parameters in non-vegetarians. The vegetarians had a relative increase in HDL (High-density lipoprotein). No significant disparity was found with BMI values in both non-vegetarians and vegetarians. Conclusion: This study shows that a vegetarian diet tends to develop nutrient deficiency anemia, especially iron deficiency anemia whereas a non-vegetarian diet has an increased risk of developing dyslipidemias. It is recommended that a non-vegetarian diet having an adequate daily intake of fruits, vegetables, and polyunsaturated fatty acids, with lifestyle modifications would prevent the development of deficiency anemias and dyslipidemias. © 2024 IEEE.

This study elaborates on the use of the ‘Random Forest’ algorithm, which is a supervised machine learning algorithm that can be deployed in regression as well as classification problems and has been deployed for prediction of loan approval. The study utilizes the data of Indian customers who want to avail loans for educational purposes from the Indian banks, which are the main loan lending agencies in India. The dataset has been curated from Indian customers who have applied for education loans from the Indian banks. Features of the dataset have been selected by reviewing the application forms of various Indian banks. From the standpoint of Indian banks, the current study shows the potential of machine learning algorithms such as the decision tree classification algorithm, with a particular emphasis on the random forest method, in predicting education loan eligibility with a high accuracy of 90.6%. The model has been evaluated using the following most pertinent metrics in the classification problem: accuracy, recall, F1 score, and precision. Significantly, our approach eliminates the need for credit scores, making it an ideal solution for education loan applicants who typically have limited or no credit history. To the best of our knowledge, there is a significant gap in the literature because no study has particularly examined the prediction of education loan approval in the context of Indian banks. Our research fills this critical void, providing valuable insights for stakeholders in the Indian education loan sector. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

This study examines and assesses the features and user experiences of popular smartphone applications for food delivery, including Uber Eats, Zomato, Swiggy, and EatSure. Through assessments were carried out regarding parameters including order processing speed, delivery dependability, user interface design, and general satisfaction. Data collection techniques included user surveys, app analytics, and in-person interactions with customer service representatives. The importance of precise and rapid order delivery, user- friendliness, the effectiveness of promotional offers, pricing strategies, and shipping costs are all heavily emphasized in the study. The results showed that the apps differed and that certain platforms outperformed others in specific domains. The research paper provides useful insights for users and service providers equally, highlighting the necessity of continual evaluation and innovation to meet evolving consumer needs in the food delivery industry. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

The purpose of this research is to upgrade accessibility and cross-language information retrieval by fashioning a Multilingual Text Recognition and Interpretation System. This system seeks to control communication gaps and create an alternative inclusive digital environment in response to the growing need for digital tools that can exercise and understand text in different languages. The system recognizes text from diversified sources, including digital files, handwritten notes, and printed documents, using sophisticated machine learning models including Transformer-based models and Convolutional Neural Networks (CNNs). Users may obtain and penetrate information in languages they may not be familiar with thanks to the system's integration of Optical Character Recognition (OCR) and Neural Machine Translation (NMT), which transforms identified text into the appropriate language. This research develops a reliable system that can work with various scripts and languages and provide high accuracy in real-Time performance. Common issues include managing low-resource languages, enhancing the ability to concede intricate scripts, and maximizing performance for real-Time applications, which will all be addressed by the solution. © 2025 IEEE.

In this paper, we presented a crop recommendation system for precision agriculture that is based on the machine learning models providing crop recommendations based on the environmental and soil context. Introducing Explainable AI (XAI) in the designed system using LIME, we hope to enhance the level of trust to its recommendations through the understanding of the process which goes into each of them by the farmers. This has been made possible by the system's architecture that consists of backend in Flask and frontend in React. The system was subjected to a rigorous assessment of its reliability and efficiency once developed to obviate measurement errors and ensure its effectiveness and efficiency. Moreover, the feedback on the integration of XAI also reveals an enhancement of interpretability by demonstrating the explanations of such features as pH of the soil, temperature, and moisture of the soil to have a positive effect to the system. Besides improving the chances of accurate crop yields estimation, this approach enables an accurate assessment of farming inputs with regards to sustainability. © 2025 IEEE.

Current research combines multiple machine learning models to provide comprehensive insights into tailor-made medicine for lung cancer. The study identified micropatterns of malignant disorders using a dataset of 3222 CT and MRI scans using VGG16 architecture, K-Nearest Neighbors (KNN), Support Vector Machines (SVM), and Convolutional Neural Networks (CNN). A comprehensive understanding of the characteristics of pulmonary arteries is obtained. After extensive evaluation, the models exhibit exceptional accuracy and precision, with VGG16 performing particularly well. Confusion matrices provide comprehensive insight into model predictions. This work provides important new information in the field of personalized precision medicine through the potential of standardized methods for lung cancer diagnosis and treatment. Future research should focus on sample design they will be enhanced, and their application will be extended to multiple datasets. © 2024 IEEE.

This study is based on the creative design and production of educational toy prototypes utilizing wood waste. It mainly targeted the aspects of sustainability, safety, and durability. A qualitative research approach, based on various literature reviews learning eco-friendly materials, has been used to emphasize the potential of utilizing wood wastes in the development of new, innovative toy prototypes for pre-school children. These prototypes reduce environmental impacts caused by waste wood as it is aligned with the principles of green design, promoting the development of eco-friendly educational toys. The results have been designed to provide the basis for subsequent research on the potential and flexibility for the manufacturing of wooden toys through waste materials. Some types of wood can be used based on their inherent physical and chemical properties, ideal for the elaboration of very durable, safe, and engaging toys. The innovation of materials for this end has an underpinning with the evolution of needs in educational toys toward sustainable, eco-friendly, and enriching learning for children. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

Empathy or the ability to interpret and communicate another individual’s emotions and condition could be considered the cornerstone of the healthcare professionals’ relationship with patients. It is important for healthcare professionals to practice empathy while connecting with patients to ensure their well-being and motivation. Nurses especially need to interact with patients and their families more frequently, and therefore need to constantly demonstrate empathetic behavior and compassion. While the Indian Nursing curriculum includes training in communication skills, there is a need to practice these behaviors in real contexts and be able toobserve measurable improvements. There are also several factors such as upbringing, the education system, shortage of manpower and resources, and a skewed nurse to patient ratio which lead to higher stress factors and lack of empathy in Indian Nurses. There is a need for training and mentorship for nurses and an environment that helps them practice empathetic behavior, without impacting their daily work or the wellbeing of patients. If this training is available online or technology enabled, it can also free up time for senior nurses who may otherwise be providing mentorship to younger nurses. Virtual Reality (VR) provides the opportunity of creating an environment where nurses can experience a real hospital setting, with realistic scenarios and interactions with patients built in to practice empathetic behavior. This study, based on a review of existing literature and conversations with nurses, explores the role of a nurse and identifies areas where empathetic behavior can be practiced. It also researches existing segments where VR has been successfully implemented in the healthcare context. It arrives at a framework for creating VR-based training for Indian Nurses on empathy. This framework can also be extended to building other VR-based soft skill training for healthcare professionals. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

The high prevalence of mental illnesses in adolescents makes it crucial to design innovative ways to identify and intervene at an early stage. This paper presents a new framework based on Bidirectional Encoder Representations from Transformers to detect potential mental health risks of teenagers from their online social network activities. The framework captures the subtle linguistic nuances used by teenagers experiencing mental health struggles. This is achieved by fine-tuning a BERT-based sequence classification model on a large dataset of social media posts, categorized across multiple mental health conditions, with the intent of proactive identification of at-risk teenagers. It presents an evaluation of this model, developed using a stratified k-fold cross-validation approach, scoring an average accuracy of 81.42%, precision of 81.98%, recall of 81.49%, and F1-score of 81.42%. The research is thus important for the field of artificial intelligence in mental health screening and points toward strong potential for natural language processing to help better refine early intervention strategies for adolescent mental health. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2026.

This research endeavors to revolutionize the precision of photovoltaic (PV) power generation predictions through the introduction of a novel Hybrid Ensembled Model. Motivated by the imperative to enhance the reliability of renewable energy technologies, the study amalgamates the strengths of fuzzy logic, support vector regression, ensemble learning, and evolutionary computation. The model's design addresses the inherent challenges of PV power generation prediction, notably uncertainties and non-linear dynamics. The investigation unfolds in distinct phases, beginning with a meticulous exploration of the significance of precise PV power generation prediction. Acknowledging the limitations of existing methodologies, the research establishes a robust foundation for the development of the Hybrid Ensembled Model. Methodologically, the incorporation of Double-Input-Fuzzy-Modules (DIFM), Extreme Learning Machine (ELM), and evolutionary computation ensures a comprehensive and adaptive framework. This research propels the discourse on renewable energy prediction models, presenting a Hybrid Ensembled Model as a pioneering solution for precise and reliable PV power generation forecasts. The implications extend beyond academia, envisioning a future where sustainable energy systems are underpinned by accurate predictions, fostering optimal resource utilization and contributing to a resilient and sustainable energy landscape. © 2024 IEEE.

Video game addiction is a rapidly growing concern among young people, emerging as a major issue due to its harmful impact on their mental health. This study investigates how predictive models might reveal subtle patterns in gaming habits by examining the relationship between video games, youth development, and machine learning. The research will investigate social dynamics as well as cognitive, academic, and mental health implications in order to educate individuals about responsible gaming habits and teaching tactics. The study covers ethical considerations and long-term societal ramifications, emphasizing the significance of a compatible way to navigating the IT based world because of young people's prosperity. Its purpose is to evaluate whether a gamer is career-minded. That's why we gathered 804 data points from them, of which 780 were utilized. The key point that we utilized were name, gender, age, amount of time spent studying, amount of sleep or wakefulness, amount of time spent playing games, reason for playing games, amount of time spent playing games more than with family, time spent honing skills, and time spent worried about a career. and after that, they underwent reprocessing and were checked before being used with certain machine learning algorithms. Various prediction and find ways use machine learning, artificial intelligence, and deep learning method. Our approaches include Gaussian Naive Bayes (GNB), Random Forest (RF), Adaptive Boosting (ADA Boosting), Decision Tree (DT), Extra Trees (ET), Gradient Boosting (GB) classifier, KNN (K-Nearest Neighbors), Support Vector Machine (SVM). The Random Forest classifier performed better in our analysis than the other eight methods. While all eight classifiers have great prediction accuracy, Random Forest Classifier (RF) has the highest accuracy of 97.00%. This study analyses the various behaviours or habits of the current youth society. Based on the results obtained from that analysis, a machine learning-based model has been proposed. This proposed model will help in understanding the behavior of young people and their current situation. © 2025 Bharati Vidyapeeth, New Delhi.

Mango fruits are widely recognized for their diverse applications in the food, beverage, and various industries, making them a global commodity of significance. the escalating demand for premium-quality mangoes highlights the necessity for dependable and efficient grading techniques. Traditional mango grading heavily relies on human expertise, often susceptible to subjectivity, errors, and time-intensive procedures. In this study, we introduce a CNN-based mango grading model to automate and enhance the accuracy ofmango quality assessment. Our proposed model employs CNNs in a modular, sequential approach to extract relevant features from mango images and classify them into distinct quality grades. We utilize a comprehensive dataset of mango images with corresponding quality labels for training and validation. Experimental outcomes underscore the effectiveness of the CNN-based method in accurately grading mangoes, presenting a promising solution to improve efficiency and objectivity in mango grading. In conclusion, the CNN-based mango grading model represents a significant step forward in modernizing and optimizing mango quality assessment processes. the combination of machine learning techniques with image analysis offers a robust solution to address longstanding challenges associated with traditional grading methods. As this technology evolves, it has the potential to revolutionize quality control practices in the food industry. This CNN-based approach outperforms other classifiers for this specific mango grading problem. © 2024 IEEE.

Precision agriculture has become an essential practice for optimizing crop yield and resource utilization. This research integrates classification and regression models to enhance crop prediction and yield estimation. The study employs a Random Forest Classifier for crop recommendation and utilizes regression models such as Linear Regression, Decision Tree Regressor, and K-Neighbors Regressor for yield estimation. Feature Importance Analysis is also applied to identify key factors influencing crop growth. The proposed hybrid approach aims to improve prediction accuracy, scalability, and practical applicability, providing a comprehensive decision-support system for farmers. © 2025 IEEE.

This paper describes a full pledged website and mobile app to provide a solution for people who are searching for medical treatment location where they are unaware about its location. With the help of Machine learning algorithms, medical diagnosis will be done and based on the reports of medical diagnosis with the tracking facility of GPS technology, our suggested system will give recommend a nearby hospitals based on user symptoms. This suggested hospital will give the best care for their medical condition; in this paper, we have features with option based on symptom-based prediction, location-based prediction and real-time prediction to book ambulances within the application and also from the selected hospital and a real-time distance will be displayed to the suggested hospital from the user location. This project provides a user-friendly interface which includes the Random forest Machine Learning prediction for using the services which is designed and implemented which gives the precise location tracking for smooth communication and safer travel. Software require some of cover tools for developing mobile app, such as Android and React Native, while hardware requirements of GPS modules, GSM modems, and microcontrollers being used with IoT facility. This project aim to increase patient benefiting features by improving availability to access the healthcare services for people who are in unfavorable conditions being helped using location-based services and also with mobile technologies. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2026.

In the banking or financial sector, the major problem is credit card fraud and the financial institutions find it challenging. This fundamental study will examine how machine learning combats this issue and suggest possible ways to detect and avoid credit card fraud detection. This article highlights the methods and resources for identifying fraudulent credit card purchases. The study examines the prevalence of credit card fraud and the reasons it must be quickly detected, preparing the data is a crucial step in identifying credit card fraud by examining the various datasets to identify fraudulent activities. The study also describes the importance of cleaning, modifying, and creating characteristics for various kinds of data related to credit card fraud detection. Highlighting the importance of having a different proportion of actual and fraudulent transactions is also emphasized. © 2025 Author(s)

The rapid growth of Internet of Things (IoT) devices increases their vulnerability to attacks, necessitating stronger security measures. The machine learning technique is one of the most effective ways to detect anomalies and malicious activity in IoT contexts. The use of machine learning for Internet of Things attack detection is reviewed in this paper. We have looked into the use of numerous popular machine learning techniques in this field, including Deep Neural Networks, Random Forests, and Support Vector Machines. The goal of this paper is to provide important insights to researchers and practitioners who want to use machine learning to fortify IoT systems against cyberattacks. We discuss the various ways attackers can attack, examine a summary of potential risks, and discuss the likely Machine Learning solutions. This paper provides a review of working in the field of IoT security with important insights by highlighting the significant potential of machine learning techniques in IoT attack detection. © 2025 IEEE.

Managing student attendance by hand or through ERP can be a big task for teachers and faculty members. The smart and automatic attendance system is easier to make the process in an efficient manner. During the analysis process, feedback from the user is done. The proposed system helps to stop issues like student marking attendance for friends(proxies) who aren't actually in the class. It's always live through video feed and can be used for attendance. OpenCV captures the video frames, and the system also uses dlib to detect and recognize faces. It recognizes faces that are matched with a database of student photos to mark attendance. The proposed model gives an effective Automated tool that tracks the student time presence. © 2024 IEEE.

The identification of age and gender from facial images is also difficult because of the variations in lighting condition, different facial poses, the presence of occlusions, as well as the problem of imbalanced gender data set. Consequences of inaccurate prediction include the following: This could be observed in surveillance systems resulting in the failure of offering a productive and effective individual marketing the result of these inaccuracies could also be seen in social and demographic studies where unsuitable results could be generated. To overcome these difficulties, this paper utilizes a novel CNN architecture that is designed specifically for the task. The CNN model design includes convolutional and max-pooling layers and fully connected layers with dropout to reduce the overfitting problem. This means that the accuracy and reliability of age and gender prediction is improved by the use of this approach, making it suitable for security services, marketing realms and human-computer interface systems. There exists a reasonable improvement in the model when benchmark datasets are applied; thus, the custom CNN model is a reliable tool for practical tasks. © 2024 IEEE.

Indian equity has been one of the most vibrant and continuously changing financial markets and has witnessed bull and bear phases in the last two decades. This paper aims to focus on the following objectives: identifying and exploring the complex dynamics that underpin these cycles and their consequences for market participants, Indian policymakers, and the economy more generally. Using quantitative and qualitative methodologies, this research examines bull and bear spans within the Indian equity market across the 2003 to 2022 period. The research data is collected through a secondary research approach that incorporates historical and numerical data analysis, and through experts' opinions that give a quantitative and qualitative data trend view on the market, The methodologies include time-series analysis, correlation analysis, and event analysis to determine the market phase and its causes. Here are some main conclusions where quite clearly it is shown that the Indian stock market possesses a high degree of stability and development opportunities, even though it experiences negative impacts from the world crisis getting back on its feet to reach a new nominee. Moreover, the strong relationship between macroeconomic variables and market returns is confirmed, supporting good economic management. The study points to the growing globalization of Indian markets, the changing pattern of market entry, and the emerging role of individual investors. The findings indicate that establishing strong and effective rules and regulations, and a market environment for managing cycles is essential. Although various cycles remain discernible, it should be noted that the general tendency in the Indian equity market has been considered to be rather upward, and the Indian market has been gradually maturing. The paper reiterates the market's significance in India's economic goals before outlining research directions for future work, such as technological development on the market or sustainability in investing. © 2025 IEEE.