Browse Items (11858 total)
Sort by:
-
Investigating Key Contributors to Hospital Appointment No-Shows Using Explainable AI
The healthcare sector has suffered from wastage of resources and poor service delivery due to the significant impact of appointment no-shows. To address this issue, this paper uses explainable artificial intelligence (XAI) to identify major predictors of no-show behaviours among patients. Six machine learning models were developed and evaluated on this task using Area Under the Precision-Recall Curve (AUC-PR) and F1-score as metrics. Our experiment demonstrates that Support Vector Classifier and Multilayer Perceptron perform the best, with both scoring the same AUC-PR of 0.56, but different F1-scores of 0.91 and 0.92, respectively. We analysed the interpretability of the models using Local Interpretable Model-agnostic Explanation (LIME) and SHapley Additive exPlanations (SHAP). The outcome of the analyses demonstrates that predictors such as the patients' history of missed appointments, the waiting time from scheduling time to the appointments, patients' age, and existing medical conditions such as diabetes and hypertension are essential flags for no-show behaviours. Following the insights gained from the analyses, this paper recommends interventions for addressing the issue of medical appointment no-shows. 2024 IEEE. -
Enhancing Stroke Prediction: Leveraging Ensemble Learning for Improved Healthcare
Stroke, a potentially deadly medical disorder, requires excellent prediction and prevention measures to minimize its impact on individuals and healthcare systems. In this study, ensemble learning techniques are employed to enhance the accuracy of stroke prediction. The method combines four different machine learning algorithms, Adaboost, CatBoost, XGBoost, and LightGBM, to produce a strong predictive model. The data was composed of a rich set of demographic, medical, and lifestyle information. The data was preprocessed and features were engineered to maximize predictive performance. Results showed that the stacked ensemble model, which is composed of Adaboost, CatBoost, XGB, LightGBM, and Logistic Regression, meta-model, outperformed other models. The model has the potential to be used as a decision support tool in an early stroke risk assessment system, enhancing clinician decision-making and improving healthcare outcomes. 2024 IEEE. -
Smart Vehicle Recognition System on Indian Roads Under Rainy Conditions
Recognition of vehicles under the different weather condition is very challenging. This work aims to recognize vehicles on Indian road in accordance with their visibility. It is important to recognize the surround roadside objects, particularly front and rare vehicles to avoid the accidents. Especially in raining conditions vehicle recognition is rate traffic surveillance cameras get decreases due to water droplets. Hence, we proposed a method for recognition of vehicles on road in rainy condition using image processing in computer vision techniques to improve the recognition rate. In the proposed method, an instance segmentation technique is used to segment the vehicles in Indian road scene and the visual noise and texture features are analysed and computed in the segmented images to recognize the vehicles more accurately in rainy conditions. By integrating the visual noise features with the texture feature and instance segmentation, the accuracy of vehicle recognition is improved. The experimental findings demonstrated that the suggested approach could more accurately predict the visibility of vehicles in rainy weather conditions. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
Machine Learning-Based Imputation Techniques Analysis and Study
Missing values are a significant problem in data analysis and machine learning applications. This study looks at the efficacy of machine learning (ML) - based imputation strategies for dealing with missing data. K-nearest Neighbours (KNN), Random Forest, Support Vector Machines (SVM), and Median/Mean Imputation were among the techniques explored. To address the issue of missing data, the study employs k-nearest neighbors, Random Forests, and SVM algorithms. The dataset's imbalance is considered, and the mean F1 score is employed as an evaluation criterion, using cross-validation to ensure consistent results. The study aims to identify the most effective imputation strategy within ML models, offering crucial insights about their adaptability across various scenarios. The study aims to determine the best plan for data preprocessing in machine learning by comparing approaches. Finally, the findings help to improve our knowledge and application of imputation techniques in real-world data analysis and machine learning. 2024 IEEE. -
Contextual Recommendation System: A Revolutionary Approach Using Hadoop, Spark, NLP and LLMs
This study presents a novel framework for contextual recommendations on platforms like Wikipedia, integrating Hadoop, Spark, NLP, and LLMs. Leveraging these technologies, the framework aims to enhance user experiences by delivering personalized article suggestions aligned with their current interests. Through scalable data processing, advanced NLP techniques, and LLM-powered semantic understanding, the framework offers a transformative approach to recommendation systems, promising to revolutionize knowledge exploration on digital platforms. 2024 IEEE. -
A Heuristic Approach to Resolve Priority-Driven Unbalanced Transportation Problem (PUTP)
This research addresses the priority-driven unbalanced transportation Problem (PUTP), characterized by a situation where the overall demand surpasses the available supply. We propose the Max-flow Min-cost Priority-driven Unbalanced Transportation Problem (MMPUTP) as a heuristic approach to handle this issue effectively. The strategy of MMPUTP focuses on optimizing resource allocation and reducing costs, making it highly effective in fulfilling high priority needs in a cost-efficient manner. Through a comparison with Vogel's Approximation Method (VAM) over different sets of problems ranging in size from 5?5 to 50?50, the effectiveness of the MMPUTP algorithm is evident. The findings underscore the significance of choosing the right algorithm based on the size and complexity of the problem set in the context of the Priority-driven Unbalanced Transportation Problem, with MMPUTP proving to be a flexible and reliable option in various situations. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
Smart Certificates Using Blockchain: A Review
When making job offers, it is usual practice for businesses to check applicants academic credentials. The organization that issued the certificate must authenticate it for the employer to ensure that it is genuine. Because of the lengthy process involved in certificate verification, the selection process takes longer overall while establishing the legitimacy of a certificate. To address this issue, blockchain provides a verified distributed ledger along with a cryptography mechanism to thwart academic credential counterfeiting. The Blockchain also provides a standardized platform for document access, storage, and verification that takes the least amount of time. The review of the methodologies and performance of the same has been covered in the paper. The Author(s), under exclusive license to Springer Nature Switzerland AG 2024. -
Analyzing the Consumer Buying Behavior by Adapting Artificial Intelligence (AI)
In any business consumers or customers are important part of the market, so it is necessary to attract more customers for increasing the profits. The current research in this area has demonstrated that artificial intelligence (AI) has a substantial impact on the end customer, contrary to the widespread notion that it has more of an impact on industry than other manufacturers. There are many studies on the various applications of AI in analyzing and visualizing the consumer behavior. Thus, it is been observed the behavior of consumer is not same for same businesses, it varies from consumer to consumer. In other respects, AI is changing how consumers act right now. In coming year's use of AI will become common where the human dependable businesses also get automated with time. 2024 IEEE. -
Application of AI in Determining the Strategies for the Startups
Startups face unique challenges in developing viable strategies to create a competitive advantage and achieve long-term success in today's rapid and global business climate. Using tools powered by AI and algorithms, startups may harness massive amounts of data, generate relevant insights, and make intelligent choices across a variety of business processes. The research begins with an examination of the fundamental concepts beneath AI and how they are used in the business sector. It illustrates how organizations may simplify and improve important activities such as market research, customer segmentation, and trend analysis using artificial intelligence (AI), natural language understanding (NLU), as well as predictive analytics. The report also delves into extensive case studies and real-world examples of businesses that have effectively integrated AI into their business decision-making processes. These examples highlight the practical benefits of AI-driven insights that include enhanced resource allocation, customer targeting, as well as operational performance. It highlights the importance of ethical AI methodologies, transparency, and safeguards to ensure unbiased and fair decision-making. Finally, this study demonstrates how AI has the potential to profoundly transform how entrepreneurs design and implement their strategies. By leveraging AI-driven perspectives, startups may handle complex market dynamics with more precision and agility, increasing their chances of enduring and succeeding in a competitive business climate. The study's findings provide a road map for organizations wishing to apply AI in strategic decision-making processes. 2024 IEEE. -
TumorInsight: GAN-Augmented Deep Learning for Precise Brain Tumor Detection
In addition to the shortage in data as well as the low quality of MRI images, one of the most difficult tasks in contemporary medical imaging is the diagnosis of tumors in brain. This work presents a new approach to enhance diagnostic accuracy using sophisticated preprocessing techniques. Combining BRATS 2023 and Cheng et al. datasets to apply cutting-edge deep learning preprocessing methods with Generative Adversarial Networks (GANs), specifically DCGAN, Contrast Limited Adaptive Histogram Equalization (CLAHE), and gamma correction, it aims to significantly improve the quality of MRI images. As a result, updated data should be generated with greater precision and detail, making it possible to identify tumor-affected areas with greater accuracy. Thorough assessment, demonstrated by metrics such as Accuracy (0.98), Specificity (0.99), Sensitivity (0.99), AUC (0.65), Dice Coefficient (0.67), and Precision (0.71), highlights possible advancements in brain tumor identification and treatment, thereby highlighting the effectiveness of the suggested approach. 2024 IEEE. -
The Design of Driver Fatigue Detection Based on Eye Blinking and Mouth Yawing
In modern era, the Intelligent Transportation System (ITS) is very essential for the betterment of transport management, autonomous vehicles and especially for safe driving. The statistics suggest that the major severe accidents occur because of drivers drowsiness. The main objective of this work is to give the alert alarm when the driver is falling asleep. In the proposed study, the driver's face is detected using the Viola Jones algorithm, and a novel approach to detecting eye blinks using template matching and a similarity measure. For effective eye tracking, the normalized correlation coefficient is calculated. The correlation score is used to identify eye blinks since a blink causes a significant change in the correlation score. In tracking of mouth yawing finding the darkest region between the lips. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
Privacy-preserving federated learning in healthcare: Fundamentals, state of the art and prospective research directions
Recent collaborations in medical diagnostic systems are based on data private collaborative learning using Federated Learning (FL). In this approach, multiple organizations train a machine-learning model at the same time eventually leading to global model generation. This paper reviews the fundamentals of FL and its evolution path in Healthcare. The objective of this review is to scope a wide variety of healthcare applications in FL. Exactly what research direction is moving in interesting for research communities to guide their future course. This review uniquely focuses on examining numerous FL-based healthcare implementations, detailing their core methodologies and performance metrics, which, to our knowledge, have not been previously available. Privacy-preserving collaborative distributed learning through federated learning in healthcare enhances research collaborations, thereby resulting in better-performing models. This comprehensive review will act as a valuable reference for researchers exploring new FL applications in the healthcare domain. 2024 IEEE. -
Enhancing Cybersecurity: Machine Learning Techniques for Phishing URL Detection
Phishing attacks exploit user vulnerabilities in cybersecurity awareness by tricking them to fake websites designed to steal confidential data. This study proposes a method for detecting phishing URLs using machine learning. The proposed method analyzes various URL characteristics, such as length, subdomain levels, and the presence of suspicious patterns, which are key indicators of phishing attempts. Gradient Boosting was selected due to its robustness in handling complex, non-linear relationships between features, making it particularly effective in distinguishing between legitimate and phishing URLs, by evaluating the Gradient Boosting classifier on a dataset with 10,000 entries and 50 features, the method achieves an accuracy of 99%.This approach has the potential to enhance web browsers with add-ons or middleware that alert users from potential phishing sites which will be based solely on URL. 2024 IEEE. -
Insights of Evolving Methods Towards Screening of AI-Enhanced Malware in IoT Environment
Internet-of-Things (IoT) has been encountering a series of potential form of threats since past half decades. Artificial Intelligence (AI), which is frequently seen to be adopted to solve various challenges in IoT operation, has now been adopted even by attackers for their malicious purposes. Of all forms of threats, AI-enhanced malwares are one of the most potential forms of threats which has its extensive effectiveness towards the complete operation of the entire IoT environment. Hence, this manuscript discusses existing detection and prevention approaches evolved in current literatures to understand various taxonomies of solution-based methodologies for circumventing such threats. The paper also contributes towards highlighting the potential open-ended issues that are yet to be addressed. The Author(s), under exclusive license to Springer Nature Switzerland AG 2024. -
Enhancing Early Detection of Cardiovascular Disease through Feature Optimization Methods
cardiovascular diseases are the most common reason for mortality around the world. Early detection of the ailment can help to reduce the mortality rate considerably. The ever-growing technologies like machine learning algorithms and deep learning models can be used for this purpose. The AI models thus developed can be used for health sector for assisting doctors in assessing the stage of the disease and detection and tracking of the clots in the cardio blood vessels. The proposed work uses two benchmark datasets for analysing the performance of various machine learning algorithms including KNN, Nae Bayes, Decision Tree and Random Forest. The performance was compares based on the AUC %. The method feature reduction were used here to reduce the computational complexity of the model. The results show that Random Forest Algorithm gave the best result when compared to other algorithms in case of UCI dataset and MLP classifier gave best results for Kaggle dataset. 2024 IEEE. -
Impact of Artificial Intelligence on the Social Media Marketing Strategies
The use of social media is increasing as the use of smartphones is increasing, various applications on smartphones are now becoming good platform for market. As the use of smartphones is increasing the use of different artificial intelligence (AI) technologies making the phones smarter. The social media is now one of the most globally crowded platforms with millions of users. Most of the businesses are now turning their marketing strategies by highlighting the digital marketing from various platforms. Thus, the focus of study is to find the increasing impact of AI on the social media related marketing strategies. Different studies highlight the different impacts of social media and marketing using different AI tools and platforms which makes customer to find the best product as per their choice. So, social media marketing has become simpler and more adaptable thanks to the development of artificial intelligence. 2024 IEEE. -
Experimental Investigation of Nano Hexagonal Boron Nitride Reinforcement in Aluminum Alloys Through Casting Method
Aluminum metal matrix composites (AlMMCs) have a significant impact on a variety of industries that seek for innovation, efficiency, and sustainability. AlMMCs are substantial because of the special combination of properties that make them an essential part of contemporary production and design. Custom made properties of the AlMMCs can be obtained by the reinforcing different ceramic particles. Among the reinforcements, nano hexagonal boron nitride were rarely used. Hexagonal boron nitride particles have self-lubrication properties and it is one of the promising substitutes of graphite. The incorporation of hexagonal boron nitride (hBN) as a reinforcement material in aluminum alloys has garnered significant attention in recent years. This paper provides an overview of the reinforcement of nano hBN in aluminum alloys through casting method and highlights the mechanical and thermal properties of these alloys. The results show that the wear rate of the composite at 2wt.% is 9.91% lower for a load of 40 N when compared to unreinforced composite. Furthermore, the impact of hBN content, dispersion, and processing parameters on the properties of the composites is analyzed. The unique structural and thermal properties of hBN, along with excellent lubricating abilities, make it a promising candidate for reinforcing aluminum composites. The Author(s), under exclusive license to Springer Nature Switzerland AG 2024. -
Digital Forensics Chain of Custody Using Blockchain
In todays world, Digital Forensics is a crucial subject with much scope as data storage becomes more decentralised. The collection and preservation of digital media is a topic of concern across the Cyber Security and Digital Forensics field. With Cloud Infrastructure and other technologies, data is not permanently stored in one place and gathering and analysing it can become a headache for Forensic Investigators. Blockchain, however, works as a decentralised, distributed peer-to-peer network and thus can be considered a suitable solution for the mentioned problems. With the help of a blockchain network and Smart Contracts, Digital Forensics can be significantly improved to adapt to modern digital architecture. The Author(s), under exclusive license to Springer Nature Switzerland AG 2024. -
Design and Stress Analysis of the Frame for an Electric Bike
Global emissions have been on the rise since the industrial era because of the increased energy-intensive human activities, which is a direct cause of global warming and climate change. Of the total emissions, around 17% is from the transportation sector, which significantly contributes to the emissions. One of the easiest ways to be more sustainable is to choose electric vehicles instead of Internal combustion engines. Almost 75% of the vehicles registered in India are two-wheelers, but there are no affordable and reliable electric two-wheelers. This research works to optimize and analyze the design of a step-through frame design for an electric bicycle. The frame design is analyzed by providing boundary and loading conditions with two different materials (Steel-AISI4130 and Aluminum AL6061). The numerical analysis is carried out using ANSYS APDL. The result of von Mises stress is 166MPa and 160.4MPa for steel and aluminum, respectively. The result of stress and displacement is within the acceptable limit. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
An Alternative Deep Learning Approach for Early Diagnosis of Malaria
Considering the malaria disease-related moralities prevailing mainly in underdeveloped countries, early detection and treatment of malaria must be an essential strategy for lowering morbidity and fatality rates. Detection of Malaria using traditional investigation methods through blood samples and expert judgments was found to be time-consuming. In this paper, the authors introduced a Machine Learning automated system to eliminate the need for human intervention, which in turn enables early detection of malaria. The study has used various Deep Learning techniques such as traditional Convolutional Neural Network (CNN), VGG19, ConvNeXtXLarge, ConvNeXtBase, ConvNeXtSmall, ConvNeXtTiny, InceptionResnetv2, Xception, DenseNet169, EfficientNetB7, MobileNet, ResNet50, and NasNetLarge as base models. These models have been trained and tested with microscopic blood smear images dataset and observed that ConvNeXtXLarge detects malarial parasites with an accuracy of 96%. The proposed method outperforms the existing approaches in terms of both accuracy and speed. The findings of this work can contribute to the development of more accurate and efficient automated systems for early detection of Malaria. 2024 IEEE.