Browse Items (3095 total)
Sort by:
-
Optimizing Healthcare: Enhancing Disease Management with Recommendation Systems
This paper explores a data-driven disease recommendation system for medical professionals based on symptoms. The technology examines symptom patterns to recommend diseases from large datasets by utilizing collaborative filtering and data analytics. To provide individualized disease recommendations based on symptom severity, it goes through data preprocessing and uses techniques like collaborative filtering and cosine similarity. Even if the technology is promising, disease predictions might be strengthened. It seeks to support early disease prediction and offer patients and healthcare professionals individualized guidance. This system demonstrates the potential of technology in healthcare decision-making using a basic Tkinter application. More improvements are anticipated as a result of data-driven approach advancements, which will improve patient care and optimize healthcare procedures. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
Data-Driven Strategies for Twitter Engagement: Hashtag Recommendations and API Insights
Twitter is a great way to connect with people worldwide, and one of the best ways to do that is by using hashtags. A hashtag is a keyword or phrase attached to a particular topic, and users can use it to find related tweets. Using a hashtag relevant to the needs or for business can increase the tweets visibility and make it easier for people to see the content they want. It can hugely help content creators who want to increase engagement and influence their tweets. This research introduces TagMate, a hashtag recommender system for Twitter that offers significant benefits. By accessing the tweets using Twitter API and after analyzing and performing algorithms, recommendations for hashtags can be obtained. The Twitter API allows access to various information about the account, including followers, tweets, content, etc. This information can be used to generate recommendations for hashtags related to the business. The system will generate hashtags according to the tweet and recommend trending or popular hashtags to increase their reach or engagement on the Twitter platform. Using the API, a dashboard can be created showing which hashtags are being used most frequently and which are most popular. This information can help create more relevant and engaging tweets, attracting more followers and interest. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
A Comprehensive Analysis of the Impact of Demographic, Socioeconomic Factors, and Social Media Advertisements on Consumers Online Shopping Behavior
This research comprehensively explores the intricate relationship between social media, demographic factors, socioeconomic status, and online shopping behavior. It investigates how various elements, such as brand, quality, price, and reviews, intersect to influence consumers virtual shopping experiences. Recognizing the diversity among consumer populations, the study considers demographic traits like age, gender, and socioeconomic status. It identifies a research gap concerning the moderating effects of these factors on virtual shopping behavior and the influence of different advertisement sources. The research proposes a methodology involving machine learning approaches to address these gaps. Using a dataset collected from 331 respondents in Kolkata, the study employs statistical tests and machine learning models to analyze the correlations and predict online shopping behavior. The results underscore the importance of understanding consumer behavior in the evolving e-commerce landscape. Logistic Regression emerges as the most accurate algorithm for predicting online shopping decisions. Additionally, the research investigates the reasons limiting consumers use of online shopping platforms, providing practical insights for tailored marketing strategies. These findings empower businesses to adapt their marketing strategies to the evolving landscape of consumer behavior. Overall, the study enhances our understanding of the multifaceted variables shaping customers decisions in online marketplaces. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
Harnessing Insights for Optimizing Healthcare: Disease Prediction and Beyond
This study offers a novel method for developing classification approaches for disease prediction. Exploratory analysis and meticulous data preprocessing were conducted to understand the relationships between symptoms and illnesses. The research involved assessing various machine learning models, including the random forest classifier, through cross-validation techniques, resulting in the identification of a high-performing model with an impressive accuracy rate. In addition, this study incorporates data visualization techniques to gain insights into symptomdisease connections. The studys focus on data visualization and optimization strategies in health demonstrates the potential to transform healthcare by providing precise diagnoses and predicting diseases, ultimately improving patient outcomes. This research underscores the efficacy of data-driven techniques and their integration into recommendation and disease prediction systems, emphasizing the significance of data visualization and optimization strategies in health. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
Unveiling Patterns, Visualizations, and Trends from Patient Diabetes Data
The important role that exploratory data analysis, or EDA, plays in the context of diabetes prediction is explored in this work. EDA is used as a key component of a multimodal strategy to identify unique characteristics linked to diabetes. EDA offers insights that aid in the creation of prediction models by sifting through the complex patterns present in the medical data. The focus is on using EDA to fully grasp the data landscape while also comprehending the distinct features of diabetes. This investigation is critical to accurately categorizing people into discrete risk groups and emphasizes the use of domain-specific knowledge in enhancing diabetes prediction techniques. The research suggests using specific EDA techniques to gain deep insights and lead proactive responses. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
Advancing Software Defect Detection and Prevention: Bridging Gaps in Early-Stage and Evolving Software Systems
Software defect prediction (SDP) is a critical method in modern software development, saving costs while ensuring the delivery of high-quality software systems. This study investigates the vital importance of SDP, focusing on its function in detecting and correcting software faults that might lead to system failures. SDP forecasts defect-proneness and optimizes software-testing processes by using software metrics such as lines of code and change information. The chapter examines the progress of SDP research since the turn of the century, emphasizing the academic emphasis on refining static characteristics and establishing efficient learning methods for building high-performance defect predictors. Recognizing the economic consequences of software flaws, particularly in major engineering projects, this chapter emphasizes the importance of SDP in limiting project failures and economic losses in the twenty-first century. Several defect prediction methods are investigated in the context of software quality, with an emphasis on ongoing attempts to prevent and discover errors early in the software development lifecycle. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
IoT-Powered Health Monitoring System for Protecting Vital Organs Through Cloud-Based Diagnosis
The main objective of this research was the development and evaluation of an IoT- and machine learning-based health monitoring system capable of protecting patients vital organs through cloud diagnosis. This could be achieved by connecting a set of sensors, including temperature, pressure, heart rate, and oxygen sensors, to the patient and allowing them to communicate with the cloud to transmit real-time data via IoT technologies. The data could be further analyzed and predicted using cloud-based machine learning algorithms. This study investigated the performance of different machine learning models, including Artificial Neural Networks (ANN), Support Vector Machines (SVM), Naive Bayes (NB), and Negative Decision Trees (DT), for the purpose of patients health prediction based on the sensor data. After experimentation and evaluation, we found that the ANN model demonstrated the best predictive ability, with an accuracy level of 99.45%. The SVM, NB, and DT models also demonstrated good performance, with the accuracies of 96.5%, 94.34%, and 91.2%, respectively. Therefore, this research demonstrated that IoT and machine learning technologies could be successfully employed in healthcare for remote patient monitoring and timely prediction. The created system allows for real-time monitoring, which enables early prediction, potentially leading to improved patient outcomes, cost savings, and higher efficiency of provided care. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
Comparative Analysis of Classification Models Using Various Feature Sets
Feature selection is a fundamental step in Machine Learning (ML) that involves choosing some input data that would enhance the model performance. The model is able to run faster using lesser computational resources while giving reasonable results. Hence, feature selection as important as selection of a good model. In this chapter the aim is to analyze how the performance of different multiclass classification algorithms is affected on different features. The algorithms used are K-Nearest Neighbor (KNN), Support Vector Machines (SVM), Linear Discriminant Analysis (LDA), and Convolutional Neural Network (CNN) on the CIFAR-10 dataset. To obtain the new dataset with modified features, we use dimension reduction methods on the original dataset. The new dataset is at least 500x smaller, and we have noticed that in the best case scenarios reducing dimensions reduces the accuracy score only marginally. The SVM is the most consistent among the experimented models. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
Optimizing Diabetes Prediction Models for Enhanced Health Data Processing
Diabetes prediction is crucial for early intervention and personalized treatment. This study uses a multimodal strategy, including prediction algorithms, downsampling, feature engineering, exploratory data analysis (EDA), cross-validation, and classification techniques. EDA is used to understand diabetes-specific features, while downsampling ensures fair representation of instances with and without diabetes. Classification algorithms categorize people into appropriate diabetes risk groups using machine learning. Cross-validation evaluates predictive models in various data scenarios. The study emphasizes the value of specialized methods and domain-specific expertise in diabetes prediction, emphasizing the need for accurate risk assessment in healthcare decision-making and the potential for proactive interventions. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
A Comprehensive Study Using Convolutional Neural Networks as a Method for Multi-class Skin Cancer Image Classification
Skin disorders occur more frequently than other kinds of diseases. Skin diseases can be attributed to a number of aspects, like fungi, bacteria, viruses, allergies, and so on. The rapid advancement of healthcare centered around lasers and photonics has rendered it feasible to diagnose skin disorders in a more accurate and timely manner. However, the cost of such a diagnostic remains extremely limited and prohibitively expensive. As a result, the use of image processing methods is beneficial in the initial phases of designing a computerized dermatology screening system. The retrieval of characteristics is an extremely important step in classifying skin disorders. The use of computer vision may play a crucial role in the diagnosis of a variety of skin conditions using a variety of approaches. The strategy we have proposed is straightforward and quick and requires no expensive technology besides a computer and a camera. When applied to the inputs of a colored picture, the method is successful. After that, resize a portion of the image to retrieve attributes with a pretrained convolutional neural network. The attribute was then classified using the multi-class XGBoost program. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
Deep Learning-Based Dynamic Vision: Classifying Hand Gestures
In the field of hand gesture recognition, this research introduces novel approaches by utilising a variety of state-of-the-art deep learning models, including YOLOv6, YOLOv8, VGG16, VGG19, and ResNet50. Our work involved rigorous dataset annotation and preprocessing, coupled with custom data augmentation techniques tailored for real-world scenarios. The results were excellent, as YOLOv6 exhibited remarkable precision, achieving an impressive Average Precision (AP) of 97.4% and recall (AR) of 90%. Meanwhile, YOLOv8s prowess shone in specific classes, where it attained a remarkable mean Average Precision (mAP) of 89%. We further explored the capabilities of classical Convolutional Neural Networks (CNNs) such as VGG16 and VGG19. These models demonstrated solid performance with an average accuracy of 74 and 67%, respectively. Our study also explored the utilization of ResNet50, which, despite its popularity in other computer vision tasks, showed a lower accuracy of 33% in the context of hand gesture recognition. This research showcases a significant leap beyond the conventional CNN-based research in hand gesture recognition, as we integrated both object detection and image classification models into the evaluation framework. Looking ahead, our research opens doors to exploring ensemble models that synergize the strengths of YOLOv6, YOLOv8, VGG16, and VGG19, promising a harmonized performance across all classes. Moreover, we advocate for further research into transfer learning techniques, anticipating even higher accuracy levels in scenarios constrained by limited training data. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
Enhancing Weed Recognition in Cotton Fields Through Advanced Imaging and Learning Techniques
This research investigates the efficacy of weed recognition models in cotton fields through advanced imaging and machine learning techniques. Utilizing 10 trials, the models, namely K-NN and GBM, were evaluated across multiple performance metrics. Results reveal that GBM consistently outperformed K-NN in accuracy, precision, recall, and F1 score, with average values of 0.88, 0.89, 0.86, and 0.88, respectively, compared to K-NN's averages of 0.85, 0.87, 0.82, and 0.85. Moreover, GBM exhibited higher AUC values (0.94) than K-NN (0.92) in ROC curve analysis, indicating superior discrimination ability. Additionally, k-fold cross-validation demonstrated GBM's higher mean accuracy (0.89) and F1 score (0.88) compared to K-NN (mean accuracy: 0.86, mean F1 score: 0.85). Additionally, integrating temporal data analysis could improve the models ability to detect weed growth patterns over time. Real-time monitoring capabilities and automated decision-making systems could streamline weed management practices in agricultural settings. Furthermore, expanding the study to encompass diverse geographical regions and crop types would provide valuable insights into the generalizability and robustness of the developed models. Overall, continued research in this domain holds the potential to revolutionize weed management strategies and contribute to sustainable agriculture practices. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
Forecasting Breast Cancer with Integrated Pre-trained CNN and Machine Learning Framework from CT Images
This article investigates machine learning techniques effectiveness at using computed tomography (CT) images to forecast breast cancer, hoping to expedite early identification and plan treatment. Drawing on many different machine learning models, such as CNN, SVM, VGG16, RNN and RF, we did extensive work to measure their performance distinguishing between malignant and benign breast tissue regions. The dataset includes 2,430 CT pictures, with 70% for training and 30% for testing. It has been carefully selected and prepared in order to guarantee robustness and consistency. The precision, and in-sensitivity measure the accuracy, sensitivity, specificity is used as analytic indicators to measure the models ability to predict regions of breast cancer accurately. Our findings show that the proposed CNN model achieved an accuracy of 98.75%, superior performance. Other machine learning models are also highlighted in this study, demonstrating how breast cancer can be predicted using various methods. This research will determine the forms and technologies suitable for breast cancer forecasting. Medical imaging and clinical decision-making can move forward because of this research, offering a glimpse into how integrated machine-learning systems can bring greater precision to diagnosis and prognosis. By careful experimentation and analysis, we hope to prepare people for early intervention and personalized treatment methods. This will make for improved patient outcomes in fighting breast cancer. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
Multilingual Voice-Assisted for Traffic Sign Detection and Classification in Adverse Weather Conditions
In a world where millions of people are wounded in auto accidents each year due to negligence, a lack of understanding of traffic laws, and bad weather, there is an urgent need for greater road safety. This is particularly the case in India, where a disproportionately high number of traffic accidents lead to numerous fatalities. Ignoring traffic signs raises these risks and endangers not only vehicles but also passengers and pedestrians. This project addresses the significant issue of traffic sign recognition in bad weather and offers voice-based instruction in many languages to increase road safety. Using a mix of state-of-the-art technologies, including YOLOv8 for real-time sign detection and the Google Translate API, which supports NLP tasks, this research offers a full solution. The model's remarkable precision and efficacy underscore its capacity to revolutionize traffic safety and furnish a more secure and expedient driving encounter. With the world moving towards more autonomous mobility, this study is laying the groundwork for safer and more effective driving in the future. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
Effective Methods of Waste Management Practices in Green Hotels Toward Green Brand Image: An Empirical Study
The changes in consumer tastes are a significant motivating factor for hotels to adopt environmentally friendly practices. Recently, there has been a significant focus on the perils of climate change and the significance of adopting sustainable practices. As a result, environmentalism now influences almost every consumer decision. With the increasing awareness of environmental sustainability in the hospitality industry, the options for eco-friendly hotels are expanding, providing a wider range of choices for potential customers. Thus, this study seeks to examine the efficient strategies employed by green hotels for trash management to enhance their green brand image. Customer data from hotels was gathered and examined using SPSS 25 software. The findings suggest that implementing energy efficiency measures, promoting water conservation, and adopting sustainable and environmentally conscious building practices are effective approaches to waste management that can improve a companys brand image. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
Voluntary Carbon Markets: Bridging Climate Solutions Through Sustainable Finance
Voluntary carbon markets (VCMs) enable companies and individuals to offset their carbon footprint by acquiring carbon credits from programs that decrease or eliminate emissions, such as renewable energy or forest conservation activities. As climate change mitigation becomes more prominent on corporate and investor agendas, voluntary carbon markets provide a framework for channeling funding toward emissions-reduction operations. The threat of climate change necessitates immediate action, pushing sustainable finance to the forefront. Carbon offsetting appears as an intriguing but contentious option in this changing ecology. It enables entities to compensate for their greenhouse gas emissions by funding programs that reduce emissions elsewhere. However, worries about the integrity of some offsets remain, emphasizing the need for strong standards and verification systems. This includes aligning lending, insurance, and investment agendas with sustainability objectives such as the Paris Agreement. As the global economy undergoes structural transformations to meet climate targets, sustainable finance invests in VCM projects to direct funds towards programs that reduce carbon emissions, boost renewable energy use, and support sustainable practices. This linkage strengthens the financial sectors role in combating climate change while also promoting economic growth and resilience. Integrating VCMs into sustainable finance strategies improves transparency and accountability in carbon offsetting methods, ensuring that investments contribute to emissions reductions and environmental benefits. This conceptual study seeks to contribute to a thorough understanding of the collaboration between VCMs, sustainable finance, and climate action, emphasizing the significance of collective action in companies and individuals achieving environmental sustainability and minimizer by the effects of climate change. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
The Green Evolution: Transforming Supply Chains for a Sustainable Future
The paper analyses systemic, multi-stakeholder effort required to overcome continual challenges. Proactive solutions that prioritize supplier collaboration and transparency, aided by emerging technologies and public-policy guidance, can drive the necessary sustainability transformation. The required sustainability change can be initiated by proactive solutions that place a high priority on supplier collaboration and transparency, supported by new technology and recommendations from public policy. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
SMOTE-Based Sampling for Addressing Class Imbalance
Various real-world applications, including as text categorization, categorization of gender in facial recognition for medical evaluation, fraud detection, and satellites analysis of images for oil-spill monitoring, are frequently plagued by imbalanced data. The majority class is commonly the primary focus of machine learning algorithms, with the minority samples being ignored or classified in a secondary manner. Nevertheless, despite their rarity, these minority samples are very important. When it comes to classification tasks, the issue of class imbalancewhere one class is underrepresented relative to anotherpresents a significant barrier. Specialized approaches including SMOTE, ADASYN, and cost-sensitive voting classifiers have been developed to address this problem. The minority class is oversampled in these methods, synthetic samples are created adaptively, and different prices are placed on misclassification mistakes in order to solve the issue of class imbalance. As a result, rigorous assessment utilizing pertinent metrics and cost considerations are required. The efficacy of these strategies, however, depends on dataset features and problem-specific factors. Class imbalance is still a hot topic for study, and there has been constant innovation in novel methods that are adapted to certain dataset characteristics and application fields. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
Linking the Path to Zero Hunger: Analysing Sustainable Development Goals Within the Context of Global Sustainability
A global framework, the Sustainable Development Goals of the United Nations, are designed to tackle the most urgent global issues. SDG 2, which stands for Zero Hunger, demonstrates a robust interconnection with the remaining seventeen goals since achieving food security and improved nutrition requires an all-encompassing approach that addresses the interconnected challenges presented by poverty, health, education, gender equality, climate change, and sustainable resource management. Within this framework, the research endeavors to ascertain the interrelationships among SDG 2 and other goals and analyze the critical goals that drive the achievement of SDG 2. Furthermore, the study provides an exhaustive analysis of the positions adopted by different nations concerning SDG 2. The results indicate that the SDGs are interconnected; while SDG 2 is closely linked to several other SDGs, their respective impacts differ. Furthermore, it has been determined that policies are crucial to attaining the SDGs. Without a transformation in agri-food systems that enhances resilience and facilitates the provision of affordable, nutritious foods and healthy diets, the current state of affairs will persist. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
A Comprehensive Survey of Methods for Identifying Counterfeit Banknotes Using Image Processing and Machine Learning
The world economy is threatened by counterfeit currencies. Counterfeit currencies are often difficult, time-consuming and ineffective to identify manually. Automated methods based on image processing techniques and machine learning algorithms are helpful in detecting counterfeit notes. This survey paper reviews the current strategies on fake banknote detection using image processing techniques and machine learning algorithms. We discuss various stages of the detection process, including image acquisition, preprocessing, feature extraction and classification. Furthermore, we analyze the limitations and comparative performance of different algorithms and approaches mentioned in the literature. The survey aims to provide insights into the various methodologies, challenges and future directions in the field of fake banknote detection, facilitating the development of more robust and effective counterfeit detection systems. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.