Articles

Social media platforms like X (formerly Twitter) generate vast volumes of user-generated content that provide real-time insights into public sentiment. Despite the widespread use of traditional machine learning methods, their limitations in capturing contextual nuances in noisy social media text remain a challenge. This study leverages the Sentiment140 dataset, comprising 1.6 million labeled tweets, and develops predictive models for binary sentiment classification using Naive Bayes, Logistic Regression, and the transformer-based BERT model. Experiments were conducted on a balanced subset of 12,000 tweets after comprehensive NLP preprocessing. Evaluation using accuracy, F1-score, and confusion matrices revealed that BERT significantly outperforms traditional models, achieving an accuracy of 89.5% and an F1-score of 0.89 by effectively modeling contextual and semantic nuances. In contrast, Naive Bayes and Logistic Regression demonstrated reasonable but consistently lower performance. To support practical deployment, we introduce SentiFeel, an interactive tool enabling real-time sentiment analysis. While resource constraints limited the dataset size and training epochs, future work will explore full corpus utilization and the inclusion of neutral sentiment classes. These findings underscore the potential of transformer models for enhanced public opinion monitoring, marketing analytics, and policy forecasting.

This research is a collaboration involving a university team, a partnering corporation, and a hemodialysis clinic, which is a cross-disciplinary research initiative in the field of Artificial Intelligence of Things (AIoT) within the medical informatics domain. The research has two objectives: (1) The development of an <i>Internet of Things (IoT)</i><i>-based</i><i> Information System</i> customized for the hemodialysis machines at the clinic, including transmission bridges, clinical personnel dedicated web/app, and a backend server. The system has been deployed at the clinic and is now officially operational; (2) The research also utilized de-identified, anonymous data (collected by the officially operational system) to train, evaluate, and compare <i>Deep Learning-based Intradialytic Blood Pressure</i><i> (BP)</i><i>/Pulse </i><i>Rate (PR) </i><i>Predictive Models</i>, with subsequent suggestions provided. Both objectives were executed under the supervision of the Institutional Review Board (IRB) at Mackay Memorial Hospital in Taiwan. The system completed for objective one has introduced three significant services to the clinic, including automated hemodialysis data collection, digitized data storage, and an information-rich human-machine interface as well as graphical data displays, which replaces traditional paper-based clinical administrative operations, thereby enhancing healthcare efficiency. The graphical data presented through web and app interfaces aids in real-time, intuitive comprehension of the patients’ conditions during hemodialysis. Moreover, the data stored in the backend database is available for physicians to conduct relevant analyses, unearth insights into medical practices, and provide precise medical care for individual patients. The training and evaluation of the predictive models for objective two, along with related comparisons, analyses, and recommendations, suggest that in situations with limited computational resources and data, an Artificial Neural Network (ANN) model with six hidden layers, SELU activation function, and a focus on artery-related features can be employed for hourly intradialytic BP/PR prediction tasks. It is believed that this contributes to the collaborating clinic and relevant research communities.

Intersectionality has significantly enhanced our understanding of how overlapping social identities—such as race, ethnicity, gender, sex, class, and sexual orientation—interact to shape individual experiences. However, despite its theoretical importance, much of the existing literature has relied on qualitative approaches to define and study intersectionality, limiting its application in predictive modeling, comparative analysis, and policy development. This paper introduces the concept of Quantitative Intersectionality Scoring System (QISS), a novel approach that assigns numerical scores to intersecting identities, thereby enabling a more systematic and data-driven analysis of intersectional effects. We argue that QISS can substantially enhance the utility and predictive validity of quantitative models by capturing the complexities of multiple, overlapping social determinants. By presenting concrete examples, such as the varying impacts of socioeconomic mobility on life expectancy among different intersectional groups, we demonstrate how QISS can yield more precise and reliable forecasts. Such a shift would allow policymakers and service providers to dynamically assess economic and health needs, as well as the uncertainties around them, as individuals move through different social and economic contexts. QISS-based models could be more responsive to the complexities of intersecting identities, allowing for a more quantified and nuanced evaluation of policy interventions. We conclude by discussing the challenges of implementing QISS and emphasizing the need for further research to validate these quantifications using robust quantitative methods. Ultimately, adopting QISS has the potential to improve the accuracy of predictive models and the effectiveness of policies aimed at promoting social justice and health equity.

The management of supply chains is increasingly complex. This study provides a comparative analysis of the cost-benefit analysis for managing various risks. It identifies the financial implications of leveraging artificial intelligence in supply chains to better address risk. Empirical results show a business case for managing some sources of risk more proactively facilitated through predictive modeling techniques offered by AI. Across investigation streams, the use of AI results in an average total cost saving ranging from 41,254 to 4,099,617. Findings from our research can be used to inform managers and theorists about the implications of integrating AI technologies to manage risks in the supply chain. Our work also highlights areas for future research. Given the growing interest in studying sub-second forecasting, our research could be a point of departure for future investigations aimed at considering the impact of forecasting horizons such as an intra-day basis. We formulate a conceptual framework that considers how and to what extent performance evaluation metrics vary according to differences in the fidelity of predictive models and factor importance for identifying risks. We also utilize a mixed-method approach to demonstrate the applicability of our ideas in practice. Our results illustrate the financial implications of integrating AI predictive tools with business processes. Results suggest that real-world companies can circumvent inefficiencies associated with trying to manage many classes of risk via the use of AI-enhanced predictive analytics. As managers need to justify investment to top management, our work supports decision-making by providing a means of conducting a trade-off analysis at the tactical level.

The effects on the elderly are disproportionately Alzheimer’s disease (AD) is one of the most prevalent and chronic types of dementia. Alzheimer's disease (AD), a fatal illness that can harm brain structures and cells long before symptoms appear, is currently incurable and incurable. Using brain MRI pictures from a publicly accessible Kaggle dataset, this study suggests a prediction model based on Convolutional Neural Networks (CNNs) to help with the early detection of Alzheimer's disease. Four levels of dementia have been applied to the 6,400 photos in the collection: not demented, slightly demented, moderately demented, and considerably mildly demented. Pixel normalization, class balancing utilizing data augmentation techniques, and picture scaling to 128×128 pixels were all part of a thorough workflow for data preparation. To improve the gathering of spatial dependence in volumetric MRI data, a 3D convolutional neural network (CNN) architecture was used. We used important performance measures including F1-score, recall, accuracy, precision, and log loss to gauge the model's effectiveness. A review of the available data indicates that the total F1-score, accuracy, recall, and precision were 99.0%, 99.0%, and 99.38%, respectively. The findings demonstrate the model's potential for practical use in early AD diagnosis and establish its robustness with the help of confusion matrix analysis and performance curves.

The growing complexity and variability in healthcare delivery and costs within Medicare Advantage (MA) and Medicare Supplement (Medigap) plans present significant challenges for improving health outcomes and managing expenditures. Neural networks, a subset of artificial intelligence (AI), have shown considerable promise in optimizing healthcare processes, particularly in predictive modeling, personalized treatment recommendations, and risk stratification. This paper explores the application of neural networks in enhancing health outcomes within the context of Medicare Advantage and Supplement plans. We review how deep learning models can be leveraged to predict patient risk, optimize resource allocation, and identify at-risk populations for preventive interventions. Additionally, we discuss the potential for neural networks to improve claims processing, reduce fraud, and streamline administrative burdens. By integrating various data sources, including medical records, claims data, and demographic information, neural networks enable more accurate and efficient decision-making processes. Ultimately, this approach can lead to better patient care, reduced healthcare costs, and improved satisfaction for beneficiaries of these programs. The paper concludes by highlighting the current limitations, ethical considerations, and future directions for AI adoption in the Medicare Advantage and Supplement sectors.

While several application domains are exploiting the added-value of analytics over various datasets to obtain actionable insights and drive decision making, the public policy management domain has not yet taken advantage of the full potential of the aforementioned analytics and data models. To this end, in this paper authors present an overall architecture of a cloud-based environment that facilitates data retrieval and analytics, as well as policy modelling, creation and optimization. The environment enables data collection from heterogeneous sources, linking and aggregation, complemented with data cleaning and interoperability techniques. An innovative approach for analytics as a service is introduced and linked with a policy development toolkit, which is an integrated web-based environment to fulfil the requirements of the public policy ecosystem stakeholders [1]. Large information databases on various public issues exist, but their usage for public policy formulation and impact analysis has been limited so far, as no cloud-based service ecosystem exists to facilitate their efficient exploitation. With the increasing availability and importance of both public big and traditional data, the need to extract, link and utilize such information efficiently has arisen. Current data-driven web technologies and models are not aligned with the needs of this domain, and therefore, potential candidates for big data, cloud-based and service-oriented public policy analysis solutions should be investigated, piloted and demonstrated [2]. This paper presents the conceptual architecture of such an ecosystem based on the capabilities of state-of-the-art cloud and web technologies, as well as the requirements of its users.

Predictive modeling, supported by machine learning technology, aims to analyze data in order to guide decision-making towards actions generating desired values in the future. It encompasses the set of techniques used to build models that estimate the value of a certain variable predicting a forthcoming event from the past or current values of relevant attributes. In predictive healthcare modeling, the built models represent the relationship among the data concerning customer, provider, production, and other aspects of the healthcare environment in order to assist the decision processes in the prevention of diseases and in the planning of preventive actions by detection of high-risk patients. Contrary to trend analysis, whose goal is to describe past events, predictive models aim to provide useful indications regarding future events and changes. Predictive healthcare modeling supports actions that try to prevent the manifestation of diseases in healthy individuals or try to diagnose as early as possible the incidence of a disease in patients at risk. A sound predictive analysis encompasses not only the model-training task, but also the aspects of data quality, preprocessing, and fusion during its entire implementation lifecycle to ensure appropriate input data preparation. The robustness of the predictive model and its results depends highly on data quality. Due to the variety of data sources in healthcare environments, it becomes essential to use preprocessing in order to remove noise and inconsistencies. The increasing number of endorsable data exchange standards makes each data exchange achievable, but it demands the implementation of a data-governance program. In addition, the influence of the hospital-database architect on the architecture of an early-diagnosis model is important to guarantee appropriate input-formatting modularity.

Innovations in computing and communication technologies are reshaping finance. The seismic changes are casting uncertainty about the future of financial services. On one hand, fintech evangelists project a rosy future, asserting that the fast-moving algorithms can deliver low-cost financial services intuitively, customized to meet robust consumer expectations. On the other hand, many finance veterans fret that the traditional banking model could disintermediate, bleeding banks via a ‘death by a thousand cuts’, reducing them to passive portfolio holders with no direct customer relationship, eclipsed by digital giants which use their enormous treasure troves of customer data to offer banking as an added service with nearly free cost. Amidst the upbeat technological promises and apocalyptic forebodings, there are two constant, mostly agreed-upon, truths. The first is the vital importance of data. Advances in the internet, cloud computing, and record-keeping technologies are producing an ‘exponential growth in the volume and detail of data’. Some of this big data are personal information. Smartphones are deployed in almost all developed and emerging economies, serving as little spies; tracking, recording location histories, social networks, and app usage of their unsuspecting owners; often with a great degree of precision. ‘People are walking data-factories’ in this ‘mobile digital society’. Data are the fermentation of these global exchanges, electronic commerce and communication, and financial transactions. To just take Facebook as an example, it shares 30 million people a day through updates and posts, hosting personal information on 2.23 billion users. To the alarm of the uninformed public, much of this information is available for commercial harvest. The second constant is the rise of intelligent solutions. Consumers today—be it disclosed or not—are fed tailored clothes, music, film, holiday packages—almost anything you like, notably dynamic pricing, varying in accordance with individual profiles, or personalized search results. The availability of powerful computers has enabled comparable applications that are intended to make the system more responsive to their customer profiles and desires, or to capitalize competitive business possibilities. Such changes will transform the financial industry and occupy a prominent position among the mechanisms of policy competition, reshaping the way in which financial services are bestowed and led on the demand side.

This study aims to achieve a new understanding of how, why, and when consumer behavior is shaped, enacted, and experienced inside and across integrated digital ecosystems related to large-scale trackable goods, all in service of helping marketers optimize their business performance in the new economy. The pioneering understanding begins by exploring what motivates the choices of a homogeneous group of consumers to organize their consumption of national and store brand varieties of consumer package goods in a certain manner. Thereafter, the essay explores how, if at all, the other digital activities of consumers across various product-related digital spaces and on various platforms build expertise and interest in these products such that they exert an effect on the purchase choices for these products. The essay then advances to asking how online information seeking, in various product-related digital spaces, on various platforms, and from various sources, and taking place at various points in the purchase journey affects online-offline dynamics in purchasing these products. Thereafter, the research examines how paid digital communication in various product-related digital spheres and forms, enabled by consumer advertising engagement on various platforms, boosts the offline sales of these products. Finally, by employing a new methodology that combines consumer scanning data, self-reported online activity data, and transaction data collected from an ad-tech partner, the research presents a fresh set of marketing action levers and performance outcomes on chosen products. Along the way, progress is made on four under-investigated topics in the advertising literature – the role of consumer actors and their expertise in the online-offline purchasing dynamics for ads, advertising engagement, consumer digital spaces, and consumer digital activity investment.