Big Data最新文献

The demand for intensive care units (ICUs) is steadily increasing, yet there is a relative shortage of medical staff to meet this need. Intensive care work is inherently heavy and stressful, highlighting the importance of optimizing these units' working conditions and processes. Such optimization is crucial for enhancing work efficiency and elevating the level of diagnosis and treatment provided in ICUs. The intelligent ICU concept represents a novel ward management model that has emerged through advancements in modern science and technology. This includes communication technology, the Internet of Things (IoT), artificial intelligence (AI), robotics, and big data analytics. By leveraging these technologies, the intelligent ICU aims to significantly reduce potential risks associated with human error and improve patient monitoring and treatment outcomes. Deep learning (DL) and IoT technologies have huge potential to revolutionize the surveillance of patients in the ICUs due to the critical and complex nature of their conditions. This article provides an overview of the most recent research and applications of linical data for critically ill patients, with a focus on the execution of AI. In the ICU, seamless and continuous monitoring is critical, as even little delays in patient care decision-making can result in irreparable repercussions or death. This article looks at how modern technologies like DL and the IoT can improve patient monitoring, clinical results, and ICU processes. Furthermore, it investigates the function of wearable and advanced health sensors coupled with IoT networking systems, which enable the secure connection and analysis of various forms of patient data for predictive and remote analysis by medical professionals. By assessing existing patient monitoring systems, outlining the roles of DL and IoT, and analyzing the benefits and limitations of their integration, this study hopes to shed light on the future of ICU patient care and identify opportunities for further research.

Our online lives generate a wealth of behavioral records-digital footprints-which are stored and leveraged by technology platforms. These data can be used to create value for users by personalizing services. At the same time, however, it also poses a threat to people's privacy by offering a highly intimate window into their private traits (e.g., their personality, political ideology, sexual orientation). We explore the concept of cloaking: allowing users to hide parts of their digital footprints from predictive algorithms, to prevent unwanted inferences. This article addresses two open questions: (i) can cloaking be effective in the longer term, as users continue to generate new digital footprints? And (ii) what is the potential impact of cloaking on the accuracy of desirable inferences? We introduce a novel strategy focused on cloaking "metafeatures" and compare its efficacy against just cloaking the raw footprints. The main findings are (i) while cloaking effectiveness does indeed diminish over time, using metafeatures slows the degradation; (ii) there is a tradeoff between privacy and personalization: cloaking undesired inferences also can inhibit desirable inferences. Furthermore, the metafeature strategy-which yields more stable cloaking-also incurs a larger reduction in desirable inferences.

Physical therapists specializing in sports rehabilitation detection help injured athletes recover from their wounds and avoid further harm. Sports rehabilitators treat not just commonplace sports injuries but also work-related musculoskeletal injuries, discomfort, and disorders. Sensor-equipped Internet of Things (IoT) monitors the real-time location of medical equipment such as scooters, cardioverters, nebulizer treatments, oxygenation pumps, or other monitor gear. Analysis of medicine deployment across sites is possible in real time. Health care delivery based on digital technology to improve access, affordability, and sustainability of medical treatment is known as digital health care. The challenging characteristics of such sports injury rehabilitation for digital health care are playing position, game strategies, and cybersecurity. Hence, in this research, health care IoT-enabled body area networks (HIoT-BAN) have been designed to improve sports injury rehabilitation detection for digital health care. The health care sector may benefit significantly from IoT adoption since it allows for enhanced patient safety; health care investment management includes controlling the hospital's pharmaceutical stock and monitoring the heat and humidity levels. Digital health describes a group of programmers made to aid health care delivery, whether by assisting with clinical decision-making or streamlining back-end operations in health care institutions. A HIoT-BAN effectively predicts the rise in sports injury rehabilitation detection with faster digital health care based on IoT. The research concludes that the HIoT-BAN effectively indicates sports injury rehabilitation detection for digital health care. The experimental analysis of HIoT-BAN outperforms the IoT method in terms of performance, accuracy, prediction ratio, and mean square error rate.

Liver cirrhosis stands as a prominent contributor to mortality, impacting millions across the United States. Enabling health care providers to predict early mortality among patients with cirrhosis holds the potential to enhance treatment efficacy significantly. Our hypothesis centers on the correlation between mortality and laboratory test results along with relevant diagnoses in this patient cohort. Additionally, we posit that a deep learning model could surpass the predictive capabilities of the existing Model for End-Stage Liver Disease score. This research seeks to advance prognostic accuracy and refine approaches to address the critical challenges posed by cirrhosis-related mortality. This study evaluates the performance of an artificial neural network model for liver disease classification using various training dataset sizes. Through meticulous experimentation, three distinct training proportions were analyzed: 70%, 80%, and 90%. The model's efficacy was assessed using precision, recall, F1-score, accuracy, and support metrics, alongside receiver operating characteristic (ROC) and precision-recall (PR) curves. The ROC curves were quantified using the area under the curve (AUC) metric. Results indicated that the model's performance improved with an increased size of the training dataset. Specifically, the 80% training data model achieved the highest AUC, suggesting superior classification ability over the models trained with 70% and 90% data. PR analysis revealed a steep trade-off between precision and recall across all datasets, with 80% training data again demonstrating a slightly better balance. This is indicative of the challenges faced in achieving high precision with a concurrently high recall, a common issue in imbalanced datasets such as those found in medical diagnostics.

This study investigates customers' product design requirements through online comments from social media, and quickly translates these needs into product design specifications. First, the exponential discriminative snowball sampling method was proposed to generate a product-related subnetwork. Second, natural language processing (NLP) was utilized to mine user-generated comments, and a Graph SAmple and aggreGatE method was employed to embed the user's node neighborhood information in the network to jointly define a user's persona. Clustering was used for market and product model segmentation. Finally, a deep learning bidirectional long short-term memory with conditional random fields framework was introduced for opinion mining. A comment frequency-invert group frequency indicator was proposed to quantify all user groups' positive and negative opinions for various specifications of different product functions. A case study of smartphone design analysis is presented with data from a large Chinese online community called Baidu Tieba. Eleven layers of social relationships were snowball sampled, with 14,018 users and 30,803 comments. The proposed method produced a more reasonable user group clustering result than the conventional method. With our approach, user groups' dominating likes and dislikes for specifications could be immediately identified, and the similar and different preferences of product features by different user groups were instantly revealed. Managerial and engineering insights were also discussed.

We present an efficient incremental learning algorithm for Deterministic Finite Automaton (DFA) with the help of inverse query (IQ) and membership query (MQ). This algorithm is an extension of the Identification of Regular Languages (ID) algorithm from a complete to an incremental learning setup. The learning algorithm learns by making use of a set of labeled examples and by posing queries to a knowledgeable teacher, which is equipped to answer IQs along with MQs and equivalence query. Based on the examples (elements of the live complete set) and responses against IQs from the minimally adequate teacher (MAT), the learning algorithm constructs the hypothesis automaton, consistent with all observed examples. The Incremental DFA Learning algorithm through Inverse Queries (IDLIQ) takes $O\left(\left|\Sigma \right|N+\left|P_c\right|\left|F\right|\right)$ time complexity in the presence of a MAT and ensures convergence to a minimal representation of the target DFA with finite number of labeled examples. Existing incremental learning algorithms; the Incremental ID, the Incremental Distinguishing Strings have polynomial (cubic) time complexity in the presence of a MAT. Therefore, sometimes, these algorithms even fail to learn large complex software systems. In this research work, we have reduced the complexity (from cubic to square form) of the DFA learning in an incremental setup. Finally, we prove the correctness and termination of the IDLIQ algorithm.

Deep learning and big data techniques have become increasingly popular in traffic flow forecasting. Deep neural networks have also been applied to traffic flow forecasting. Furthermore, it is difficult to determine whether neural networks can be used for accurate traffic flow prediction. Moreover, since the network model is poorly structured and the parameter optimization technique is inappropriate, the traffic flow prediction is inaccurate because of the lack of certainty. The proposed system overcomes these problems by combining multiple simple recurrent long short-term memory (LSTM) neural networks with time traits to predict traffic flow using a deep gated stacked neural network. To deepen the model, the hidden layers have been trained using an unsupervised layer-by-layer approach. This approach provides a systematic representation of the time series data. A systematic representation of hidden layers improves the accuracy of time series forecasting by capturing information at multiple levels. Furthermore, it emphasizes the importance of model structure, random weight initialization, and hyperparameters used in stacked LSTM to enhance predictive performance. The prediction efficacy of the deep gated stacked LSTM model is compared with that of the gated recurrent unit model and the stacked autoencoder model.

Pre-COVID-19, most of the supply chains functioned with more capacity than demand. However, COVID-19 changed traditional supply chains' dynamics, resulting in more demand than their production capacity. This article presents a multiobjective and multiperiod supply chain network design along with customer prioritization, keeping in view price discounts and outsourcing strategies to deal with the situation when demand exceeds the production capacity. Initially, a multiperiod, multiobjective supply chain network is designed that incorporates prices discounts, customer prioritization, and outsourcing strategies. The main objectives are profit and prioritization maximization and time minimization. The introduction of the prioritization objective function having customer ranking as a parameter and considering less capacity than demand and outsourcing differentiates this model from the literature. A four-valued neutrosophic multiobjective optimization method is introduced to solve the model developed. To validate the model, a case study of the supply chain of a surgical mask is presented as the real-life application of research. The research findings are useful for the managers to make price discounts and preferred customer prioritization decisions under uncertainty and imbalance between supply and demand. In future, the logic in the proposed model can be used to create web application for optimal decision-making in supply chains.

This study contributes to the research on Internet of Things data visualization for business intelligence processes, an area of growing interest to scholars, by conducting a systematic review of the literature. A total of 237 articles published over the past 11 years were obtained and compared. This made it possible to identify the top contributing and most influential authors, countries, publishers, institutions, papers, and research findings, together with the challenges facing current research. Based on these results, this work provides a thorough insight into the field by proposing four research categories (Technology infrastructure, Case examples, Final-user experience, and Big Data tools), together with the development of these research streams over time and their future research directions.

Predicting propagation cascades is crucial for understanding information propagation in social networks. Existing methods always focus on structure or order of infected users in a single cascade sequence, ignoring the global dependencies of cascades and users, which is insufficient to characterize their dynamic interaction preferences. Moreover, existing methods are poor at addressing the problem of model robustness. To address these issues, we propose a predication model named DropMessage Hypergraph Attention Networks, which constructs a hypergraph based on the cascade sequence. Specifically, to dynamically obtain user preferences, we divide the diffusion hypergraph into multiple subgraphs according to the time stamps, develop hypergraph attention networks to explicitly learn complete interactions, and adopt a gated fusion strategy to connect them for user cascade prediction. In addition, a new drop immediately method DropMessage is added to increase the robustness of the model. Experimental results on three real-world datasets indicate that proposed model significantly outperforms the most advanced information propagation prediction model in both MAP@k and Hits@K metrics, and the experiment also proves that the model achieves more significant prediction performance than the existing model under data perturbation.