The early and accurate diagnosis of Type 2 Diabetes (T2D) is crucial for effective management and prevention of complications. This study explores a novel approach to T2D diagnosis by integrating Bayesian, uncertain, and deterministic strategies. The proposed methodology leverages Bayesian inference to handle uncertainty in patient data, while deterministic models provide clear decision rules based on clinical parameters. By combining these strategies, the model aims to improve diagnostic accuracy, reduce uncertainty in predictions, and offer robust support for clinical decision-making. This hybrid approach addresses the challenges of incomplete or noisy data, providing a more reliable framework for the timely identification of T2D in diverse patient populations. The evaluation of this hybrid model is based on more scenarios and comparisons. This evaluation proves that our hybrid model is more efficient than any uncertain, Bayesian, and deterministic model.
Evaluating cars is a common multi-attribute decision-making (MADM) problem because it requires considering multiple criteria at once. However, assessing these attributes often involves subjective judgments and incomplete data. Consequently, the decision-making process contains inherent uncertainties that necessitate advanced models to handle imprecise and ambiguous evaluations. This paper is aimed at developing novel complex hybrid hypersoft set structures: the possibility complex fuzzy hypersoft set (pCFHS), the possibility complex intuitionistic fuzzy hypersoft set (pCIFHS), and the possibility complex neutrosophic hypersoft set (pCNHS), by simultaneously integrating complex fuzzy hypersoft sets, complex intuitionistic fuzzy hypersoft sets, and complex neutrosophic hypersoft sets with possibility degree-based arrangements. By using the aggregations of three proposed frameworks, three algorithms are presented and validated through the MADM problem of car evaluation. This evaluation considers four main attributes along with eight corresponding sub-attribute values. Importantly, the study incorporates fuzzy values to reflect degrees of possibility. This possibility grade is meant to assess the biasing trends of decision-makers during the evaluation. In the end, the rankings of the three proposed algorithms are compared to identify the most reliable structure.
Data envelopment analysis (DEA) has been widely used to assess the technical performances of decision-making units with inputs and outputs. However, there are few research studies on performance assessment in service-oriented (without inputs) systems in the presence of contextual variables. In the theoretical part of the paper, we develop a without-input-DEA model with contextual variables for situations where we face variability and uncertainty in the data. In the application part, the work process in healthcare sector is considered as a service-oriented process with contextual variables, desirable and undesirable outputs. A two-step procedure consisting of the use of the proposed model along with a regression analysis is developed to evaluate performance in the healthcare sector and the impact of contextual variables on efficiency. The results revealed that 27% of the hospitals studied were classified as inefficient. We also found that hospitals are inefficient in terms of mortality rate and infectious waste. Regarding the Number of inpatients, on average, the performance of the hospitals was acceptable. The coefficient was 0.0046, indicating a direct impact on efficiency.
Given the nonlinearity, instability and uncertainty of wind speeds, effectively harnessing wind energy to ensure stable operation of power systems is critical. Based on uncertainty theory, this paper proposes a hybrid CEEMDAN-UAR-BiLSTM-Attention wind speed prediction model. The CEEMDAN algorithm decomposes the original wind speed time series into multiscale Intrinsic Mode Function (IMF) components, thereby reducing data complexity and mitigating noise interference. Subsequently, each component is modeled separately, the UAR model captures the linear trend of low-frequency components, while high-frequency components are fed into a bidirectional LSTM equipped with an attention mechanism. By leveraging the bidirectional structure of LSTM, the model reveals long-term dependencies while focusing on critical predictive information. During the prediction process, the combined model achieves reduced RMSE and MAE, attaining a coefficient of determination [Formula: see text] of 0.89. This provides robust support for wind power planning and operation, showcasing significant practical value and broad application prospects.
This paper extends the research and applications of the uncertain Gordon–Schaefer model. We examine two stability criteria for the equation solutions: stability in mean and stability in [Formula: see text]-th moment. Then we collect [Formula: see text] data points from the International Pacific Halibut Commission and analyze the catch data using the uncertain Gordon–Schaefer equation. Through the uncertain hypothesis test, we demonstrate that the uncertain Gordon–Schaefer equation provides an adequate fit. Finally, we establish why the stochastic Gordon–Schaefer equation fails to characterize the halibut catch data effectively.
