To address the strong mutual coupling issue in the integration of a four-petal antenna 2×2 array for satellite communication, 5G base stations, and radar systems, this paper designs a gate-shaped passive parasitic bridge decoupling structure suitable for the 4.8~5.2 GHz frequency band. Without modifying the antenna body, this structure cancels the inherent mutual coupling between elements through reverse coupling fields. Electromagnetic simulation results show that within the target frequency band, the return loss S11 of the elements is stable at -12~-14 dB, the mutual coupling S21 between adjacent elements reaches -22~-23 dB, the coupling S31 between diagonal elements reaches -20~-22 dB, and the coupling S41 between remote elements reaches -17~-18 dB. Meanwhile, the main lobe gain remains stable at approximately 7.17 dBi with minimal performance degradation. Compact and easy to manufacture, this structure can provide a reliable technical reference for mutual coupling suppression in similar arrays.

This meta-analysis aimed to systematically evaluate the efficacy (including improvement of biochemical indicators and clinical outcomes) and safety of the molecular adsorbent recirculating system (MARS) in the treatment of hepatic encephalopathy associated with liver failure. PubMed, EMBASE, Cochrane Library, Web of Science, CNKI, Wanfang, and VIP databases were systematically searched from inception to April 2024. Predefined inclusion criteria were used to assess study quality, and statistical analyses were performed using RevMan 5.4. Seven studies were included. Results showed that MARS significantly improved hemoglobin [standard mean difference (SMD): -0.81, 95% confidence interval (CI): -1.42 to -0.19, P  = 0.01], creatinine (SMD: -0.46, 95% CI: -0.68 to -0.24, P  < 0.0001), and international normalized ratio (INR) (SMD: -0.22, 95% CI: -0.43 to -0.01, P  = 0.004). However, effects on albumin (SMD: 0.60, 95% CI: -0.22 to 1.41, P  = 0.15) and bilirubin (SMD: -0.21, 95% CI: -0.78 to 0.35, P  = 0.46) were NS. No significant association was observed between MARS treatment and increased risk of bleeding (risk ratio: 1.23, 95% CI: 0.83 to 1.81, P  = 0.30) or infection (risk ratio: 1.23, 95% CI: 0.97 to 1.56, P  = 0.09). In conclusion, preliminary analysis suggests that MARS may help improve creatinine and INR levels in hepatic encephalopathy patients without significantly increasing bleeding or infection risks. However, because of the limited number and suboptimal quality of included studies and presence of heterogeneity, current evidence is insufficient. Rigorously designed, adequately powered, high-quality randomized controlled trials are needed to validate the precise efficacy and safety of MARS in hepatic encephalopathy treatment.

As a complex and destructive natural disaster, the characteristics of typhoons are closely related to human activities, and their accurate categorization is of vital significance for improving disaster warning and management capabilities. This study highlights the key role of typhoon clustering in analyzing typhoon behaviors, aiming to provide reliable support for disaster prevention and control. Based on the NOAA meteorological dataset from 2003 to 2024, this study firstly adopts the K-means clustering algorithm to classify typhoons into seven categories and then utilizes eight machine learning models to train and validate the classification results, and introduces the Shapley's additive interpretation (SHAP) algorithm to enhance the interpretability of the models. The study data covers a variety of features such as air temperature, wind speed, atmospheric pressure, and weather station observations, etc. After a systematic preprocessing process, a feature matrix containing key variables such as typhoon intensity and moving speed is constructed. The results show that the XGBoost model outperforms others across multiple evaluation metrics (Accuracy: 0.992, Precision: 0.989, Recall: 0.992, F1.5 Score: 0.990), highlighting its exceptional capability in managing complex weather classification tasks. The seven categories of typhoon types classified by K-means exhibit different feature patterns, while the SHAP analysis further reveals the effects of each feature on the classification and its potential interactions. This study not only verifies the effectiveness of K-means combined with machine learning in typhoon classification but also lays a solid scientific foundation for accurate prediction, risk assessment and optimization of management strategies for typhoon disasters through the in-depth analysis of feature impacts.

The core challenge in basketball tactic modeling lies in efficiently extracting complex spatial-temporal dependencies from historical data and accurately predicting various in-game events. Existing state-of-the-art (SOTA) models, primarily based on graph neural networks (GNNs), encounter difficulties in capturing long-term, long-distance, and fine-grained interactions among heterogeneous player nodes, as well as in recognizing interaction patterns. Additionally, they exhibit limited generalization to untrained downstream tasks and zero-shot scenarios. In this work, we propose a Spatial-Temporal Propagation Symmetry-Aware Graph Transformer for fine-grained game modeling. This architecture explicitly captures delay effects in the spatial space to enhance player node representations across discrete-time slices, employing symmetry-invariant priors to guide the attention mechanism. We also introduce an efficient contrastive learning strategy to train a Mixture of Tactics Experts module, facilitating differentiated modeling of offensive tactics. By integrating dense training with sparse inference, we achieve a 2.4x improvement in model efficiency. Moreover, the incorporation of Lightweight Graph Grounding for Large Language Models enables robust performance in open-ended downstream tasks and zero-shot scenarios, including novel teams or players. The proposed model, TacticExpert, delineates a vertically integrated large model framework for basketball, unifying pretraining across multiple datasets and downstream prediction tasks. Fine-grained modeling modules significantly enhance spatial-temporal representations, and visualization analyzes confirm the strong interpretability of the model.

AI-aided clinical diagnosis is desired in medical care. Existing deep learning models lack explainability and mainly focus on image analysis. The recently developed Dynamic Uncertain Causality Graph (DUCG) approach is causality-driven, explainable, and invariant across different application scenarios, without problems of data collection, labeling, fitting, privacy, bias, generalization, high cost and high energy consumption. Through close collaboration between clinical experts and DUCG technicians, 46 DUCG models covering 54 chief complaints were constructed. Over 1,000 diseases can be diagnosed without triage. Before being applied in real-world, the 46 DUCG models were retrospectively verified by third-party hospitals. The verified diagnostic precisions were no less than 95%, in which the diagnostic precision for every disease including uncommon ones was no less than 80%. After verifications, the 46 DUCG models were applied in the real-world in China. Over one million real diagnosis cases have been performed, with only 17 incorrect diagnoses identified. Due to DUCG's transparency, the mistakes causing the incorrect diagnoses were found and corrected. The diagnostic abilities of the clinicians who applied DUCG frequently were improved significantly. Following the introduction to the earlier presented DUCG methodology, the recommendation algorithm for potential medical checks is presented and the key idea of DUCG is extracted.

Since machine learning is applied in medicine, more and more medical data for prediction has been produced by monitoring patients, such as symptoms information of diabetes. This paper establishes a frame called the Diabetes Medication Bayes Matrix (DTBM) to structure the relationship between the symptoms of diabetes and the medication regimens for machine learning. The eigenvector of the DTBM is the stable distribution of different symptoms and medication regimens. Based on the DTBM, this paper proposes a machine-learning algorithm for completing missing medical data, which provides a theoretical basis for the prediction of a Bayesian matrix with missing medical information. The experimental results show the rationality and applicability of the given algorithms.

Finding an optimum way to identify stocks with less delisting risk is critical for every investor in the stock market. However, this procedure is often done based on personal experience, which doesn’t fully utilize the historical delisting records. This convention of selecting stocks might result in a greater loss since it merely involves subjective judgment, especially for individual investors. Our research proposes a probabilistic approach for identifying the delisting risk associated with different industry sectors, given the P/B ratio level distribution. And this research offers a customized guide for individual investors to better choose the safer investment options related to the stocks’ industry sectors. The completion of our conditional probability matrix is operated under the high-rank assumption, together with the features of Bayesian matrices. The experimental results for our domestic delisting stocks supports the validity and usefulness of our method.

Due to the surge in COVID-19 cases, hospitals have had to receive many more patients than before, which has brought unprecedented pressure to the hospital system. Therefore, the emphasis of medical decision-making has shifted from reaching the best treatment effect to prioritizing the treatment of COVID-19 patients by hospitals, which is key to relieving the pressure on the hospital system and reducing the overall mortality rate of COVID-19. There is no doubt that establishing the prioritization of COVID-19 cases is fundamental and pivotal for hospitals to achieve the shift in medical decision-making. Prioritization of COVID-19 cases in previous studies was mostly based on one patient characteristic, mainly including age, health conditions, and gender. This paper focuses on two patient characteristics at the same time. The probability that a COVID-19 patient who died had a given health condition in a given age group is calculated using the matrix completion technique based on the high-rank assumption of Bayesian matrices and the properties of Markov matrices. The calculated results show that doctors should give patients over 55 with respiratory diseases, patients over 65 with circulatory diseases, and patients over 65 with diabetes a higher prioritization in COVID-19 treatment.

With the increasing scale of the urban subway, the total energy consumption of the subway has increased dramatically and poses a great challenge to the comfort of passengers and the punctuality of train operation. In order to ensure on-time train operation and passenger comfort, and at the same time reduce the energy consumption of subway operation, this paper proposes a Proximal Policy Optimization (PPO)-based optimization algorithm for the optimal control of subway train operation. Firstly, a reinforcement learning architecture for optimal control of subway train operation is constructed with the position and speed of train operation as the reinforcement learning state, energy consumption and comfort as the optimization objectives, and train operation time as the constraint. The proposed reinforcement learning model is trained by the PPO algorithm, and the reward scaling is added to the training process to accelerate the training speed and improve the efficiency of the algorithm. The experimental results show that the proposed PPO with reward scaling algorithm can effectively reduce train energy consumption and improve passenger comfort while ensuring on-time train operation.

The development of virtual coupling technology provides solutions to the challenges faced by urban rail transit systems. Train tracking control is a crucial component in the operation of virtual coupling, which plays a pivotal role in ensuring the safe and efficient movement of trains within the train and along the rail network. In order to ensure the high efficiency and safety of train tracking control in virtual coupling, this paper proposes an optimization algorithm based on Soft Actor-Critic for train tracking control in virtual coupling. Firstly, we construct the train tracking model under the reinforcement learning architecture using the operation states of the train, Proportional Integral Derivative (PID) controller output, and train tracking spacing and speed difference as elements of reinforcement learning. The train tracking control reward function is designed. Then, the Soft Actor-Critic (SAC) algorithm is used to train the virtual coupling train tracking reinforcement learning model. Finally, we took the Deep Deterministic Policy Gradient as the comparison algorithm to verify the superiority of the algorithm proposed in this paper.

This paper presents an optimization model to complete the missing data in the energy matchup matrix which compares different players’ energy. The optimization model searched for the solutions that can make the eigenvalue of the energy matrix the biggest that should be the principle one. To test the model, we built a Bayesian matrix of energy matchup which mixes the judgments of energy in a matchup between players given by experts and the statistical data gotten from each game in different positions. The proposed method can complete the probabilities given by the Bayesian matrix. Finally, an implementation shows the effectiveness and rationality of the model.

For patients in the early stages of diabetes, it is crucial for patients and doctors to make treatment decisions to prevent the condition from getting worse and developing complications such as diabetic eyes and feet. Both undertreatment and overtreatment can do harm to the patient. In this paper, the probability of taking each treatment option for different symptoms is complemented by a database of existing successful cases of diabetic treatment options to recommend an appropriate treatment for patients with specific symptoms. The matrix completion process adopts the assumption of high rank and completes the matrix based on the unique properties of Bayesian matrices. The results demonstrate the practicality and effectiveness of this algorithm.

Diabetes departments of hospitals provided more and more professional data from each case for data-driven learning. However, experts in hospitals argue that their experience can provide more reliable information than statistical data. This paper presents a Data-Experience intelligent model to predict the possibilities of diabetes complications with the occurrence of abnormal signs. The model merges data and experience mathematically by mixing human judging behavior from traditional Chinese medicine and statistical patient data from western medicine, which is under a particular situation. The probabilities given by the model can value the distributions of each diabetic complication while estimating the posterior when multi-abnormal signs occurred. An implement in a specific case is analyzed based on the statistical data and human judging performance, which shows the effectiveness and rationality of the Data-Experience Intelligent model for predicting diabetic complications proposed in this paper. The results give a more comprehensive prediction by synthesizing subjective information and objective information.

Bayesian learning has been successfully used in many fields to make decision or sense the outcome of causes or influences of events. However, the relations between causes and observed events are more complicated than the Bayesian inference can learn, like prediction in some psychology experiments, which should consider human experience and the interdependence of the events. This paper raises Bayesian learning to intelligent learning by putting events in a network to analysis the complicated criterions and interdependences among events with the individual probability judgments. Finally, the results of experiment used by intelligent learning illustrate more complex relations than Bayesian learning.

Concerning basketball injuries, trainers should carry out specific training programs for athletes based on their characteristics to reduce the risk of injury. In basketball games, BMI is a main index for players. Therefore, this paper proposes a Bayesian matrix completion algorithm to learn the correlation probability between BMI and various injuries of players, which can assist trainers to make customized training plans for various players. And this algorithm will be applied to the existing injury cases in NBA statistical database. The high rank assumption is adopted in the process of matrix completion, and the matrix is completed according to some special attributes of Bayesian matrix. Experimental results show that the algorithm is practical and effective.

Risk management is a key factor for smart city running. There are many risk events in a strict process like transportation management of a smart city or a medical surgery in a smart hospital, and every step may lead to one kind of risk or more. In view of the fact that the occurrence of the flow risks follows the sequence formed by each process step, this paper presents a Bayesian network under strict chain (BN_SC) to model this situation. In this model, the probabilistic reasoning formula is given according to the sequence of process steps, and the probabilities given by the model can do risk factor analysis to support the system to find an effective way to improve the process like machine manufacturing or a medical surgery. Finally, an example is analyzed based on the information given by doctors according to the situation of LC in their hospital located in Sichuan Province of China, which shows the effectiveness and rationality of the proposed BN_SC model.

This paper proposes a learning model of basketball players structure based on the Bayesian network. By analyzing the data of NBA 's players, we complete the structural learning of basketball players network based on the +/- values of 5 resident players in the Portland Trail Blazers team. We finally obtain a winning and losing models for the five resident players of the Portland Trail Blazers team, and we make suggestions for coaches about player rotation based on the analysis of the models.

Triangular intuitionistic fuzzy numbers (TIFNs) are effective and flexible to characterize the fuzziness and uncertainty in real-world problems. The theories of TIFNs have been used in multi-attribute decision making but are rarely applied in a two-sided matching decision. Therefore, it is important and necessary to investigate the two-sided matching problem with TIFNs. This paper develops a decision method for two-sided matching with triangular intuitionistic fuzzy numbers and applies it to smart environmental protection. First, a similarity measure between generalized triangular fuzzy numbers (TFNs) is presented. Then, a novel similarity measure between TIFNs is extended, where the maximum membership degrees and minimum non-membership degrees, areas, and perimeters are considered. With respect to the two-sided matching problem with TIFNs, the two-sided matching model with TIFNs is established. Using similarity measures between TIFNs, the similarity matrices of triangular intuitionistic fuzzy preference matrices are constructed by using the positive idea vectors. Then, the two-sided matching model with similarity measures is obtained. Using the arithmetic mean, normalization formulas and linear weighting, the two-sided matching model with similarity measures is transformed into a mono-objective model. The optimum matching scheme is obtained by solving the model. Thus, a similarity measure-based two-sided matching decision method for TIFNs is proposed. Finally, a matching example in smart environmental protection is provided to illustrate the advantages of the proposed method.

For named entity recognition technology in a specific domain,there are various identification methods corresponding to different fields.Different fileds of texts have their own unique textual features,which leads to the existing identification method is difficult to adapt to new specific domain.In order to solve this problem,this paper proposes a method based on conditional ran-dom field,semi-supervised learning and active learning,which forms a unified technical framework to adapt to the named entity recognition in each specific domain.This method constructs the feature set based on characteristics of rail transit text,then trains CRF to recognize named-entity of rail traffic text,and selects the samples with lower confidence level than the selected threshold, and then manually extends the training samples to achieve high goals.In order to validate the method, this paper carries on the experiment in the field of rail transit.The experimental results show that the method is effective and has a good recognition effect in the field of rail transit.

This paper combines the theory of hesitant fuzzy linguistic term sets (HFLTSs) with two-sided matching decision making (TSMDM). The related definitions of HFLTSs and two-sided matchings (TSMs) are introduced. Then, the problem of TSMDM with HFLTSs is presented. For solving this problem, a model of TSMDM with HFLTSs is developed. The AHP method is used to determine the important degrees of agents of each side. On this base, the model of TSMDM can be changed into a double-goal model with HFLTSs. Then, the double-goal model with HFLTSs is changed into the double-goal model with scores through using the proposed score function. Furthermore, the double-goal model can be changed into a single-goal model by using the linear weighting technique once again. The scheme of TSM can be obtained through solving the single-goal model. At last, an example with sensitive analysis is provided for the illustration of the presented approach of TSM.