Abstract The lung cancer incidence and mortality in China have always been high. Moreover, due to the limited level of professional technology, misdiagnosis and missed diagnosis of lung cancer often occur. To improve the accuracy of diagnosis, this paper proposes an interpretable diagnostic method for lung cancer based on Chinese electronic medical records (EMRs). First, to overcome the difficulty in word segmentation of clinical texts in Chinese EMRs, a dictionary construction method is proposed based on the idea of maximal clique, and 730 medical professional terms related to lung diseases are identified. Then, the ProbSparse self‐attention mechanism and self‐attention distilling operation in Informer are used to improve the Bidirectional Encoder Representations from Transformer (BERT) to realize the representation of long clinical texts with lower time complexity and memory consumption. Finally, the convolutional neural network with an attention mechanism is employed to process the representation results to realize the interpretable prediction of lung cancer. This method is applied to the lung cancer diagnosis of inpatients in a tertiary hospital in Hunan Province, obtaining excellent results of about 0.9 for area under the receiver operating characteristic curve (AUROC) and area under the precision‐recall curve (AUPRC). In addition, the results of the comparative analysis with existing dictionaries, word embedding methods and diagnostic methods further confirm the superiority of the proposed method. Specifically, the proposed method improves the precision by at least 6%, the recall by at least 2.6%, the F1 score by at least 5.2%, AUROC by at least 7.3% and AUPRC by at least 7.7% compared with all these state‐of‐the‐art methods.

Abstract With the excessive consumption and pollution of water resources, the sustainable development of water resources poses a serious threat currently. How to perceive and prevent the degradation of water resource in advance is an urgent problem. The water resource carrying capacity (WRCC) is a significant indicator to reflect the condition of water resources in a region. Resounding to these circumstances, our research establishes a decision support framework to solve WRCC risk evaluation issues. First, hesitant probabilistic fuzzy sets (HPFSs) are selected as a representation for the evaluation information in the expert group. Aimed at existing studies of HPFSs, some limitations are overcome involving the distance and comparison rule. Secondly, a TODIM‐PROMETHEE II based multi‐criteria group decision making (MCGDM) method is developed to overcome the inherent restrictions of PROMETHEE II method and make it suitable for a practical decision‐making condition with bounded rationality. Subsequently, a case study is utilized to demonstrate the feasibility of our newly proposed decision support framework, followed by a sensitivity analysis and a comparison analysis. The outcome indicates that the framework has an excellent performance to solve this kind of MCGDM issues.

Abstract In the multi‐criteria group decision‐making (MCGDM) problems with great uncertainty, making full use of participants' evaluation information could help improve the accuracy and reliability of decision results. Probabilistic linguistic term set (PLTS) is an effective tool to represent qualitative data and can fully express the hesitation and preference of decision makers. Therefore, this paper aims to propose an MCGDM method based on PLTSs. In the proposed method, the projection of PLTSs is explored to measure the distance and angle differences between two objects, and Bayesian best–worst method (Bayesian BWM) is used to determine the aggregated final weights of criteria. Besides, the elimination and choice translating reality III (ELECTRE III) method combined with distillation algorithm deals with the projection of PLTSs to obtain the alternatives' ranking of each decision maker. Then, the weighted convex median voting rule is developed to integrate the rankings results regarding all decision makers, which can solve the conflict of ranking results among experts and ensure that the comprehensive ranking results are reasonable and practical. Finally, a case study of health‐care waste management is designed and comparative analyses are implemented to show the effectiveness and advantages of the proposed method.

Decision makers (DMs) have different cognitive levels in practical experience, information reserve, and thinking ability. Thus, decision information is often not completely reliable. As a tool that can effectively represent information reliability, Z-number has been studied by many scholars in recent years. Current research on Z-number assumes that differences in various parts of a Z-number can complement one another. However, in many cases, the preference of DMs for each part is difficult to determine, or DMs believe that the differences in various parts cannot be complementary. Therefore, to solve such decision problems, this paper attempts to extend the traditional MABAC method to the Z-information environment by introducing the directed distance and regret theory. The proposed method simultaneously considers the randomness and fuzziness of Z-number. An example about regional circular economy development program selection is provided to illustrate the feasibility of the proposed method. Results show that the proposed method can solve complex decision problems rationally and effectively, and it has broad application prospects.

Abstract Probabilistic linguistic term sets (PLTSs) are an important expression for hesitant linguistic preference information under group decision‐making circumstances. This study investigates problems of multicriteria group decision making (MCGDM) with PLTSs. A novel and rational comparison method is first proposed, and two distance measures for PLTSs are defined. The weight of each criterion is then obtained via maximum deviation method. Subsequently, extended Techniques for Order Preference by Similarity to Ideal Solution (TOPSIS) ‐ VIseKriterijumska Optimizacija I Kompromisno Resenje a Serbian name (VIKOR) and TODIM (an acronym in Portuguese of interactive and multiple attribute decision making) methods are developed as decision support models to handle MCGDM problems. An illustrative example is also analysed to demonstrate the rationality and feasibility of the proposed methods.