The core difficulty the researchers set out to solve lies in a family of techniques known as Multi-Attribute Decision-Making, or MADM. These methods are designed to rank alternatives—here, scenic spots—against a set of weighted criteria. They work elegantly when the number of attributes is small and the data are homogeneous. But modern tourism evaluation data, harvested at scale from platforms such as Ctrip, violate both assumptions. Hundreds of attributes may be involved, many of them correlated, some redundant, and some outright contradictory in how they rank the same destinations. Feeding such high-dimensional data directly into a decision model can distort results, because redundant or conflicting attributes exert an outsized influence on the final score.

The team’s solution unfolds in three stages. The first stage is a consistency-driven dynamic clustering algorithm. Instead of clustering the scenic spots themselves, the method clusters the evaluation attributes, grouping together those criteria that produce similar ranking patterns across the destinations being compared. In other words, if two attributes—say, visitor ratings of cleanliness and ratings of facility maintenance—consistently order the scenic spots in nearly the same way, they are placed in the same cluster. The word dynamic is important: the algorithm adjusts cluster assignments as it processes the data rather than fixing them in advance, and it uses measures of rank consistency to decide when attributes genuinely belong together. This step reduces the dimensionality of the problem in a principled way, without discarding any information outright.

The second stage applies a well-established decision tool within each cluster: TOPSIS, the Technique for Order Preference by Similarity to Ideal Solution. TOPSIS, introduced in the operations research literature in the 1980s and refined through the 1990s, ranks alternatives by measuring how close each one is to an ideal positive solution and how far it is from a negative ideal solution. Within a single cluster of similar attributes, this is a reasonable and well-conditioned calculation, because the attributes behave coherently. Running TOPSIS separately inside each cluster yields a set of local rankings—one ordered list of scenic spots per cluster—each reflecting the perspective of a coherent family of evaluation criteria.

The third stage is where the method borrows from an unexpected field: voting theory. The researchers use the Borda count, an aggregation procedure devised in the eighteenth century by Jean-Charles de Borda, to fuse the local rankings into a single global ranking. Under the Borda count, each local ranking awards points to each scenic spot according to its position, and the spot with the greatest accumulated total across all clusters wins the top place. This is the same arithmetic that determines winners in some ranked-ballot elections, and its appeal here is the same as in politics: it aggregates many partial, imperfect orderings into one socially robust verdict while dampening the influence of any single anomalous cluster. The authors note that prior theoretical work has argued the Borda count is in an important sense the optimal ranking method, which makes it a natural choice for fusion.

The pipeline was tested on evaluation data for China’s 5A scenic spots, collected from publicly accessible tourism review information on the Ctrip platform. The 5A designation is the pinnacle of China’s tourist rating system, so the destinations involved are already elite; the question was whether a data-driven ranking built from raw, high-dimensional visitor evaluations could reproduce and refine the official assessments. The empirical results showed a high degree of consistency between the method’s output rankings and existing tourism ratings, suggesting that the clustering-and-fusion pipeline captures the same underlying quality signals that expert evaluators respond to, but derives them transparently from visitor data rather than from a fixed administrative rubric.

Why not simply cluster the destinations instead of the attributes? The authors’ design choice reflects the structure of the problem. Tourists evaluate the same scenic spot along many dimensions, and the statistical relationships among those dimensions—how they correlate, which ones move together across destinations—carry the real information. By grouping attributes with similar ranking behaviour, the method effectively discovers latent evaluation perspectives: bundles of criteria that visitors treat as a single experiential quality. Each cluster then functions as a coherent judge, and the Borda count lets these judges vote. This architecture mirrors recent work on large-scale group decision-making, in which clustering and consensus techniques are used to manage hundreds or thousands of participants, and it connects to a growing literature on ranking fusion for massive online review datasets.

The technical payoff is robustness against heterogeneity. Real-world evaluation data are rarely generated by a single consistent process: different visitors weight different concerns, different attributes suffer different amounts of noise, and some criteria may be systematically inflated or deflated on particular platforms. A monolithic MADM model must average all of these effects together, which can wash out genuine distinctions between destinations. The cluster-then-rank-then-fuse design isolates each coherent evaluation channel, ranks destinations within that channel, and only then combines the verdicts. Conflicting attributes are no longer forced into a single weighted sum; they are allowed to speak in their own cluster and then reconciled at the voting stage. The authors position the framework as particularly valuable for scenarios involving high-dimensional and heterogeneous evaluation data, where conventional approaches struggle.

The practical implications extend beyond holiday planning. Tourism management authorities could use such a framework to monitor how the data-driven ranking of attractions evolves as visitor sentiment shifts, identifying destinations whose reputations are improving or deteriorating and tracing those changes back to specific clusters of attributes. Travel platforms could integrate the method into recommendation engines that must reconcile millions of reviews across many criteria. And the underlying algorithmic idea—cluster the criteria, rank within clusters, fuse by voting—is domain-agnostic, potentially transferable to any multi-attribute decision problem drowning in high-dimensional data, from product selection to service quality assessment.

The study also offers a candid account of its data foundations. The datasets were drawn from publicly accessible tourism evaluation information on Ctrip, and the full raw dataset cannot be redistributed because of third-party platform terms of use and intellectual-property restrictions. To support reproducibility, the published paper documents the data source, structure, preprocessing, variable construction, clustering procedure, parameter settings and experimental design in detail, with derived statistical information and non-restricted processed data available from the corresponding author on reasonable request. The research was supported by the Hunan Provincial Natural Science Foundation of China and the Scientific Research Fund of the Hunan Provincial Education Department. As online reviews continue to swell into one of the largest bodies of human judgement ever collected, methods like this one—part statistics, part social choice theory—point toward a future in which the wisdom of crowds is not merely accumulated, but properly counted.

Subject of Research: A dynamic clustering and ranking fusion method for multi-attribute evaluation of 5A scenic spots using tourist review data

Article Title: Dynamic clustering and ranking fusion for tourism recommendations: A data-driven approach to 5A scenic spot evaluation

Article References: Li, W., Xiao, Z., Yu, B., & Xu, Z. (2026). Dynamic clustering and ranking fusion for tourism recommendations: A data-driven approach to 5A scenic spot evaluation. Applied Intelligence, 56(15), Article 433. https://doi.org/10.1007/s10489-026-07488-4

Image Credits: AI Generated

DOI: 10.1007/s10489-026-07488-4

Keywords: tourism recommendation, dynamic clustering, ranking fusion, TOPSIS, Borda count, multi-attribute decision-making, high-dimensional data, 5A scenic spots, online reviews, Ctrip, consistency-driven clustering, Applied Intelligence

To understand why this matters, it helps to trace the lineage of the underlying mathematics. Fuzzy sets, introduced by Lotfi Zadeh in 1965, replaced the rigid yes-or-no logic of classical set theory with degrees of membership, allowing an element to belong to a set partially. Hesitant fuzzy sets, proposed by Vicenç Torra in 2010, went a step further by permitting several possible membership values for a single element, capturing the situation where a panel of experts offers conflicting assessments rather than a single consensus figure. Dual hesitant fuzzy sets then added a second dimension: alongside multiple possible membership degrees, they allow multiple possible non-membership degrees, so an evaluator can simultaneously express hesitation about how well an option performs and hesitation about how badly it fails.

The new paper builds on two prominent extensions of that dual structure. Probabilistic dual hesitant fuzzy sets attach probability weights to each possible membership and non-membership value, acknowledging that some expert opinions are more likely or better supported than others. Interval-valued dual hesitant fuzzy sets, by contrast, replace single numbers with intervals, which is useful when an assessor can only bound a judgment rather than pin it down. Each extension captures something the other misses. Probabilities encode the relative credibility of individual assessments; intervals encode the inherent imprecision of the assessments themselves. Yan’s contribution is the probabilistic interval-valued dual hesitant fuzzy set, or PIVDHFS, which fuses both properties into a single representation. In a PIVDHFS, each element carries a set of interval-valued membership degrees and a set of interval-valued non-membership degrees, and each of those interval values is associated with a probability. The result is a data structure that can model disagreement among experts, imprecision within each expert’s judgment, and the unequal reliability of different judgments, all simultaneously.

Representing information this way is only useful if you can compute with it, and that requires a way to measure how different two PIVDHFSs are. Distance measures are the workhorses of fuzzy decision theory: they underpin ranking methods, clustering, pattern recognition, and medical diagnosis applications. The paper defines the operational laws of PIVDHFSs and then derives a family of distance formulas. The probabilistic interval-valued dual hesitant fuzzy weighted distance treats each attribute with a fixed importance weight. The ordered weighted distance instead assigns weights to positions in an ordered sequence, so that the largest deviations can be emphasized or downplayed depending on the decision-maker’s risk posture. The hybrid weighted distance combines both ideas, weighting attributes and positions at the same time, and the generalized hybrid weighted distance adds a tunable parameter that lets the analyst adjust how aggressively large differences are penalized. Together these measures give decision-makers a flexible toolkit for comparing alternatives described in this rich uncertain format.

A second, often overlooked problem in multi-attribute decision-making is determining how much each criterion should matter in the first place. When attribute weights are unknown, subjective assignment invites bias, so objective weighting methods infer importance directly from the data. The study enhances one such method, CRITIC, which stands for criteria importance through intercriteria correlation. The original CRITIC, introduced by Diakoulaki and colleagues in 1995, assigns higher weights to criteria that carry more information, measured by standard deviation, and that are less redundant with other criteria, measured by correlation. Yan’s version incorporates the Spearman correlation coefficient, a rank-based measure of association, within the PIVDHFS context. Rank correlation is more robust than the Pearson coefficient when the underlying data are ordinal or non-linear, which is precisely the character of hesitant fuzzy assessments. The enhanced CRITIC therefore extracts objective weights from the structure of the probabilistic interval-valued data itself, reducing the influence of arbitrary human judgment on the weighting stage.

With distances and weights in hand, the framework turns to ranking. TOPSIS, the technique for order of preference by similarity to ideal solution, is one of the most widely used methods in operations research. Its logic is intuitive: the best alternative is the one that is simultaneously closest to a hypothetical positive ideal solution and farthest from a negative ideal one. The paper extends TOPSIS to operate directly on PIVDHFSs, using the newly developed distance measures to compute each alternative’s separation from the ideal and anti-ideal points. The extended PIVDHF-CRITIC-TOPSIS pipeline thus proceeds in three stages: encode expert evaluations as PIVDHFSs, derive objective attribute weights with the Spearman-enhanced CRITIC method, and rank alternatives by their relative closeness computed through the new distance family.

The framework was put to the test on a problem with genuine industrial stakes: digital supplier selection for a new energy vehicle manufacturer in China. Digital suppliers provide the software platforms, data services, and connected-technology capabilities that modern electric vehicle production increasingly depends on, and choosing among them involves criteria that are hard to quantify precisely, such as technological maturity, data security posture, service responsiveness, and long-term partnership potential. Evaluations from multiple experts naturally conflict, and confidence in each assessment varies. The PIVDHFS format captured this layered uncertainty, and the extended framework produced a complete ranking of candidate suppliers. The author validated the result in two ways. A sensitivity analysis varied key parameters, including the generalized distance parameter and the derived attribute weights, to check whether the ranking was an artifact of particular settings. A comparative study then applied alternative decision methods to the same data to confirm that the proposed approach yields consistent and scientifically rational outcomes.

What makes this work notable in the broader landscape of decision science is its position at the confluence of several research threads that had previously advanced separately. Probabilistic dual hesitant fuzzy sets have been applied to risk evaluation, sustainable supplier selection, and rainfall analysis, while interval-valued dual hesitant fuzzy sets have supported emergency response evaluation and aggregation-based decision models. Distance measures for dual hesitant structures have found uses ranging from image segmentation to medical diagnosis. By unifying the probabilistic and interval-valued extensions and wiring them into an objective weighting scheme and a classical ranking method, the new framework offers a template that other researchers can adapt. The same architecture could, in principle, support green supplier selection, renewable energy assessment, medical diagnosis, or any domain where expert judgment is hesitant, imprecise, and unevenly reliable.

The practical implications extend to the fast-growing field of digital supply chain management. Recent literature on supplier selection in digital manufacturing has emphasized resilience, cyber risk, and data-driven simulation, and machine learning approaches have been proposed for modeling supplier performance under disruption. Yet the front end of those pipelines, the structured elicitation of expert judgment about candidates, still relies heavily on decision frameworks that assume crisp numbers or single-layer fuzziness. A representation that preserves probabilities, intervals, and dual hesitation through every stage of weighting and ranking could feed cleaner, more honest inputs into downstream analytics. For industries such as electric vehicles, where supply chains span continents and technology cycles are measured in months, the difference between a robust ranking and a fragile one can translate into millions in procurement outcomes.

There are, of course, limits worth noting. The framework’s sophistication comes at the cost of computational and elicitation overhead: asking experts to supply interval-valued memberships, non-memberships, and probabilities for every attribute of every alternative is demanding, and the quality of the output depends on the quality of those inputs. The study itself reports validation through a single real-world case, supplemented by sensitivity and comparative analyses, which is a solid but not exhaustive evidence base. Still, the paper represents a careful piece of mathematical engineering, defining operational laws, deriving a graded family of distance measures, enhancing an objective weighting method with a rank correlation coefficient, and demonstrating the whole pipeline on a decision that manufacturers actually face. As uncertain, expert-mediated judgments continue to pervade high-stakes industrial and environmental choices, tools that quantify hesitation rather than papering over it are likely to become standard equipment in the decision scientist’s toolkit.

Subject of Research: Multi-attribute decision-making using probabilistic interval-valued dual hesitant fuzzy sets with distance measures and an extended TOPSIS framework

Article Title: Multi-attribute decision-making based on probabilistic interval-valued dual hesitant fuzzy sets and extended TOPSIS

Article References: Yan, L. (2026). Multi-attribute decision-making based on probabilistic interval-valued dual hesitant fuzzy sets and extended TOPSIS. International Journal of Machine Learning and Cybernetics, 17(10), Article 486. https://doi.org/10.1007/s13042-026-03294-z

Image Credits: AI Generated

DOI: 10.1007/s13042-026-03294-z

Keywords: fuzzy sets, multi-attribute decision-making, TOPSIS, CRITIC method, distance measures, dual hesitant fuzzy sets, supplier selection, uncertainty modeling, operations research, new energy vehicles, Spearman correlation, decision science