Expert SystemsPub Date : 2026-10-21DOI: 10.1111/exsy.70466
Alberto Falcone, Alessandro Sorrenti, Alfredo Garro
{"title":"Combining Model-Driven and Data-Driven Approaches for Predictive Maintenance","authors":"Alberto Falcone, Alessandro Sorrenti, Alfredo Garro","doi":"10.1111/exsy.70466","DOIUrl":"https://doi.org/10.1111/exsy.70466","url":null,"abstract":"<div>\u0000 \u0000 <p>Predictive maintenance is increasingly supported by data-driven methods, which can effectively learn from operational data but may overlook important structural and behavioural knowledge about the monitored system. This paper presents H-PdM, a hybrid method that combines information derived from system models with machine-learning-based analysis of operational data. The two sources provide complementary information that is combined to support fault diagnosis and maintenance decisions. The method is evaluated on a triplex pump digital twin considering different fault mechanisms and their combinations. Several machine learning classifiers are compared, and the final configuration is evaluated under non-ideal operating conditions, including measurement noise, sensor bias, drift and missing data. The results show that the hybrid approach is more robust than either the data-driven or model-driven component used alone. An explainability analysis also shows how system-derived rules and data-driven features contribute differently depending on the detected fault. Overall, the results support the use of model and data-driven information together to achieve more reliable and interpretable predictive maintenance.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"43 12","pages":""},"PeriodicalIF":3.4,"publicationDate":"2026-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149258499","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Expert SystemsPub Date : 2026-10-21DOI: 10.1111/exsy.70420
Qinxue Li, Yonggui Liu, Xiaofen Yang, Xuhuan Xie, Rui Li
{"title":"Generative Adversarial Imputation With Relation-Aware Self-Attention for Robust Data Recovery","authors":"Qinxue Li, Yonggui Liu, Xiaofen Yang, Xuhuan Xie, Rui Li","doi":"10.1111/exsy.70420","DOIUrl":"https://doi.org/10.1111/exsy.70420","url":null,"abstract":"<div>\u0000 \u0000 <p>Owing to the highly networked and cyber-enabled nature of cyber-physical systems (CPSs), they are particularly susceptible to various cyber-attacks and communication anomalies. Such vulnerabilities can lead to critical data loss, which in turn impairs subsequent tasks such as state estimation. Moreover, the above issues may render the prior information of the physical topology unreliable or unavailable in real time. To address these challenges, an improved data-driven generative adversarial imputation network (GAIN) is proposed to recover CPS data subjected to denial of service (DoS) attacks or missing-data scenarios. Building upon the original GAIN framework, we improve the generator by incorporating a RASA mechanism, which explicitly models both explicit and implicit token relationships to capture complex data interdependencies. Furthermore, this mechanism enhances the traceability and interpretability of attention allocation, making the proposed approach particularly well-suited for time-series data processing. Furthermore, the joint optimization training framework is introduced, through which the synergistic effect of the masked imputation task (MIT) and the observed reconstruction task (ORT), combined with the parallel processing advantage of the self-attention mechanism, collectively enhances the model's imputation performance, robustness and generalization capability, thereby further strengthening the data recovery capacity of the proposed GAIN with RASA (RASA-GAIN). Experimental results on multiple datasets, including secure water treatment (SWaT), demonstrate that the proposed RASA-GAIN achieves an average reduction in RMSE of 53.45% on Dataset A of SWaT, 8.59% on the Letter dataset and 23.25% on the News dataset, compared to the original GAIN. Furthermore, to assess its proficiency in time-series data recovery, we deploy the proposed RASA-GAIN to mitigate missing-data challenges in power systems facing both DoS attacks and random packet losses. The results indicate that RASA-GAIN outperforms five mainstream imputation models in most scenarios, delivering superior data recovery performance across missing rates ranging from 14.594% to 57.229%.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"43 12","pages":""},"PeriodicalIF":3.4,"publicationDate":"2026-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149258291","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Expert SystemsPub Date : 2026-10-21DOI: 10.1111/exsy.70462
Eytan Ellenberg
{"title":"Hierarchical Attribution Networks: Composing Attribution-Magnitude Flows Across Predictive Modules","authors":"Eytan Ellenberg","doi":"10.1111/exsy.70462","DOIUrl":"https://doi.org/10.1111/exsy.70462","url":null,"abstract":"<div>\u0000 \u0000 <p>Complex prediction systems often contain multiple fitted modules arranged as a directed acyclic graph (DAG), yet standard feature attribution typically explains one module at a time. We present Hierarchical Attribution Networks (HAN), a post hoc composition rule for tracing non-negative attribution magnitude through a predictive DAG. For each observation and module, signed local feature attributions are retained, while their absolute magnitudes are normalized into routing coefficients. Terminal magnitude is the absolute deviation of the final model output from its explainer baseline. Flow is propagated upstream, stops at unresolved nodes whose attribution denominator is numerically negligible, and is aggregated over complete paths. This construction conserves routed magnitude by definition; it is not a new Shapley value, a causal decomposition or a claim of global efficiency. We evaluate the rule in a five-module cascade using 2408 adults from the 2017–2018 National Health and Nutrition Examination Survey. Repeated nested out-of-fold prediction yielded weak but permutation-detectable intermediate models (mean <span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 <msup>\u0000 <mi>R</mi>\u0000 <mn>2</mn>\u0000 </msup>\u0000 <mo>=</mo>\u0000 <mn>0.011</mn>\u0000 </mrow>\u0000 <annotation>$$ {R}^2=0.011 $$</annotation>\u0000 </semantics></math>, <span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 <mn>0.040</mn>\u0000 </mrow>\u0000 <annotation>$$ 0.040 $$</annotation>\u0000 </semantics></math>, and <span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 <mn>0.088</mn>\u0000 </mrow>\u0000 <annotation>$$ 0.088 $$</annotation>\u0000 </semantics></math>) and a final diabetes classifier with mean area under the receiver operating characteristic curve of 0.744. The direct age-to-risk path was dominant for 91.0% of participants. Synthetic experiments exposed signed-ratio instability under cancellation and quantified sensitivity to collinearity and graph misspecification. Participant-level bootstrap refitting, a KernelSHAP comparison and heterogeneous-model analysis assessed stability and technical portability. HAN provides an auditable descriptive view of how local attribution magnitude is routed across modular predictors, conditional on the supplied graph, models and explainers.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"43 12","pages":""},"PeriodicalIF":3.4,"publicationDate":"2026-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149258290","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Expert SystemsPub Date : 2026-10-20DOI: 10.1111/exsy.70459
Mirza Aqib Ali, Usama Athar, Adnan Maqsood
{"title":"SITL Validation of Federated Reinforcement Learning for Priority-Based Dynamic Task Allocation in UAV Swarms","authors":"Mirza Aqib Ali, Usama Athar, Adnan Maqsood","doi":"10.1111/exsy.70459","DOIUrl":"https://doi.org/10.1111/exsy.70459","url":null,"abstract":"<div>\u0000 \u0000 <p>The deployment of reinforcement learning (RL) controllers for unmanned aerial vehicle (UAV) swarm navigation has been extensively studied in abstract simulation environments. However, a critical validation gap persists between algorithmic development and operationally representative testing under realistic flight dynamics. This work presents a Software-In-The-Loop (SITL) validation of a federated reinforcement learning (FRL) framework for multi-UAV swarm navigation, implemented using ArduPilot flight controller stack integrated with the Gazebo Harmonic physics simulator and orchestrated through a Python-based middleware architecture that connects asynchronous RL policy inference with real-time MAVLink communication. Federated Proximal Policy Optimization (PPO) framework is tested across four progressively complex operational scenarios: (i) single-agent point-to-point navigation (96.0% success rate, path efficiency <span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 <mi>ρ</mi>\u0000 <mo>=</mo>\u0000 <mn>0.847</mn>\u0000 </mrow>\u0000 <annotation>$$ rho =0.847 $$</annotation>\u0000 </semantics></math>), (ii) five-UAV leader-follower formation flight (94.0% success rate, zero inter-agent collisions, mean formation error <span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 <mn>1.24</mn>\u0000 <mo>±</mo>\u0000 <mn>0.38</mn>\u0000 </mrow>\u0000 <annotation>$$ 1.24pm 0.38 $$</annotation>\u0000 </semantics></math> m), (iii) dynamic mid-mission target reassignment requiring 143° heading correction (94.0% success rate, mean redirection time <span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 <mn>4.2</mn>\u0000 <mo>±</mo>\u0000 <mn>1.3</mn>\u0000 </mrow>\u0000 <annotation>$$ 4.2pm 1.3 $$</annotation>\u0000 </semantics></math> s) and (iv) priority-based dynamic task allocation, wherein a secondary target with assigned priority emerges mid-mission, triggering autonomous priority-proportional swarm partitioning (90.0% dual-target success rate, mean reallocation time <span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 <mn>5.3</mn>\u0000 <mo>±</mo>\u0000 <mn>1.6</mn>\u0000 </mrow>\u0000 <annotation>$$ 5.3pm 1.6 $$</annotation>\u0000 </semantics></math> s). Across all scenarios, the framework maintained a near-zero inter-agent collision rate (0.0% in Scenarios 2 and 4; a single incident in Scenario 3) and total federation communication overhead at or below 1.50 MB per episode, with sub-group policy divergence (<span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 <msub>\u0000 <mi>D</mi>\u0000 <mi>KL</mi>\u0000 </msub>\u0000 <mo>=</mo>\u0000 ","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"43 12","pages":""},"PeriodicalIF":3.4,"publicationDate":"2026-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149258070","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Expert SystemsPub Date : 2026-10-18DOI: 10.1111/exsy.70460
Lorenzo Ricciardi Celsi, Junxin Chen
{"title":"Ethics and Governance in the Age of Generative Artificial Intelligence: A Lifecycle Engineering Perspective","authors":"Lorenzo Ricciardi Celsi, Junxin Chen","doi":"10.1111/exsy.70460","DOIUrl":"https://doi.org/10.1111/exsy.70460","url":null,"abstract":"<div>\u0000 \u0000 <p>Trustworthy Generative Artificial Intelligence (GenAI) is often treated as a model property addressed through better training data, alignment, evaluation or guardrails. We argue instead that trustworthiness is a lifecycle governance capability that must be engineered, evidenced, monitored and adapted across data acquisition, training and alignment, safe inference, public deployment and post-deployment use. We develop a lifecycle engineering framework integrating four governance layers – technical, organizational, human-centred and regulatory – and translate ethical principles and major governance instruments into stage-specific controls, responsibilities, evidence artefacts, metrics and review triggers. The framework is constructed through a structured conceptual synthesis of peer-reviewed literature, NIST AI RMF, ISO/IEC 42001, ISO/IEC 23894, ISO/IEC 42005, ALTAI, the EU AI Act and complementary security and agentic-governance guidance. A source register, seven-field extraction schema, explicit mapping rules, worked extraction example and weighting sensitivity analysis make the construction logic inspectable. A public-facing retrieval-augmented generation case and a contrastive agentic-system stress test illustrate operationalization through measurable indicators, responsible actors, review triggers and capability boundaries without being presented as independent empirical validation. We further address hallucination, prompt injection, privacy leakage, overreliance, unsafe tool use, memory, planners and multi-agent workflows. The perspective positions trustworthy GenAI as an evidence-driven, adaptive socio-technical capability rather than a one-time model or compliance claim.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"43 12","pages":""},"PeriodicalIF":3.4,"publicationDate":"2026-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149258256","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Expert SystemsPub Date : 2026-10-16DOI: 10.1111/exsy.70456
Rana Husni Al Mahmoud, Hanan Saleet
{"title":"Benchmarking Arabic Transformer-Based Language Model Embeddings for Document Clustering: A Comparative Study","authors":"Rana Husni Al Mahmoud, Hanan Saleet","doi":"10.1111/exsy.70456","DOIUrl":"https://doi.org/10.1111/exsy.70456","url":null,"abstract":"<div>\u0000 \u0000 <p>Recent advances in transformer-based language models have substantially improved semantic text representations for Arabic natural language processing. However, their effectiveness for unsupervised document clustering has not been comprehensively investigated across different clustering algorithms, preprocessing strategies and document representations. This study presents a comparative evaluation of six transformer-based embedding models, namely AraGPT2, arabertv0.2, CAMeLBERT, AraELECTRA, AraBERTv2, and Multilingual-MiniLM, together with a TF-IDF lexical baseline, for Arabic document clustering. The evaluation was conducted on five Arabic datasets spanning both general-domain and domain-specific corpora using four clustering algorithms: <i>K</i>-Means++, Spherical <i>K</i>-Means, Hierarchical Agglomerative Clustering (HAC) and Spectral Co-Clustering. To investigate the influence of text representation, six preprocessing configurations were considered by varying stemming and document-length reduction strategies. Clustering performance was assessed using Normalised Mutual Information (NMI), Accuracy, Purity, Adjusted Rand Index (ARI), Macro <i>F</i>1-score, V-Measure and Entropy. Furthermore, the experimental results were statistically validated using the Friedman test, average-rank analysis, and Holm-corrected Wilcoxon signed-rank post hoc comparisons. The experimental results demonstrate that preserving the original lexical and morphological structure of Arabic text generally yields superior clustering performance for contextual embedding models compared with light stemming. The influence of document length was found to be dataset- and representation-dependent, with full-document representations achieving the strongest overall performance, while shorter representations frequently remained competitive. TF-IDF provided a strong lexical baseline, particularly for general-domain datasets. Among the evaluated embedding models, AraBERTv2 achieved the best overall average rank, closely followed by arabertv0.2, and the post hoc statistical analysis revealed no significant difference between the two models. CAMeLBERT consistently exhibited stable and competitive performance across all datasets, whereas AraELECTRA produced the weakest clustering results. From the clustering perspective, Spherical <i>K</i>-Means demonstrated the most consistent overall performance, while HAC achieved the highest clustering quality in several of the best-performing embedding configurations. Overall, this study provides a comprehensive benchmark for transformer-based Arabic document clustering and offers practical guidance for selecting embedding models, preprocessing strategies, document representations, and clustering algorithms for unsupervised Arabic text analysis.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"43 12","pages":""},"PeriodicalIF":3.4,"publicationDate":"2026-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149205621","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Interpretable Dynamic Graph Learning via Dual-Granularity Large Language Model Alignment","authors":"Yingjie Xie, Wenjun Wang, Tianpeng Li, Xiaoming Li, Lin Yuanbo Wu","doi":"10.1111/exsy.70454","DOIUrl":"https://doi.org/10.1111/exsy.70454","url":null,"abstract":"<div>\u0000 \u0000 <p>Dynamic graph-level embedding aims to learn whole-graph representations from temporal snapshots. Existing methods predominantly follow a single-vector paradigm that encodes each snapshot as a continuous latent vector, suffering from two key limitations: (1) Stage–Snapshot Entanglement, where within-stage fluctuations and cross-stage transitions are conflated without explicit granularity distinction; and (2) Implicit Structural Semantics, where meaningful structural properties remain buried in latent coordinates and are never exposed as interpretable, language-compatible units. To address these issues, we propose DyGLLM, an LLM-compatible framework that constructs and aligns dual-granularity structural representations for dynamic graph understanding. Specifically, DyGLLM maps temporally coherent evolution regimes into selective structural tokens <span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 <mfenced>\u0000 <msub>\u0000 <mi>SEG</mi>\u0000 <mi>k</mi>\u0000 </msub>\u0000 </mfenced>\u0000 </mrow>\u0000 <annotation>$$ leftlangle {mathrm{SEG}}_krightrangle $$</annotation>\u0000 </semantics></math>, providing explicit stage-level anchors in the LLM token space. It further constructs graph fingerprints to verbalize each snapshot's prototype deviation, temporal variation and topological profile into language-readable structural descriptions. A hybrid structure QA alignment mechanism with LoRA fine-tuning is then introduced to align structural tokens and graph fingerprints with the LLM's representation space, enabling the model to interpret and reason over explicit structural contexts. Extensive experiments on five dynamic graph benchmarks demonstrate that DyGLLM achieves consistently competitive performance across graph similarity ranking, temporal anomaly detection and change point detection, validating the effectiveness of the proposed dual-granularity structural representation.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"43 12","pages":""},"PeriodicalIF":3.4,"publicationDate":"2026-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149205624","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Expert SystemsPub Date : 2026-10-16DOI: 10.1111/exsy.70457
Yinghe Wang, Yuankun Xu, Chulei Xiang
{"title":"Heterophily-Aware Motif-Guided Contrastive Learning for Robust Social Recommendation","authors":"Yinghe Wang, Yuankun Xu, Chulei Xiang","doi":"10.1111/exsy.70457","DOIUrl":"https://doi.org/10.1111/exsy.70457","url":null,"abstract":"<div>\u0000 \u0000 <p>The homophily principle serves as the core foundation of graph neural network (GNN) based social recommendation, which assumes socially connected users share similar preferences and enables neighbourhood aggregation for user representation learning. Most existing methods directly aggregate information from original social graphs, but this strategy is severely suboptimal due to the prevalent low preference-aware homophily in real-world social networks. We further reveal that nonpreference-driven social connections introduce massive noise into GNN aggregation, leading to representation degradation, over-smoothing and even performance decline. Existing efforts mostly focus on individual subproblems of social denoising, high-order modelling or representation enhancement and fail to provide a systematic solution for these cascading challenges. In this paper, we propose H2Rec, a heterophily-aware motif-guided contrastive learning framework for low-homophily social recommendation. It integrates a dual-branch social aggregation module, a homophily-enhanced contrastive alignment objective and a user-adaptive fusion mechanism to suppress noise and learn robust user representations end-to-end. Extensive experiments on real-world benchmarks show that H2Rec significantly outperforms state-of-the-art methods on mainstream evaluation metrics.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"43 12","pages":""},"PeriodicalIF":3.4,"publicationDate":"2026-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149205622","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Rethink-RAG: Dynamic Retrieval Decision and Cooperative Verification for Reliable Multi-Hop Question Answering","authors":"Zhiming Li, Chao Gao, ZiXuan Wang, Xincheng Wang, YanChang Lu, Dianlong You","doi":"10.1111/exsy.70455","DOIUrl":"https://doi.org/10.1111/exsy.70455","url":null,"abstract":"<div>\u0000 \u0000 <p>Reliable multi-hop question answering is important for knowledge-intensive expert systems, where a wrong intermediate fact can propagate into an incorrect final decision. Existing Retrieval-Augmented Generation (RAG) methods often retrieve for every sub-question, even when the language model can answer a step from parametric knowledge, and this indiscriminate retrieval can introduce irrelevant or contradictory evidence. We propose Rethink-RAG, a training-free framework that combines dynamic retrieval decision with a cooperative answer verification pipeline. The dynamic retrieval decision module generates diverse rewrites of each sub-question and applies majority voting to decide whether external retrieval is required. The cooperative verification pipeline then filters retrieved documents, quantifies support and opposition through token-logit evidence, and arbitrates each candidate sub-answer with an adaptive confidence threshold. Experiments on 2WikiMultihopQA, HotpotQA, and MuSiQueQA show that Rethink-RAG improves answer reliability on knowledge-intensive reasoning tasks, achieving 49.00% ACC on 2WikiMultihopQA and 49.60% ACC on HotpotQA with the Llama3-8B evaluator. Additional revision experiments quantify the latency/API-call trade-off, sensitivity to the decision-boundary factor, and the limitations observed on MuSiQueQA, giving a more complete view of the framework's practical operating range.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"43 12","pages":""},"PeriodicalIF":3.4,"publicationDate":"2026-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149205620","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Expert SystemsPub Date : 2026-10-04DOI: 10.1111/exsy.70443
Wenlong Cui, Weiwei Li, Pingtao Yi
{"title":"Superiority-Probabilistic-Based Products Ranking Method: Integrating Users' Consensus Experiences and Consumers' Individual Preferences","authors":"Wenlong Cui, Weiwei Li, Pingtao Yi","doi":"10.1111/exsy.70443","DOIUrl":"https://doi.org/10.1111/exsy.70443","url":null,"abstract":"<div>\u0000 \u0000 <p>In the information edge, online reviews have emerged as a pivotal factor in consumers' purchasing decisions. In reality, we are accustomed to obtaining rigid product rankings from online reviews. However, online reviews naturally contain varied and sometimes conflicting viewpoints; the rigid ranking results are not completely reasonable. To address this problem, the paper proposes a superiority-probabilistic-based ranking model by integrating both users' consensus experiences and consumers' individual preferences. Firstly, the users' experiences, primarily represented by star ratings and textual reviews, are transformed into the individual evaluation domains (IEDs) through sentiment analysis. Then, the IEDs are further fused into the comprehensive evaluation domain (CED) by highlighting the users' consensus experiences. Subsequently, an individual-preference-oriented indicator weighting method is introduced by referring to the sequential relationship analysis (SRA) method. Based on the above preparations, the superiority-probability-based pairwise comparison matrix (SPM) is calculated through a thorough sampling of CED and the iterative aggregation of it with the indicator weights. The SPM reflects the relative superiority probability among any products, from which the probabilistic ranking of products can be derived. The case study illustrates the validity and practicability of the proposed methods. The research contribution establishes a superiority-probabilistic-based products ranking framework, providing consumers with more detailed purchase recommendations.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"43 11","pages":""},"PeriodicalIF":3.4,"publicationDate":"2026-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149145644","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}