International Journal of Approximate Reasoning最新文献

{"title":"Feasible strategies in three-way conflict analysis with three-valued ratings","authors":"Jing Liu ,&nbsp;Mengjun Hu ,&nbsp;Guangming Lang","doi":"10.1016/j.ijar.2025.109516","DOIUrl":"10.1016/j.ijar.2025.109516","url":null,"abstract":"<div><div>Most existing work on three-way conflict analysis has focused on trisecting agent pairs, agents, or issues. While these trisections lay the groundwork for understanding the nature of conflicts, further actions need to be formulated to address conflict resolution. One of the widely studied approaches is to construct feasible strategies. This paper aims to investigate feasible strategies from two perspectives of consistency and non-consistency. Particularly, we begin with computing the overall rating of a clique of agents based on positive and negative similarity degrees. Afterwards, considering the weights of both agents and issues, we propose weighted consistency and non-consistency measures, which are respectively used to identify the feasible strategies for a clique of agents. Algorithms are developed to identify feasible strategies, <em>L</em>-order feasible strategies, and the corresponding optimal ones. Finally, to demonstrate the practicality, effectiveness, and superiority of the proposed models, we apply them to two commonly used case studies on NBA labor negotiations and development plans for Gansu Province and conduct a sensitivity analysis on parameters and a comparative analysis with existing state-of-the-art conflict analysis approaches. The comparison results demonstrate that our conflict resolution models outperform the conventional approaches by unifying weighted agent-issue evaluation with consistency and non-consistency measures to enable the systematic identification of not only feasible strategies but also optimal solutions.</div></div>","PeriodicalId":13842,"journal":{"name":"International Journal of Approximate Reasoning","volume":"186 ","pages":"Article 109516"},"PeriodicalIF":3.2,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144297784","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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":"Computationally efficient variational-like approximations of possibilistic inferential models","authors":"Leonardo Cella ,&nbsp;Ryan Martin","doi":"10.1016/j.ijar.2025.109506","DOIUrl":"10.1016/j.ijar.2025.109506","url":null,"abstract":"<div><div>Inferential models (IMs) offer provably reliable, data-driven, possibilistic statistical inference. But despite the IM framework's theoretical and foundational advantages, efficient computation is a challenge. This paper presents a simple yet powerful numerical strategy for approximating the IM's possibility contour, or at least its <em>α</em>-cut for a specified <span><math><mi>α</mi><mo>∈</mo><mo>(</mo><mn>0</mn><mo>,</mo><mn>1</mn><mo>)</mo></math></span>. Our proposal starts with the specification of a parametric family that, in a certain sense, approximately covers the credal set associated with the IM's possibility measure. Akin to variational inference, we then propose to tune the parameters of that parametric family so that its <span><math><mn>100</mn><mo>(</mo><mn>1</mn><mo>−</mo><mi>α</mi><mo>)</mo><mtext>%</mtext></math></span> credible set roughly matches the IM contour's <em>α</em>-cut. This parametric <em>α</em>-cut matching strategy implies a full approximation to the IM's possibility contour at a fraction of the computational cost associated with previous strategies.</div></div>","PeriodicalId":13842,"journal":{"name":"International Journal of Approximate Reasoning","volume":"186 ","pages":"Article 109506"},"PeriodicalIF":3.2,"publicationDate":"2025-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144271892","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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":"Conflict management in a distance to prototype-based evidential neural network","authors":"Dănuţ-Vasile Giurgi ,&nbsp;Mihreteab Negash Geletu ,&nbsp;Thomas Josso-Laurain ,&nbsp;Maxime Devanne ,&nbsp;Jean-Philippe Lauffenburger ,&nbsp;Jean Dezert","doi":"10.1016/j.ijar.2025.109508","DOIUrl":"10.1016/j.ijar.2025.109508","url":null,"abstract":"<div><div>Despite advances in integrating reasoning based on belief functions to generalise probabilistic representations, distance-to-prototype-based evidential deep neural networks are still emerging and require further consolidation. Existing studies in segmentation or classification tasks typically perform prior initialisation and do not address or mitigate the potential conflicts that may arise during fusion. This work investigates high-conflict scenarios within an evidential neural network for segmentation in autonomous driving, focusing on the distance-to-prototypes component, where prototypes, derived from feature maps, serve as sources of evidence and may yield contradictory information. Conflict is mitigated through parameter adjustments within the evidential reasoning, enhancing consistency before fusion. This enables more reliable data integration and a valid application of fusion rules and decision-making processes. The proposed rectification is validated on two prototype configurations of a deep evidential lidar-camera cross-fusion architecture, using two distance-based decision strategies and adapted metrics. The impact on the network's predictions is demonstrated through qualitative and quantitative results on road detection tasks with the KITTI dataset.</div></div>","PeriodicalId":13842,"journal":{"name":"International Journal of Approximate Reasoning","volume":"186 ","pages":"Article 109508"},"PeriodicalIF":3.2,"publicationDate":"2025-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144307169","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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":"Constructing intuitionistic neighborhood based on intuitionistic fuzzy sets for three-way clustering","authors":"Yu Xie ,&nbsp;Jilin Yang ,&nbsp;Yiyu Luo ,&nbsp;Xianyong Zhang ,&nbsp;Junfang Luo","doi":"10.1016/j.ijar.2025.109512","DOIUrl":"10.1016/j.ijar.2025.109512","url":null,"abstract":"<div><div>Three-way clustering assigns highly uncertain samples to the boundary domains, effectively addressing the problem of misclassification caused by data uncertainty. In numerical attribute information systems, neighborhood rough sets can effectively capture the indiscernibility relations between objects. However, the conventional neighborhood relation suffers from a one-size-fits-all issue due to the fixed neighborhood radius. To solve these issues, we propose an intuitionistic neighborhood and construct a corresponding three-way clustering model. Specifically, we first capture the dual nature and uncertainty of neighborhood relations through the construction of the intuitionistic neighborhood. Then we construct a three-way clustering model with dual and single evaluation functions based on intuitionistic neighborhoods. Finally, we adaptively obtain the optimal threshold pairs by maximizing the clustering effectiveness index. Experiments conducted on twelve datasets demonstrate that our proposed method outperforms baseline methods, showing superior capability in handling the inherent uncertainty in information systems.</div></div>","PeriodicalId":13842,"journal":{"name":"International Journal of Approximate Reasoning","volume":"186 ","pages":"Article 109512"},"PeriodicalIF":3.2,"publicationDate":"2025-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144307170","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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":"Attribute reduction with pessimistic multigranulation rough sets in relation systems","authors":"Yehai Xie","doi":"10.1016/j.ijar.2025.109515","DOIUrl":"10.1016/j.ijar.2025.109515","url":null,"abstract":"<div><div>Pessimistic multigranulation rough sets (PMGRSs) are an important extension of rough sets and attribute reduction is a significant application of rough set theory. In this paper, we study attribute reduction using PMGRSs in relation systems. Recognizing that the assumptions of reflexivity-symmetry and equivalence of relations are obstacles for application, we redefine the concepts of pessimistic reduction, pessimistic lower approximate distribution (PLAD) reduction, and pessimistic upper approximate distribution (PUAD) reduction based on relations without any restrictions. Furthermore, we design reduction algorithms based on discernibility matrices to identify all pessimistic reducts, PLAD-reducts, and PUAD-reducts. Finally, we conducted comparative experiments on 18 public datasets and the experimental results confirmed the effectiveness of the proposed algorithms.</div></div>","PeriodicalId":13842,"journal":{"name":"International Journal of Approximate Reasoning","volume":"186 ","pages":"Article 109515"},"PeriodicalIF":3.2,"publicationDate":"2025-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144297782","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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":"A minimal base or a direct base? That is the question!","authors":"Jaume Baixeries ,&nbsp;Amedeo Napoli","doi":"10.1016/j.ijar.2025.109509","DOIUrl":"10.1016/j.ijar.2025.109509","url":null,"abstract":"<div><div>In this paper we revisit the problem of computing the closure of a set of attributes given a basis of dependencies or implications. This problem is of main interest in logics, in the relational database model, in lattice theory, and in Formal Concept Analysis as well. A basis of dependencies may have different characteristics, among which being “minimal”, e.g., the DG-Basis, or being “direct”, e.g., the Canonical-Direct Unit Basis and the <em>D</em>-base. Here we propose an extensive and experimental study of the impacts of minimality and directness on the closure algorithms. The results of the experiments performed on real and synthetic datasets are analyzed in depth, and suggest a different and fresh look at computing the closure of a set of attributes w.r.t. a basis of dependencies.</div></div>","PeriodicalId":13842,"journal":{"name":"International Journal of Approximate Reasoning","volume":"186 ","pages":"Article 109509"},"PeriodicalIF":3.2,"publicationDate":"2025-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144271893","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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":"A formal study of a rough set model integrating relational and neighbourhood system approaches","authors":"Md. Aquil Khan,&nbsp;Ranjan","doi":"10.1016/j.ijar.2025.109492","DOIUrl":"10.1016/j.ijar.2025.109492","url":null,"abstract":"<div><div>This article introduces a rough set model that integrates two approximation operators - one induced by a relation and the other by a neighbourhood system - within a unified framework. The proposed model formalizes concept approximation in heterogeneous settings where multiple granulation mechanisms contribute to information processing. The framework effectively combines the strengths of relation-based and neighbourhood-based rough set models by employing conjunctive and disjunctive fusion rules. To provide a rigorous foundation, we develop a logical study of the resulting approximation operators using the modal language with two unary modal operators. We introduce semantics that fuses Kripke and neighbourhood interpretations of modal operators, establish sound and complete deductive systems, and investigate definability properties. This study contributes to both rough set theory and modal logic by offering a formal perspective on the fusion of approximation operators.</div></div>","PeriodicalId":13842,"journal":{"name":"International Journal of Approximate Reasoning","volume":"186 ","pages":"Article 109492"},"PeriodicalIF":3.2,"publicationDate":"2025-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144297783","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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":"Visual hallucination detection in large vision-language models via evidential conflict","authors":"Tao Huang ,&nbsp;Zhekun Liu ,&nbsp;Rui Wang ,&nbsp;Yang Zhang ,&nbsp;Liping Jing","doi":"10.1016/j.ijar.2025.109507","DOIUrl":"10.1016/j.ijar.2025.109507","url":null,"abstract":"<div><div>Despite the remarkable multimodal capabilities of Large Vision-Language Models (LVLMs), discrepancies often occur between visual inputs and textual outputs—a phenomenon we term visual hallucination. This critical reliability gap poses substantial risks in safety-critical Artificial Intelligence (AI) applications, necessitating a comprehensive evaluation benchmark and effective detection methods. Firstly, we observe that existing visual-centric hallucination benchmarks mainly assess LVLMs from a perception perspective, overlooking hallucinations arising from more advanced reasoning capabilities. We develop the Perception-Reasoning Evaluation Hallucination (PRE-HAL) dataset, which enables the systematic evaluation of both perception and reasoning capabilities of LVLMs across multiple visual semantics, such as instances, scenes, and relations. Comprehensive evaluation with this new benchmark exposed more visual vulnerabilities, particularly in the more challenging task of relation reasoning. To address this issue, we propose, to the best of our knowledge, the first Dempster-Shafer theory (DST)-based visual hallucination detection method for LVLMs through uncertainty estimation. This method aims to efficiently capture the degree of conflict in high-level features at the model inference phase. Specifically, our approach employs simple mass functions to mitigate the computational complexity of evidence combination on power sets. We conduct an extensive evaluation of state-of-the-art LVLMs, LLaVA-v1.5, mPLUG-Owl2, and mPLUG-Owl3, with the new PRE-HAL benchmark. Experimental results indicate that our method outperforms five baseline uncertainty metrics, achieving average AUROC improvements of 4% and 10% across three LVLMs. Notably, it exhibits remarkable robustness in scene perception tasks. These results validate that feature-level conflict analysis offers a scalable, cost-effective solution for enhancing LVLM trustworthiness. Our code is available at <span><span>https://github.com/HT86159/Evidential-Conflict</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":13842,"journal":{"name":"International Journal of Approximate Reasoning","volume":"186 ","pages":"Article 109507"},"PeriodicalIF":3.2,"publicationDate":"2025-06-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144291414","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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":"Attribute combination reduction in formal concept analysis: A theoretical characterization","authors":"Qin Zhang ,&nbsp;Jianjun Qi ,&nbsp;Ling Wei ,&nbsp;Siyu Zhao","doi":"10.1016/j.ijar.2025.109498","DOIUrl":"10.1016/j.ijar.2025.109498","url":null,"abstract":"<div><div>Formal concept analysis is an approach relying on hierarchies of formal concepts to acquire knowledge from formal contexts, and is developed on the foundation of lattice theory. In formal concept analysis, reduction theory mainly consists of two types: attribute reduction and concept reduction. The former achieves data reduction but inevitably leads to the loss of the original information in the formal context. The latter involves the deletion of formal concepts while preserving the original information. This paper proposes a new type of reduction called attribute combination reduction to preserve object intents, which leverages the strengths of both attribute reduction and concept reduction while avoiding the limitations of attribute reduction. First, the definition of attribute combination reducts and the judgment theorem of attribute combination consistent sets are given. Then, the relationship between attribute combination reducts and concept reducts is investigated, and the properties of attribute combination reducts are explored. In addition, to narrow the scope for searching attribute combination reducts, a lower bound and an upper bound for their cardinality are provided from the perspectives of the set dimension and the Ferrers dimension of a formal context.</div></div>","PeriodicalId":13842,"journal":{"name":"International Journal of Approximate Reasoning","volume":"186 ","pages":"Article 109498"},"PeriodicalIF":3.2,"publicationDate":"2025-06-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144253444","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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":"Mixtures of probabilistic logic programs","authors":"Damiano Azzolini","doi":"10.1016/j.ijar.2025.109497","DOIUrl":"10.1016/j.ijar.2025.109497","url":null,"abstract":"<div><div>Structure learning (SL) is a fundamental task in Statistical Relational Artificial Intelligence, where the goal is to learn a program from data. Among the possible target languages, there is Probabilistic Logic Programming. Mixture models have recently gained attention thanks to their effectiveness in modeling complex distributions by combining simpler ones. In this paper, we propose learning a mixture of probabilistic logic programs to handle SL. Our method consists of three steps: 1) generating mixture components with a specific structure, 2) applying parameter learning to each component, and 3) optimizing the weights associated with each component. Furthermore, to possibly reduce the number of components and mitigate overfitting, we also explore the use of L1 and L2 regularization. Empirical results obtained by considering both the full set of components and only a fraction of them demonstrate that our approach, despite being seemingly simple, is competitive with state-of-the-art solvers.</div></div>","PeriodicalId":13842,"journal":{"name":"International Journal of Approximate Reasoning","volume":"186 ","pages":"Article 109497"},"PeriodicalIF":3.2,"publicationDate":"2025-06-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144240419","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}