Applied Soft Computing最新文献

{"title":"Limited-budget consensus with maximum consensus level for group decision making","authors":"Huanhuan Zhang ,&nbsp;Dongjie Guo ,&nbsp;Yifeng Ma","doi":"10.1016/j.asoc.2025.113905","DOIUrl":"10.1016/j.asoc.2025.113905","url":null,"abstract":"<div><div>Group decision making usually requires in-depth discussions to form a consensus acceptable to the entire group, which has attracted more and more research in recent years. Despite extensive studies on soft consensus, the relationship between consensus level and consensus cost remains unclear. This study establishes—for the first time—a precise mathematical relationship demonstrating that higher consensus levels require proportionally greater consensus costs. This finding provides critical theoretical grounding for consensus modeling. Recognizing that the cost of achieving consensus cannot be infinite and must be within a certain budget, we develop a model to determine the maximum achievable consensus level under a limited-budget. The consensus model with non-cooperators is also explored and formulated. The proposed models are applied to online lending platforms, providing a practical framework for measuring consensus levels and achieving soft consensus between lenders and borrowers within a limited-budget. This work contributes to understanding the relationship between consensus level and consensus cost, as well as the achievement of the maximum possible consensus level under limited-budget, which is relevant in scenarios where resources are finite.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"185 ","pages":"Article 113905"},"PeriodicalIF":6.6,"publicationDate":"2025-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145183697","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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":"Simulating phenomenal consciousness using generative agents based on large language models","authors":"Hanzhong Zhang ,&nbsp;Jibin Yin ,&nbsp;Haoyang Wang ,&nbsp;Ziwei Xiang","doi":"10.1016/j.asoc.2025.113922","DOIUrl":"10.1016/j.asoc.2025.113922","url":null,"abstract":"<div><div>Large Language Models (LLMs) still face challenges in tasks requiring understanding implicit instructions and applying common-sense knowledge. In such scenarios, LLMs may require multiple attempts to achieve human-level performance, potentially leading to inaccurate responses or inferences in practical environments, affecting their long-term consistency and behavior. This paper introduces the Internal Time-Consciousness Machine (ITCM), a computational consciousness structure to simulate the process of human consciousness. We further propose the ITCM-based Agent (ITCMA), which supports action generation and reasoning in open-world settings, and can independently complete tasks. ITCMA enhances LLMs’ ability to understand implicit instructions and apply common-sense knowledge by considering agents’ interaction and reasoning with the environment. The trained ITCMA performs better than state-of-the-art (SOTA) in the seen set. Even untrained ITCMA can achieve higher task completion rates than SOTA on the seen set, indicating its superiority over traditional intelligent agents in utility and generalization. In real-world tasks with quadruped robots, the task completion rate of untrained ITCMA is close to its performance in the unseen set, demonstrating its comparable utility and universality in real-world settings.</div><div>CCS Concepts: <span><math><mo>∙</mo></math></span> Human-centered computing <span><math><mo>→</mo></math></span> Interactive systems and tools; <span><math><mo>∙</mo></math></span> Computing methodologies <span><math><mo>→</mo></math></span> Natural language processing.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"185 ","pages":"Article 113922"},"PeriodicalIF":6.6,"publicationDate":"2025-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145158924","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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":"Rural road extraction from remote sensing images based on multi-view contextual information and multi-stage features","authors":"Langping Li ,&nbsp;Jizheng Yi ,&nbsp;Pengyu Lei ,&nbsp;Hengkai Lou ,&nbsp;Xiaoyao Li ,&nbsp;Hui Lin","doi":"10.1016/j.asoc.2025.113975","DOIUrl":"10.1016/j.asoc.2025.113975","url":null,"abstract":"<div><div>Accurate extraction of rural roads from high-resolution optical remote sensing images is of great significance to the development of rural areas, road navigation, rural land resource planning and other applications. Different from urban roads, rural ones with complex terrain backgrounds are often slender and winding, thereby making them more susceptible to vegetation cover. In order to improve the reliability and accuracy of rural road extraction, a Complex Rural Road Extraction Network (CRRENet) is proposed in this work, which consists of five parts: feature encoder, Multi-view Contextual Information Extraction Module (MCIEM), Multi-stage Feature Fusion Module (MFFM), Channel Coordinate Attention Mechanism (CCAM) and feature decoder. The MCIEM extracts the multi-view contextual information by the parallel dilated convolution with different dilation rates. To avoid the loss of image details, the MFFM integrates different feature maps from the downsampling stages. By adjusting the weights of feature maps, the CCAM enables the network to self-adaptively suppress the background noise and focus on the road foreground. Ablation and comparison validate CRRENet's superiority.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"185 ","pages":"Article 113975"},"PeriodicalIF":6.6,"publicationDate":"2025-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145158927","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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 data-driven approach to tackling academic stress-coping and mental health issues in college students using spherical fuzzy MARCOS methodology","authors":"Raiha Imran ,&nbsp;Munazza Amin ,&nbsp;Kifayat Ullah ,&nbsp;Dragan Pamucar ,&nbsp;Zeeshan Ali ,&nbsp;Oumaima Saidani ,&nbsp;Vladimir Simic","doi":"10.1016/j.asoc.2025.113925","DOIUrl":"10.1016/j.asoc.2025.113925","url":null,"abstract":"<div><div>The drastically developing nature of the knowledge economy and the rising need for top-notch expertise have placed tremendous pressure on college students. As higher education becomes more accessible, masses of students are enrolling in colleges, which puts additional pressure on colleges and institutions; as a result, they cannot provide adequate resources to the students. As the class size increases, many students require mental health assistance, academic guidance, and financial aid, which then puts pressure on the teachers and the facilities. This flood of students overloads the facilities, resulting in it becoming more challenging to provide attention and concern, leading many students to feel overlooked and affecting their mental health. Due to not getting timely support, students may find it challenging to handle their academic responsibilities. Moreover, the students face a heavy workload, unclear guidance, and limited resource access. The objective of this study is to develop a structured, data-driven decision-making framework for systematically evaluating and improving student mental health and academic stress-coping strategies in a college setting. To address this, a comprehensive decision-making structure, measurement of alternatives, and ranking according to the compromised solution (MARCOS) within the spherical fuzzy (SF) environment, has been applied, which evaluates the key factors causing mental health issues by comparing the ideal and anti-ideal alternatives. The novelty of the proposed approach lies in leveraging the SF framework’s explicit ability to model hesitation (abstinence) alongside truth and falsity degrees, enabling more accurate representation of subjective psychological assessments compared to traditional fuzzy models. Furthermore, the method calculates utility functions corresponding to each alternative (coping technique), prioritizes the strategies, and selects the most effective intervention. The results reveal that personalized mental health plans emerged as the top-ranked coping strategy, highlighting the importance of tailored support in culturally and contextually diverse academic environments.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"185 ","pages":"Article 113925"},"PeriodicalIF":6.6,"publicationDate":"2025-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145159666","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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 deep reinforcement learning trader without offline training","authors":"Boian Lazov","doi":"10.1016/j.asoc.2025.113881","DOIUrl":"10.1016/j.asoc.2025.113881","url":null,"abstract":"<div><div>In this paper we pursue the question of a fully online trading algorithm (i.e. one that does not need offline training on previously gathered data). For this task we consider Double Deep <span><math><mi>Q</mi></math></span>-learning in the episodic setting with Fast Learning Networks approximating the expected reward <span><math><mi>Q</mi></math></span>. Additionally, we define the possible terminal states of an episode in such a way as to introduce a mechanism to conserve some of the money in the trading pool when market conditions are seen as unfavourable. Some of these money are taken as profit and some are reused at a later time according to certain criteria. After describing the algorithm, we test it using 1-minute-tick price data for 4 major cryptocurrencies from Binance. We see that the agent performs better than trading with randomly chosen actions on each timestep. And it does so when tested on the whole dataset for a given market as well as on different subsets, representing different market trends.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"185 ","pages":"Article 113881"},"PeriodicalIF":6.6,"publicationDate":"2025-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145158920","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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":"Causal inference and attribute correlation consistency for eliminating popularity bias in recommendation system","authors":"Bo Huang ,&nbsp;Zhenhao Wang ,&nbsp;Feifan Yan ,&nbsp;Jin Liu ,&nbsp;Hamido Fujita ,&nbsp;Xiao Liu ,&nbsp;Hanan Aljuaid","doi":"10.1016/j.asoc.2025.113928","DOIUrl":"10.1016/j.asoc.2025.113928","url":null,"abstract":"<div><div>Popularity bias has long been a persistent issue in recommendation systems, leading to misleading results and significant problems such as the Matthew Effect and the Information Cocoon Room. Existing studies have primarily focused on the elevation of long-tailed items, overlooking the crucial connection between users and items. Drawing inspiration from causal graphs, this paper introduces a novel framework called CIACC (Causal Inference and Attribute Correlation Consistency) to tackle the challenges posed by popularity bias. The framework leverages causal graphs to evaluate the compatibility between users and items and to gauge the influence of item popularity on rankings. It employs counterfactual inference to estimate the impact of item popularity on rankings and adheres to the consistent principle of attribute correlation to enhance the feature representation of long-tailed items. Through rigorous experiments conducted on three public datasets, we demonstrate that our CIACC framework outperforms state-of-the-art methods.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"185 ","pages":"Article 113928"},"PeriodicalIF":6.6,"publicationDate":"2025-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145268869","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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":"Boosting fuzzy classification rules under footprint of uncertainty of type-2 fuzzy sets","authors":"Jidong Li ,&nbsp;Jian Cui ,&nbsp;Qian Su","doi":"10.1016/j.asoc.2025.113876","DOIUrl":"10.1016/j.asoc.2025.113876","url":null,"abstract":"<div><div>This paper presents a novel approach to improving classification accuracy in fuzzy rule-based systems by selecting Embedded Type-1 Fuzzy Sets (ET1 FSs) from Footprint of Uncertainty (FOU) areas. The method consists of three stages: (1) learning Type-1 fuzzy rules from predefined linguistic variables, (2) generating FOU areas with Interval Type-2 Fuzzy Sets (IT2 FSs), and (3) using the adaboost ensemble method, where multi-class problems are decomposed into binary classification tasks, and ET1 FSs are iteratively selected via genetic algorithms. By relying on IT2 FSs for the flexible partitioning of classification boundaries, the method enhances accuracy while addressing challenges in high-dimensional and multi-class problems. Experiments were performed on 12 UCI datasets and three image classification tasks using features from pre-trained convolutional neural networks. These datasets were selected to ensure diversity in dimensionality and class distribution. Comparative analyses with several state-of-the-art classification methods demonstrate that IT2 FSs can be effectively used to develop accurate classification systems. Additionally, this work analyzes trade-offs between complexity and interpretability by tuning FOU size (distinguishability) and adjusting the number of rules. The results show that a balanced FOU size and rule count yield better accuracy gains than either alone. Furthermore, several suitable trade-off regions with their parameters are presented.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"185 ","pages":"Article 113876"},"PeriodicalIF":6.6,"publicationDate":"2025-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145159717","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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":"Multi-criteria linguistic optimization for covert communication in secure LLM-based steganography","authors":"Kamil Woźniak ,&nbsp;Marek R. Ogiela ,&nbsp;Lidia Ogiela","doi":"10.1016/j.asoc.2025.113960","DOIUrl":"10.1016/j.asoc.2025.113960","url":null,"abstract":"<div><div>This paper presents a novel framework for covert communication through secure steganography using large language models (LLMs). Our approach leverages multi-criteria linguistic optimization to encode secret information directly into stylistic features of auto-regressively generated text. This strategy balances embedding capacity with naturalness and coherence. The secret message is partitioned into fixed-size blocks. Each block is embedded into binary stylistic feature vectors via a surjective linear mapping, which introduces redundancy. This redundancy enables the use of a history-aware cost function that selects stylistic vectors to minimize abrupt transitions and preserve fluency across sentences. Candidate sentences are generated by prompting LLMs with contextual and stylistic constraints. Rejection sampling then ensures exact feature matching and high linguistic quality. Experimental evaluation in multiple LLMs, diverse text contexts, and parameter settings demonstrates effective embedding capacities of up to 0.30 bits per token while maintaining strong linguistic naturalness, validated through perplexity, lexical diversity, readability, and a linguistic acceptability metric. Importantly, decoding recovers the full secret with zero error under ideal conditions. This confirms the reliability of the method. The current work focuses on embedding efficiency and imperceptibility. Robustness against active text alterations and formal undetectability assessments remain open challenges for future research. The proposed multi-criteria linguistic optimization framework offers a promising avenue for advanced covert communication by harmonizing secure information embedding with fluent, human-like language generation.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"185 ","pages":"Article 113960"},"PeriodicalIF":6.6,"publicationDate":"2025-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145159714","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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":"Hierarchical layered multigraph network with scale importance estimation for image classification","authors":"João Pedro Oliveira Batisteli ,&nbsp;Nicolas Passat ,&nbsp;Silvio Jamil Ferzoli Guimarães ,&nbsp;Zenilton Kleber Gonçalves do Patrocínio Júnior","doi":"10.1016/j.asoc.2025.113877","DOIUrl":"10.1016/j.asoc.2025.113877","url":null,"abstract":"<div><div>This work introduces a novel image representation and processing approach using Graph Neural Networks (GNNs). We propose a multigraph representation named <strong>H</strong>i<strong>E</strong>rarchical <strong>L</strong>ayered <strong>M</strong>ultigraph (HELM), which explicitly encodes spatial and hierarchical relationships as distinct edge types, overcoming the limitations of existing methods that fail to fully exploit relational information in images. A multi-scale representation is generated through hierarchical segmentation of a superpixel base graph, enabling the computation of spatial and hierarchical relationships within and across scales. To effectively process this multi-relational information, we introduce the <strong>H</strong>i<strong>E</strong>rarchical <strong>L</strong>ayered <strong>M</strong>ultigraph <strong>Net</strong>work (HELMNet), a novel GNN architecture incorporating specialized mechanisms for selectively aggregating and fusing information from each distinct edge type. Additionally, it includes a Region Graph Readout (RGR) module that employs an attention mechanism to dynamically weight the contribution of each representation scale during the aggregation process for classification. Experimental results demonstrate the greater efficacy of HELMNet for image classification. Compared to hierarchical models that do not distinguish between edge types, HELMNet obtains substantial average accuracy gains of 2.1% (significant at 3% level) and 10% (significant at 0.1% level) on the CIFAR-10 and STL-10 datasets, respectively. On the EUROSAT dataset, HELMNet achieves over 95% accuracy, requiring only 0.73% of the best-performing state-of-the-art model size (in number of parameters). For the more demanding and high-resolution RESISC45 dataset, the proposed model still delivers impressive results, achieving an accuracy of over 85%.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"185 ","pages":"Article 113877"},"PeriodicalIF":6.6,"publicationDate":"2025-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145183693","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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":"Coverage exploration of unknown obstacle-cluttered environments using a swarm of ground robots","authors":"Khalil Al-rahman Youssefi,&nbsp;Wilfried Elmenreich","doi":"10.1016/j.asoc.2025.113964","DOIUrl":"10.1016/j.asoc.2025.113964","url":null,"abstract":"<div><div>This paper introduces a coverage exploration algorithm for unknown obstacle-cluttered environments using a swarm of ground robots. A key contribution of this work is the proposed fitness function, which balances multiple exploration objectives and encourages robots to disperse effectively, avoiding excessive overlapping visits. The robots are assumed to start from a single corner of the environment, reflecting practical situations where pre-distributing them is not feasible. This setup highlights a key feature of the algorithm, as it enables self-organization and effective distribution of the robots throughout the environment. The robustness of the method is demonstrated through experiments in various environmental setups, showing its resilience to different obstacle structures and reliable performance across diverse scenarios. The approach also leverages the benefits of swarm behavior, where an increasing number of robots improves exploration efficiency through enhanced collaboration and coverage. The algorithm is evaluated against a swarm random walk approach and two multi-robot meta-heuristic methods, significantly outperforming them in terms of coverage efficiency and robustness.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"185 ","pages":"Article 113964"},"PeriodicalIF":6.6,"publicationDate":"2025-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145159812","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}