Expert Systems最新文献

筛选
英文 中文
A Quarter-Century of Analogical Natural Language Processing: A Systematic Review 四分之一世纪的类比自然语言处理:系统回顾
IF 3.4 4区 计算机科学
Expert Systems Pub Date : 2026-11-04 DOI: 10.1111/exsy.70472
Bilel Elayeb, Myriam Bounhas, Erik Cambria, Amir Hussain
{"title":"A Quarter-Century of Analogical Natural Language Processing: A Systematic Review","authors":"Bilel Elayeb,&nbsp;Myriam Bounhas,&nbsp;Erik Cambria,&nbsp;Amir Hussain","doi":"10.1111/exsy.70472","DOIUrl":"https://doi.org/10.1111/exsy.70472","url":null,"abstract":"<div>\u0000 \u0000 <p>Analogical reasoning constitutes a foundational pillar of human cognition, enabling the seamless extrapolation of knowledge from known structural precedents to completely novel contexts. Within Computational Linguistics, the operationalization of relational proportions (<span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 <mi>a</mi>\u0000 <mo>:</mo>\u0000 <mi>b</mi>\u0000 <mo>:</mo>\u0000 <mo>:</mo>\u0000 <mi>c</mi>\u0000 <mo>:</mo>\u0000 <mi>d</mi>\u0000 </mrow>\u0000 <annotation>$$ a:b::c:d $$</annotation>\u0000 </semantics></math>) has emerged as a robust, knowledge-light paradigm for learning, inference, and data-driven generalization. This paper presents a comprehensive Systematic Literature Review (SLR) tracking the architectural taxonomy, structural transitions, and empirical milestones of analogical Natural Language Processing (NLP) over the past quarter-century (2000–2025). Adhering strictly to standardized PRISMA guidelines, we aggregate and synthesize 43 primary studies across four distinct evolutionary eras: discrete symbolic/algebraic frameworks, distributional pattern induction models, static embedding offset topologies, and deep contextualized transformer architectures. Each paradigm is critically evaluated in terms of mathematical formulations, dataset configurations, optimization metrics, and downstream operational limitations across core language processing tasks, including morphology, machine translation, and sentence semantic alignment. Methodologically, we reconcile qualitative evidence with quantitative benchmark syntheses, formalize axiomatic constraints across sentential domains, and assess potential publication bias. Our cross-paradigm synthesis exposes critical trade-offs between symbolic interpretability, vocabulary coverage limits, and continuous vector space distortions. Finally, we establish a forward-looking research agenda mapping current open problems—specifically focusing on text morphing generation, analogical dataset augmentation, personalized preference learning, and explainable neuro-symbolic interface integration—to guide future developments in data-efficient and interpretable natural language architectures.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"43 12","pages":""},"PeriodicalIF":3.4,"publicationDate":"2026-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149355988","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}
引用次数: 0
Structure-Guided Extractive Compression for Long-Document Summarization via Budget Allocation and Redundancy Filtering 基于预算分配和冗余过滤的长文档摘要结构导向抽取压缩
IF 3.4 4区 计算机科学
Expert Systems Pub Date : 2026-11-03 DOI: 10.1111/exsy.70468
Fengen Zhang, Tiegang Fan
{"title":"Structure-Guided Extractive Compression for Long-Document Summarization via Budget Allocation and Redundancy Filtering","authors":"Fengen Zhang,&nbsp;Tiegang Fan","doi":"10.1111/exsy.70468","DOIUrl":"https://doi.org/10.1111/exsy.70468","url":null,"abstract":"<div>\u0000 \u0000 <p>Long-document summarization with large language models is limited by context-window size and inference cost. A training-free compressor can reduce this cost by selecting source sentences before generation, but a single global ranking leaves the distribution of a fixed token budget across document units implicit. We separate allocation from sentence selection in a two-stage method that requires no task-specific training. It first assigns initial quotas to pre-defined units from their lengths and fixed structural weights. It then ranks sentences by similarity to the document centroid, applies a within-unit redundancy threshold and globally refills unused capacity. The resulting implementation is greedy and does not solve a linear or integer program. We evaluate one fixed Qwen2.5-7B-Instruct generator on GovReport, ArXiv, QMSum and an 84-case Multi-LexSum set whose median input length is 221,986 tokens. Against matched global-ranking controls, the method improves ROUGE-1 by 0.83 points on ArXiv and 4.04 points on QMSum, whereas GovReport results are mixed. Against the supervised MemSum extractor under a matched 1024-token evidence budget, it is weaker on ArXiv and mixed on GovReport. On Multi-LexSum, it exceeds a full-document hierarchical Qwen baseline by 3.66/1.39/1.29 ROUGE-1/2/L points overall and reduces end-to-end runtime from approximately 7.9 h to 50.6 min, although the 200K+ word subset does not show a conclusive quality advantage. The results support quota allocation as a practical, interpretable control while also showing that it is not consistently better than supervised or simple positional baselines.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"43 12","pages":""},"PeriodicalIF":3.4,"publicationDate":"2026-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149355725","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}
引用次数: 0
Hodges–Lehmann and Vision-Language Model Based Intelligent Fire Detection in Complex Environments 基于Hodges-Lehmann和视觉语言模型的复杂环境智能火灾检测
IF 3.4 4区 计算机科学
Expert Systems Pub Date : 2026-11-02 DOI: 10.1111/exsy.70469
Shenyuan Gao, Guozhong Huang, Chanseok Park, Pai Zheng, Ibrahim M. Hezam, Yusen Wang, Yanlin Sun, Xuehong Gao
{"title":"Hodges–Lehmann and Vision-Language Model Based Intelligent Fire Detection in Complex Environments","authors":"Shenyuan Gao,&nbsp;Guozhong Huang,&nbsp;Chanseok Park,&nbsp;Pai Zheng,&nbsp;Ibrahim M. Hezam,&nbsp;Yusen Wang,&nbsp;Yanlin Sun,&nbsp;Xuehong Gao","doi":"10.1111/exsy.70469","DOIUrl":"https://doi.org/10.1111/exsy.70469","url":null,"abstract":"<div>\u0000 \u0000 <p>Fires in industrial and environmental settings pose threats to process safety. Current vision-based fire detection models face challenges in balancing high accuracy with lightweight deployment, suffer from high false alarm rates, and lack a closed-loop inspection workflow. To enhance fire prevention capabilities and practical applicability, it is necessary to develop a methodology encompassing robust detection models and intelligent agents for fire safety inspection. Therefore, this study proposes a two-stage robust fire detection methodology. Particularly, a novel Hodges–Lehmann-based Robust Attention (HLRA) module is developed, utilising statistical estimators to achieve parameter-free spatial and channel attention. Based on this, a real-time detection model named RobustFireDet is proposed, integrating a triangular masked convolution and a robust HL-IoU loss to improve accuracy while reducing parameters. Subsequently, a classification model named RobustFireNet is designed to mitigate false alarms by focusing on small-target flames using the CoAlign Loss function. Experiments compare the proposed models with mainstream counterparts. RobustFireDet attains a detection [email protected] of 74.3%, surpassing the second-best YOLOv11 by 2.2 percentage points with 22.7% fewer parameters. RobustFireNet achieves a small-flame classification accuracy of 96.9%, surpassing the second-best EfficientNetV2-S by 2.5 percentage points. The two-stage pipeline further suppresses spurious fire boxes by 29.2% and false-alarm images by 25.4% in an end-to-end evaluation. On top of the two-stage detection results, a Vision-Language Model-based intelligent agent called FirePatrolAgent is further developed for scene analysis and recommendation generation, forming a closed-loop inspection system that completes detection, verification, analysis, and response within 25.43 s per risk event.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"43 12","pages":""},"PeriodicalIF":3.4,"publicationDate":"2026-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149355747","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}
引用次数: 0
Leveraging Large Language Models for Domain-Agnostic Aspect-Aware Recommendation 利用大型语言模型进行领域不可知的方面感知推荐
IF 3.4 4区 计算机科学
Expert Systems Pub Date : 2026-11-02 DOI: 10.1111/exsy.70464
Donghyeon Kim, Soohyun Woo, Seonu Park, Xinzhe Li, Jaekyeong Kim
{"title":"Leveraging Large Language Models for Domain-Agnostic Aspect-Aware Recommendation","authors":"Donghyeon Kim,&nbsp;Soohyun Woo,&nbsp;Seonu Park,&nbsp;Xinzhe Li,&nbsp;Jaekyeong Kim","doi":"10.1111/exsy.70464","DOIUrl":"https://doi.org/10.1111/exsy.70464","url":null,"abstract":"<div>\u0000 \u0000 <p>The growth of e-commerce and the corresponding increase in the number of online reviews have driven the demand for recommender systems that capture user preferences. Online reviews contain detailed evaluations of item aspects, and recent studies have employed aspect-based sentiment analysis (ABSA) to incorporate fine-grained user sentiment for improved personalisation. However, existing ABSA-based recommendation approaches rely on predefined aspect categories, which prevents them from capturing user preferences for specific aspect terms and limits their generalisation across diverse domains. To address these limitations, this study proposes a novel ABSA-based recommendation framework, domain-agnostic aspect-aware recommendation (DAAR), which leverages large language models (LLMs) to extract aspect terms and integrate them with sentiment information. First, aspect terms that represent the targets of user evaluation are extracted from review texts using LLMs. Then, semantic representations of the aspect terms are generated using Phrase-BERT based on bidirectional encoder representations from transformers (BERT), and sentiment scores for each aspect term are obtained through a sentiment analysis model based on decoding-enhanced BERT with disentangled attention (DeBERTa). Finally, the two representations are combined into a unified representation, and an attention mechanism is applied to capture the relative importance among aspect terms. The resulting representation is then integrated with user–item interactions to predict the final rating. Experimental results on the Yelp and Amazon review datasets demonstrate that DAAR outperforms the baseline models and shows strong generalisation capabilities. This indicates that the proposed framework alleviates domain dependency and more precisely captures user preferences by integrating users' sentiment towards aspect terms.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"43 12","pages":""},"PeriodicalIF":3.4,"publicationDate":"2026-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149355748","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}
引用次数: 0
Enforcing Graph-Relative Structural Compliance in LLM-Based Industrial Anomaly Explanation via Causal Graph Grounding 基于因果图基础的llm工业异常解释中的图相关结构遵从性
IF 3.4 4区 计算机科学
Expert Systems Pub Date : 2026-11-01 DOI: 10.1111/exsy.70467
Yao Zhang, Lei An, Qiang Zhao
{"title":"Enforcing Graph-Relative Structural Compliance in LLM-Based Industrial Anomaly Explanation via Causal Graph Grounding","authors":"Yao Zhang,&nbsp;Lei An,&nbsp;Qiang Zhao","doi":"10.1111/exsy.70467","DOIUrl":"https://doi.org/10.1111/exsy.70467","url":null,"abstract":"<div>\u0000 \u0000 <p>We study when large language model (LLM) root-cause reports can be trusted, and find compliance and correctness to be separable properties. We <i>ground</i> the LLM with a directed dependency graph discovered from the data: PCMCI learns the structure from the normal regime, and the report may cite only its edges—compliance with the discovered structure, not proof that mechanisms are true. Compliance is measured by a deterministic, mechanically computable <i>edge-set violation rate</i> (EVR): the share of cited edges outside the allowed set. EVR is graph-relative by construction; it verifies membership in the discovered graph, never the truth of any relation. An enum constraint with client-side validation keeps EVR at zero <i>by construction</i>—a deployment guarantee, not a finding; a soft instruction alone reaches zero at small scale (5%–6% residual at 30 seeds). Correctness is another matter: on degree-matched random graphs (30 seeds) soft-mode EVR stays near zero while chain precision halves (0.54<span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 <mo>→</mo>\u0000 </mrow>\u0000 <annotation>$$ to $$</annotation>\u0000 </semantics></math>0.28) and recall falls 0.91<span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 <mo>→</mo>\u0000 </mrow>\u0000 <annotation>$$ to $$</annotation>\u0000 </semantics></math>0.55—on this synthetic protocol, only genuinely causal structure restores correctness. A true-statistic table roughly halves numeric fabrication, and enum-constrained structured numeric fields eliminate fabrication in the constrained channel only (validity 5/5). The grounded system may probabilistically abstain at its retrieval boundary. On real plants (SWaT, WADI) compliance transfers, but root identification remains open (SWaT 3/27; WADI, assessable after a tag-matching correction, exploratory: grounded hit@1 1/3 under the redesigned unit-invariant selection, 0/3 legacy), and discovered dependencies only partially persist across attack windows. Graph-relative compliance is enforceable; causal correctness is bounded by discovery quality.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"43 12","pages":""},"PeriodicalIF":3.4,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149355815","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}
引用次数: 0
Optimization of Advertising Bidding Model With Generative AI and Deep Reinforcement Learning Collaboration: Design and Validation of Closed-Loop Simulation Framework 基于生成人工智能和深度强化学习的广告竞价模型优化:闭环仿真框架的设计与验证
IF 3.4 4区 计算机科学
Expert Systems Pub Date : 2026-10-26 DOI: 10.1111/exsy.70444
Yuanchun Xu, Erming Xu
{"title":"Optimization of Advertising Bidding Model With Generative AI and Deep Reinforcement Learning Collaboration: Design and Validation of Closed-Loop Simulation Framework","authors":"Yuanchun Xu,&nbsp;Erming Xu","doi":"10.1111/exsy.70444","DOIUrl":"https://doi.org/10.1111/exsy.70444","url":null,"abstract":"<div>\u0000 \u0000 <p>Real-time bidding (RTB) requires decisions in milliseconds under nonstationary market dynamics; traditional optimization and online RL suffer from sample efficiency, stability and generalization limitations. This study proposes a closed-loop simulation framework coupling generative AI with deep reinforcement learning. A cascaded encoder combining a conditional variational autoencoder (CVAE) with a denoising diffusion probability model (DDPM) synthesizes high-fidelity market states. A proximal policy optimization (PPO) agent trains in this environment via a three-stage ‘generate, evaluate, optimize’ mechanism, with a dynamic feedback regulator forming a co-evolutionary loop. Experiments on 47.82 million anonymized bidding records improve CTR by 23.7%, CVR by 31.2%, ROI by 23.5% over DQN and increase ROI by 6.1% over GAVE and 8.6% over DiffBid. A 2 × 2 factorial ablation confirms significant generator-agent interaction (<span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 <mover>\u0000 <mi>β</mi>\u0000 <mo>̂</mo>\u0000 </mover>\u0000 </mrow>\u0000 <annotation>$$ hat{beta} $$</annotation>\u0000 </semantics></math> = 0.37 ROI units, 95% CI [0.24, 0.51], <span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 <mi>p</mi>\u0000 </mrow>\u0000 <annotation>$$ p $$</annotation>\u0000 </semantics></math> = 0.008, <span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 <msub>\u0000 <mi>p</mi>\u0000 <mtext>Bonf</mtext>\u0000 </msub>\u0000 </mrow>\u0000 <annotation>$$ {p}_{Bonf} $$</annotation>\u0000 </semantics></math> = 0.024). Cross-dataset transfer on iPinYou preserves rankings with 7.0% ROI advantage over GAVE; counterfactual OPE on real logs (SNIPS, doubly robust, auction replay) agrees with simulator ranking. The system achieves 87.4% sample utilization (distinct from GPU utilization 92.30%) and 14.8 ms mean inference latency (p50/p95/p99 = 12.4/18.7/23.1 ms; 100% SLA at 100,000 req/s). A live A/B test on a production platform is planned as future work.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"43 12","pages":""},"PeriodicalIF":3.4,"publicationDate":"2026-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149258794","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}
引用次数: 0
A Network-Enhanced Supervised Learning Framework for Early Prediction of Corruption Risk Indicators in Public Procurement Data 用于公共采购数据中腐败风险指标早期预测的网络增强监督学习框架
IF 3.4 4区 计算机科学
Expert Systems Pub Date : 2026-10-26 DOI: 10.1111/exsy.70465
Kostas Loumponias, Andreas Kosmatopoulos, Foivos Katsetsiadis, Theodora Tsikrika, Stefanos Vrochidis, Ioannis Kompatsiaris
{"title":"A Network-Enhanced Supervised Learning Framework for Early Prediction of Corruption Risk Indicators in Public Procurement Data","authors":"Kostas Loumponias,&nbsp;Andreas Kosmatopoulos,&nbsp;Foivos Katsetsiadis,&nbsp;Theodora Tsikrika,&nbsp;Stefanos Vrochidis,&nbsp;Ioannis Kompatsiaris","doi":"10.1111/exsy.70465","DOIUrl":"https://doi.org/10.1111/exsy.70465","url":null,"abstract":"<p>A tender receiving only one offer (single bidding) is generally recognized as a sign of restricted competition and a potential corruption risk indicator. Existing corruption risk models rely predominantly on ex post information, limiting their usefulness for early stage prediction. This paper introduces a network-enhanced supervised learning framework that predicts single bidding outcomes at the time of tender announcement, using only features available at the first call for tender. The framework integrates conventional procurement attributes with historical buyer behaviour metrics and with relational features derived from annual buyer–supplier interaction networks, including centrality measures, community structure and node embeddings. To ensure temporal validity, all features are constructed exclusively from data available prior to tender publication. The methodology is evaluated across three European countries, Bulgaria, Croatia and Serbia, using country-specific models to account for institutional and market heterogeneity. Results show that the Random Forest model enriched with network-based and historical features consistently outperforms baselines relying solely on transactional attributes, achieving macro F1-scores between <span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 <mn>0.69</mn>\u0000 </mrow>\u0000 <annotation>$$ 0.69 $$</annotation>\u0000 </semantics></math> and <span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 <mn>0.80</mn>\u0000 </mrow>\u0000 <annotation>$$ 0.80 $$</annotation>\u0000 </semantics></math>. These results demonstrate that single bidding can be forecast reliably at the time of tender announcement and highlight the value of incorporating relational structures and buyer behaviour into procurement analytics. The proposed framework offers a scalable foundation for early warning systems that support proactive integrity monitoring and more transparent procurement governance.</p>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"43 12","pages":""},"PeriodicalIF":3.4,"publicationDate":"2026-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/exsy.70465","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149258793","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
StemGuard: An Expertise-Reinforced and Voice-Controlled Agent for Stem Cell Culture Surveillance StemGuard:一种专业强化和语音控制的干细胞培养监测代理
IF 3.4 4区 计算机科学
Expert Systems Pub Date : 2026-10-25 DOI: 10.1111/exsy.70452
Kai Li, Yating Guo, Zhao Chen, Yubin Niu
{"title":"StemGuard: An Expertise-Reinforced and Voice-Controlled Agent for Stem Cell Culture Surveillance","authors":"Kai Li,&nbsp;Yating Guo,&nbsp;Zhao Chen,&nbsp;Yubin Niu","doi":"10.1111/exsy.70452","DOIUrl":"https://doi.org/10.1111/exsy.70452","url":null,"abstract":"<div>\u0000 \u0000 <p>Vitro stem cell culture is vital for developing antiaging and antiallergy drugs. Cell confluence acts as a core indicator of cell proliferation and culture status. However, confluence estimation mostly depends on manual labour, as there are few artificial-intelligence (AI) models or professional software developed for the task. Although most existing AI works adopted cell segmentation networks for confluence estimation, their efficacy and efficiency need improving. Meanwhile, the growth status of the cells is daily monitored with their images captured and archived by technicians using microscopes. As a result, cell images are constantly generated, exhibiting large data variance and posing challenges to the estimation models. Therefore, we propose StemGuard, perhaps the novel whole-process AI agent specified at cell confluence estimation. It completes the basic confluence estimation task by a novel backbone that leverages categorization and feature fusion. It features an expertise-reinforced optimization mechanism, which enables adaptive learning on the dynamic data. It also includes a voice control module for better use experience in sterile laboratories. In addition, we define a novel index, Average Confusion Degree (ACD), to assess confluence estimation. It measures the confluence error range rather than the raw error, as the former is more meaningful than the latter to the technicians. Evaluated on real microscopic cell images provided by a drug company, StemGuard runs smoothly and delivers state-of-the-art cell confluence estimation performance, achieving an ACD of 0.0421 and Mean Absolute Error (MAE) of 0.0638.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"43 12","pages":""},"PeriodicalIF":3.4,"publicationDate":"2026-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149259122","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}
引用次数: 0
A Multi-Blockchain Structure in the Carbon Market Trading: A Case Analysis in the Chinese Carbon Exchange 碳市场交易中的多区块链结构——以中国碳交易为例
IF 3.4 4区 计算机科学
Expert Systems Pub Date : 2026-10-25 DOI: 10.1111/exsy.70461
Moting Su, Ruoyu Zhao, Jiajia Jiang, Xiangli Xiao, Fengshu Li
{"title":"A Multi-Blockchain Structure in the Carbon Market Trading: A Case Analysis in the Chinese Carbon Exchange","authors":"Moting Su,&nbsp;Ruoyu Zhao,&nbsp;Jiajia Jiang,&nbsp;Xiangli Xiao,&nbsp;Fengshu Li","doi":"10.1111/exsy.70461","DOIUrl":"https://doi.org/10.1111/exsy.70461","url":null,"abstract":"<div>\u0000 \u0000 <p>We propose a novel multi-blockchain framework for carbon futures trading to enhance innovation and liquidity in China's national carbon market. Our approach introduces: (1) a tailored trading framework that integrates standard operations with regulatory requirements from multiple authorities; (2) a design of four distinct blockchain types to ensure data privacy and seamless system interaction; and (3) a cross-chain mechanism for data consistency, verification and compliance. Experimental results validate the model's effectiveness. This work supports China's dual-carbon goals by advancing carbon finance through a secure and efficient trading system.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"43 12","pages":""},"PeriodicalIF":3.4,"publicationDate":"2026-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149259123","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}
引用次数: 0
TB-BasilNet: Weakly Supervised Segmentation of Tuberculosis Bacilli in Ziehl–Neelsen Sputum Smear Microscopy TB-BasilNet: Ziehl-Neelsen痰涂片镜检中结核杆菌的弱监督分割
IF 3.4 4区 计算机科学
Expert Systems Pub Date : 2026-10-22 DOI: 10.1111/exsy.70463
Cuneyt Ozdemir, Özlem Genç, Serpil Genç, Mehmet Ali Gedik
{"title":"TB-BasilNet: Weakly Supervised Segmentation of Tuberculosis Bacilli in Ziehl–Neelsen Sputum Smear Microscopy","authors":"Cuneyt Ozdemir,&nbsp;Özlem Genç,&nbsp;Serpil Genç,&nbsp;Mehmet Ali Gedik","doi":"10.1111/exsy.70463","DOIUrl":"https://doi.org/10.1111/exsy.70463","url":null,"abstract":"<div>\u0000 \u0000 <p>Automated segmentation of <i>Mycobacterium tuberculosis</i> bacilli in Ziehl–Neelsen (ZN)-stained sputum smear images remains challenging because bacilli are extremely small, sparse, and easily confused with staining artefacts. While prior studies report competitive classification and detection results, fewer provide a segmentation-focused analysis under realistic weak supervision, and fewer still evaluate on slides that were never seen during training. We present <i>TB-BasilNet</i>, a U-Net-based encoder–decoder that adds three lightweight, rod-specific components—an oriented-line stem, orientation-aware convolutions in the high-resolution encoder stages, and a full-resolution branch—at a cost of only 1.3% additional parameters, and trained from expert bounding-box annotations converted into pseudo-masks by stain-aware region-of-interest extraction. Models were evaluated on a single-centre dataset of 1512 images from 28 smears under a slide-level five-fold cross-validation protocol with three training seeds, native-resolution patch training and sliding-window inference, so that every reported score reflects performance on unseen slides. TB-BasilNet achieved the highest mean Dice among the compared architectures (0.704 versus 0.696 for U-Net) and exceeded U-Net in every fold, with the highest pixel-level recall (0.736) and bacillus-level detection recall (0.833), and with a decision threshold that degrades more gracefully than the baselines. A component study shows that global-context and attention modules, which are common in elaborate segmentation designs, reduce performance for thin sparse targets, whereas the proposed additions improve it. Inter-slide variation (per-slide Dice 0.40–0.87), stain precipitate and the box-derived super-vision are identified as the principal remaining constraints, indicating that slide-level validation, annotation fidelity and morphology preservation matter more than architectural complexity in ZN bacillus segmentation.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"43 12","pages":""},"PeriodicalIF":3.4,"publicationDate":"2026-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149258505","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}
引用次数: 0
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信
小红书