Procedia Computer Science最新文献_第9页

The Impact of Artificial Intelligence Pilot Policies on Stock Price Volatility: Evidence from Chinese Listed Firms 人工智能试点政策对股价波动的影响：来自中国上市公司的证据

Procedia Computer Science Pub Date : 2026-01-01 Epub Date: 2026-09-18 DOI: 10.1016/j.procs.2026.08.041

Xiyue Wang, Jingyu Li

引用次数: 0

Dimension-Partitioned Variable-Length Genetic Algorithm for Layered Protection Optimization of High-Value Assets 高价值资产分层保护优化的分维变长遗传算法

Procedia Computer Science Pub Date : 2026-01-01 Epub Date: 2026-09-18 DOI: 10.1016/j.procs.2026.08.046

Jue Wang, Shuren Liu, Deqiang Hu, Yitao Wang, Xianneng Li, Yang Yu

{"title":"Dimension-Partitioned Variable-Length Genetic Algorithm for Layered Protection Optimization of High-Value Assets","authors":"Jue Wang, Shuren Liu, Deqiang Hu, Yitao Wang, Xianneng Li, Yang Yu","doi":"10.1016/j.procs.2026.08.046","DOIUrl":"10.1016/j.procs.2026.08.046","url":null,"abstract":"<div><div>To address the challenge of mass approaches by unidentified Unmanned Surface Vehicle (USV) swarms toward maritime High-Value Assets (HVAs), this paper investigates the resource optimization of a layered protection system. Given that the optimal number of protection layers is uncertain, resource optimization presents a complex variable-dimensional nonlinear problem. To solve this, this paper proposes a Dimension-Partitioned Variable-Length Genetic Algorithm (DP-VLGA). The algorithm employs a hybrid variable-length encoding strategy to map protection layers, spatial layout, and Unmanned Aerial Vehicle(UAV) resource allocation onto variable-length chromosomes, enabling the adaptive evolution of protection configurations. To overcome premature convergence common in variable-dimensional search spaces, a dimension-partitioned solution pool management mechanism is introduced. Furthermore, an adaptive alignment-based crossover operator is developed to ensures physical validity during gene exchange across different dimensions. Experimental results demonstrate that DP-VLGA adaptively generates optimal protection configurations under varying approach intensities. Compared to other variable-dimensional optimization algorithms, the proposed method exhibits significant advantages in search accuracy and convergence stability, achieving higher control probabilities with lower resource consumption.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"287 ","pages":"Pages 256-263"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148981003","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

AI and Business in Green and Digital Era: Preface for ITQM 2026 绿色与数字时代的人工智能与商业：ITQM 2026前言

Procedia Computer Science Pub Date : 2026-01-01 Epub Date: 2026-09-18 DOI: 10.1016/j.procs.2026.08.001

Fouad Ben Abdelaziz, Yong Shi, Mirco Peron, Yi Qu, Simon Voorberg, Pei Quan, Jianping Li, Gang Kou, James M Tien, Daniel Berg

引用次数: 0

Measuring Architectural Efficiency in Agentic Manufacturing: The Iaea Index and its Application in the MECA Framework 计量代理制造业的建筑效率：原子能机构指数及其在MECA框架中的应用

Procedia Computer Science Pub Date : 2026-01-01 Epub Date: 2026-09-18 DOI: 10.1016/j.procs.2026.08.004

I. Lopez-Juarez, J.A. Vazquez-Lopez, S. De la Fuente-Ruiz, L.A. Castillo-Barrientos, R. Osorio-Comparan, G. Lefranc

{"title":"Measuring Architectural Efficiency in Agentic Manufacturing: The Iaea Index and its Application in the MECA Framework","authors":"I. Lopez-Juarez, J.A. Vazquez-Lopez, S. De la Fuente-Ruiz, L.A. Castillo-Barrientos, R. Osorio-Comparan, G. Lefranc","doi":"10.1016/j.procs.2026.08.004","DOIUrl":"10.1016/j.procs.2026.08.004","url":null,"abstract":"<div><div>The transition toward autonomous manufacturing in Industry 5.0 necessitates a paradigm shift from qualitative descriptions of “intelligence” to quantitative measures of architectural robustness. This paper addresses the current lack of holistic metrics for vertically integrated agentic systems by formalizing the <strong>Agentic Architectural Efficiency Index</strong> (I<sub>AEA</sub>). This novel mathematical index evaluates the synergy between cognitive reasoning and physical determinism by integrating three critical dimensions: causal traceability, vertical completeness, and cumulative latency. To validate this metric, we implement the <strong>Modular Explainable Contextual Agentic (MECA)</strong> framework, an architecture designed to bridge high-level AI intent with real-time robotic execution. The framework’s core novelty lies in the coupling of Timed Petri Nets for formal safety verification and ROS2/DDS for deterministic communication. Preliminary results from a numerical case study in a robotic polishing cell demonstrate that the MECA framework achieves a tenfold increase in architectural efficiency (I<sub>AEA</sub> ≈ 0.65) compared to baseline models (I<sub>AEA</sub> ≈ 0.06). By providing a verifiable standard for auditing agentic systems, these findings establish a rigorous path toward autonomous manufacturing where explainability and formal supervision foster operational resilience.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"287 ","pages":"Pages 25-32"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148981007","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Dynamic Ensemble Option Pricing: A Neural-Enhanced Multi-Model Framework 动态集合期权定价：一个神经增强的多模型框架

Procedia Computer Science Pub Date : 2026-01-01 Epub Date: 2026-09-18 DOI: 10.1016/j.procs.2026.08.051

Jinhao Liu, Xiyue Wang, Jingyu Li

{"title":"Dynamic Ensemble Option Pricing: A Neural-Enhanced Multi-Model Framework","authors":"Jinhao Liu, Xiyue Wang, Jingyu Li","doi":"10.1016/j.procs.2026.08.051","DOIUrl":"10.1016/j.procs.2026.08.051","url":null,"abstract":"<div><div>This paper presents a dynamic ensemble framework for equity option pricing that integrates three complementary models, including a neural-PINN SABR, a neural local-volatility surface, and a neural-enhanced Heston hybrid, through an adaptive weighting mechanism. Rather than selecting a single model, our approach employs a dedicated neural network that analyzes real-time market features (moneyness, time-to-expiry, volatility characteristics) to generate context-aware weights, allowing the ensemble to emphasize the most appropriate models for each pricing scenario. Trained on QQQ options data from 2020-2022 using two-stage optimization, the framework achieves a improvement in RMSE over the best individual model, with R² increasing. The system automatically adapts to different market regimes, maintains interpretability through explicit weight assignments, and reduces model risk by averaging across methodologies. This work demonstrates how machine learning can enhance traditional quantitative finance by shifting from model selection toward dynamic model management, with applications in volatility calibration, risk management, and high-frequency trading.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"287 ","pages":"Pages 295-302"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148981109","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

The impact of executives’ social media disclosure on pricing efficiency of the capital market 高管社交媒体披露对资本市场定价效率的影响

Procedia Computer Science Pub Date : 2026-01-01 Epub Date: 2026-09-18 DOI: 10.1016/j.procs.2026.08.053

Mingchuan Zhong, Xin Chen, Kedi Wang, Yanzhen Yao

{"title":"The impact of executives’ social media disclosure on pricing efficiency of the capital market","authors":"Mingchuan Zhong, Xin Chen, Kedi Wang, Yanzhen Yao","doi":"10.1016/j.procs.2026.08.053","DOIUrl":"10.1016/j.procs.2026.08.053","url":null,"abstract":"<div><div>With the rapid expansion of the digital economy, social media platforms like Weibo have become crucial frontiers for interaction between corporate executives and investors. This study investigates Shanghai and Shenzhen A-share listed companies from 2010 to 2025. By utilizing Python-based web-crawling techniques to collect post data from verified personal executive accounts, this paper empirically examines the impact of executive social media disclosure on capital market pricing efficiency. The results indicate that executives’ social media presence significantly reduces stock price synchronicity, thereby enhancing capital market pricing efficiency. This finding supports the information effect hypothesis, suggesting that executives disclose firm-specific information through social media, which effectively mitigates information asymmetry between firms and investors. Consequently, stock prices are enabled to more accurately reflect the intrinsic and independent value of the enterprise. Endogeneity tests further confirm the robustness of these conclusions after controlling for reverse causality. These findings not only enrich the literature on executive behavior and asset pricing but also offer references for regulatory authorities to standardize disclosure norms in the new media era and provide practical insights for investors leveraging unstructured data for decision-making.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"287 ","pages":"Pages 311-318"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148981111","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A stacking ensemble learning approach for scientific data classification 科学数据分类的堆叠集成学习方法

Procedia Computer Science Pub Date : 2026-01-01 Epub Date: 2026-09-18 DOI: 10.1016/j.procs.2026.08.048

Jianjun Yu, Shaowei Qin, Jingheng Zhang, Fangyu Liao, Jun Hao

{"title":"A stacking ensemble learning approach for scientific data classification","authors":"Jianjun Yu, Shaowei Qin, Jingheng Zhang, Fangyu Liao, Jun Hao","doi":"10.1016/j.procs.2026.08.048","DOIUrl":"10.1016/j.procs.2026.08.048","url":null,"abstract":"<div><div>To address the limitations of individual models in adaptive performance and the challenges of utilizing high-dimensional, sparse text features in automated scientific data classification, this paper proposes an approach based on an improved Stacking ensemble learning framework. Using dataset names and keywords as inputs, the method constructs lightweight text features via TF-IDF, bypassing the need for complex feature engineering. The experimental evaluation was conducted on a real-world scientific dataset comprising 4,505 samples across five disciplines. The performance of the proposed stacking model was benchmarked against four individual models and two simple ensemble methods (soft voting and weighted average). The results indicate that SVC exhibits the best adaptability to high-dimensional, sparse short-text features. Furthermore, ensemble models consistently outperformed individual models. Notably, the effectiveness of the stacking ensemble is determined by the quality and complementarity of the base models rather than their quantity. The proposed Stacking model achieved the highest classification performance, reaching an accuracy of 0.9024 on the test set. This lightweight, efficient, and stable method provides robust technical support for the automated classification and hierarchical governance of scientific data.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"287 ","pages":"Pages 276-281"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148981114","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Bibliometric Review of Artificial Intelligence in Green Marketing: A CiteSpace-Based Analysis 人工智能在绿色营销中的文献计量研究：基于citspace的分析

Procedia Computer Science Pub Date : 2026-01-01 Epub Date: 2026-09-18 DOI: 10.1016/j.procs.2026.08.057

Linrui Song

引用次数: 0

Visual Clickbait’s Influence on User Click Intention: The Mediating Roles of Curiosity and Perceived Credibility 视觉标题党对用户点击意图的影响：好奇心和感知可信度的中介作用

Procedia Computer Science Pub Date : 2026-01-01 Epub Date: 2026-09-18 DOI: 10.1016/j.procs.2026.08.059

Lijia Qian, Siyu Huang, Lu Huang

{"title":"Visual Clickbait’s Influence on User Click Intention: The Mediating Roles of Curiosity and Perceived Credibility","authors":"Lijia Qian, Siyu Huang, Lu Huang","doi":"10.1016/j.procs.2026.08.059","DOIUrl":"10.1016/j.procs.2026.08.059","url":null,"abstract":"<div><div>As social media shifts from text-centric to visual-centric paradigms, users face severe information overload. To compete for scarce attention, \"Visual Clickbait\"—characterized by high-impact cover images—has become increasingly prevalent. While existing literature predominantly examines the impact of textual headlines, the mechanism of cover images as primary visual cues remains underexplored. Grounded in Cognitive-Affective Theory, this study investigates the influence of visual clickbait on click intention and its underlying psychological mechanisms. Empirical results indicate that visual clickbait significantly enhances user click intention. Mediation analysis reveals a competitive mediation mechanism: although visual clickbait exerts a suppressive effect by lowering perceived credibility, it more strongly promotes behavior by stimulating curiosity, resulting in a net positive impact. Furthermore, image-title congruence serves as a moderator. Counter-intuitively, the promotional effect of visual clickbait on click intention is amplified under low congruence conditions. This research extends the theoretical framework of clickbait from the textual to the visual domain, offering theoretical insights for social media content governance and algorithmic optimization.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"287 ","pages":"Pages 359-366"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148981117","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

WheatLSNet-v2: A Local Enhancement and Auxiliary Supervision Based Method for Single-Kernel Wheat Defect Classification WheatLSNet-v2：一种基于局部增强和辅助监督的小麦单粒缺陷分类方法

Procedia Computer Science Pub Date : 2026-01-01 Epub Date: 2026-09-18 DOI: 10.1016/j.procs.2026.08.065

Xiang Chen, Yang Lu, Jiacheng Wan, Lei Zhang, Xiaoqiang Chen, Bo Mao

{"title":"WheatLSNet-v2: A Local Enhancement and Auxiliary Supervision Based Method for Single-Kernel Wheat Defect Classification","authors":"Xiang Chen, Yang Lu, Jiacheng Wan, Lei Zhang, Xiaoqiang Chen, Bo Mao","doi":"10.1016/j.procs.2026.08.065","DOIUrl":"10.1016/j.procs.2026.08.065","url":null,"abstract":"<div><div>Single-kernel wheat defect classification is a typical fine-grained visual recognition task in which subtle color differences, small local defect regions, and high inter-class similarity make accurate classification difficult. To address these challenges, this paper proposes WheatLSNet-v2, a single-kernel wheat defect classification model based on local enhancement and auxiliary supervision. Built upon the ResNet50 backbone, the proposed model introduces an HSV auxiliary branch to enhance the representation of subtle color anomalies, a local enhancement module to strengthen the response to small defect-sensitive regions, and a global-local collaborative fusion mechanism to improve the complementarity between high-level semantic features and local discriminative cues. An auxiliary classification head is further attached to the local branch during training to explicitly improve the category discriminability of intermediate local features. Experimental results show that the final model achieves 93.87% accuracy and 93.64% Macro-F1 on the validation set, outperforming the baseline ResNet50 by 3.07 and 3.12 percentage points, respectively. Comparative and ablation experiments further demonstrate that the HSV auxiliary branch and auxiliary supervision strategy both contribute positively to performance improvement, while the Shape Prior module does not provide stable gains and is therefore excluded from the final model. The proposed method effectively improves fine-grained defect recognition for single-kernel wheat images and shows practical potential for automated grain quality inspection.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"287 ","pages":"Pages 413-420"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148981166","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0