Decision Support Systems最新文献_第7页

A field study on the impact of the counter ad-blocking wall strategy on user engagement 反广告拦截墙策略对用户粘性影响的实地研究

IF 6.8 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Decision Support Systems

Pub Date : 2025-11-01 Epub Date: 2025-09-02 DOI: 10.1016/j.dss.2025.114525

Michael K. Chen , Shuai Zhao , Cristian Borcea , Yi Chen

Ad-blocking tools prevent ads from being shown to web users. Their increasingly widespread usage poses an existential risk to online publishers who provide free content and rely on display ads for revenue. Studies on counter ad-blocking strategies taken by publishers are limited, especially with regard to how these strategies affect user engagement, thus posing additional uncertainties to the selection of a suitable counter ad-blocking strategy. Through a randomized field experiment with a large global publisher, our study seeks to understand how the two most common counter ad-blocking strategies, (i) Wall and (ii) Acceptable Ads Exchange (AAX), affect user engagement differently. Our results show that the Wall strategy causes a lower overall engagement compared to AAX, mainly due to users who refuse to whitelist and leave the website. Over time, the negative impact increases, albeit at a slower speed. Furthermore, heavier users, identified based on the amount of engagement in the pre-treatment period, are less affected by the Wall strategy than lighter users; instrumental users, who read for practical purposes, are less affected than entertainment users. Finally, the Wall strategy has a bigger negative impact on the engagement of popular and new articles, compared to niche and old articles, respectively, as observed by a longer tail in engagement distribution with respect to content. These results on the heterogeneous effects of counter ad-blocking strategies on engagement offer novel and important managerial implications on a publisher’s choice of counter ad-blocking strategy and editorial decisions.

广告拦截工具可以防止广告显示给网络用户。它们日益广泛的使用给那些提供免费内容、依靠展示广告获得收入的在线出版商带来了生存风险。关于发布商采取的反广告拦截策略的研究有限，特别是关于这些策略如何影响用户参与度的研究，从而给选择合适的反广告拦截策略带来了额外的不确定性。通过对一家大型全球发行商的随机实地实验，我们的研究旨在了解两种最常见的反广告拦截策略(1)Wall和(2)Acceptable Ads Exchange （AAX）对用户粘性的不同影响。我们的研究结果显示，与AAX相比，Wall策略导致的整体参与度较低，主要原因是用户拒绝加入白名单并离开网站。随着时间的推移，负面影响会增加，尽管速度会放缓。此外，根据前处理阶段的用户粘性确定的重度用户受“墙”策略的影响要小于轻度用户；以实用为目的的工具性阅读用户受影响要小于娱乐性阅读用户。最后，与小众文章和老文章相比，墙策略对流行文章和新文章的粘性有更大的负面影响，这可以从内容粘性分布的长尾中观察到。这些关于反广告拦截策略对用户粘性的异质性影响的结果，为出版商选择反广告拦截策略和编辑决策提供了新颖而重要的管理启示。

{"title":"A field study on the impact of the counter ad-blocking wall strategy on user engagement","authors":"Michael K. Chen , Shuai Zhao , Cristian Borcea , Yi Chen","doi":"10.1016/j.dss.2025.114525","DOIUrl":"10.1016/j.dss.2025.114525","url":null,"abstract":"<div><div>Ad-blocking tools prevent ads from being shown to web users. Their increasingly widespread usage poses an existential risk to online publishers who provide free content and rely on display ads for revenue. Studies on counter ad-blocking strategies taken by publishers are limited, especially with regard to how these strategies affect user engagement, thus posing additional uncertainties to the selection of a suitable counter ad-blocking strategy. Through a randomized field experiment with a large global publisher, our study seeks to understand how the two most common counter ad-blocking strategies, (i) Wall and (ii) Acceptable Ads Exchange (AAX), affect user engagement differently. Our results show that the Wall strategy causes a lower overall engagement compared to AAX, mainly due to users who refuse to whitelist and leave the website. Over time, the negative impact increases, albeit at a slower speed. Furthermore, heavier users, identified based on the amount of engagement in the pre-treatment period, are less affected by the Wall strategy than lighter users; instrumental users, who read for practical purposes, are less affected than entertainment users. Finally, the Wall strategy has a bigger negative impact on the engagement of popular and new articles, compared to niche and old articles, respectively, as observed by a longer tail in engagement distribution with respect to content. These results on the heterogeneous effects of counter ad-blocking strategies on engagement offer novel and important managerial implications on a publisher’s choice of counter ad-blocking strategy and editorial decisions.</div></div>","PeriodicalId":55181,"journal":{"name":"Decision Support Systems","volume":"198 ","pages":"Article 114525"},"PeriodicalIF":6.8,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145097010","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}

引用次数: 0

Modeling hybrid firm relationships with graph neural networks for stock investment decisions 基于图神经网络的混合企业关系股票投资决策建模

IF 6.8 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Decision Support Systems

Pub Date : 2025-11-01 Epub Date: 2025-09-07 DOI: 10.1016/j.dss.2025.114528

Yang Du , Biao Li , Zhichen Lu , Gang Kou

The highly volatile nature of the stock market makes predicting data patterns challenging. Significant efforts have been dedicated to modeling complex stock correlations to improve stock return forecasting and support better investor decision-making. Although various predefined intrinsic associations and learned implicit graph structures have been discovered, they have limitations in fully exploring and leveraging both types of graph information. In this paper, we proposed a Hybrid Structure-aware Graph Neural Network (HSGNN) framework. Unlike models that rely solely on predefined or learned graphs, HSGNN utilizes money-flow graphs to complementarily learn implicit graph structures and applies sparse supply-chain graphs to jointly enhance stock return forecasting. Extensive experiments on real stock benchmarks demonstrate our proposed HSGNN outperforms various state-of-the-art forecasting methods, offering a robust decision-support system for financial stakeholders.

股市的高度波动性使得预测数据模式具有挑战性。大量的研究致力于建立复杂的股票相关性模型，以改善股票收益预测并支持更好的投资者决策。尽管已经发现了各种预定义的内在关联和习得的隐式图结构，但它们在充分探索和利用这两种类型的图信息方面存在局限性。本文提出了一种混合结构感知图神经网络（HSGNN）框架。与单纯依赖预定义图或学习图的模型不同，HSGNN利用现金流图互补学习隐式图结构，并应用稀疏供应链图共同增强股票收益预测。在真实股票基准上的大量实验表明，我们提出的HSGNN优于各种最先进的预测方法，为金融利益相关者提供了强大的决策支持系统。

引用次数: 0

Emotion aware session based news recommender systems 基于情感感知会话的新闻推荐系统

IF 6.8 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Decision Support Systems

Pub Date : 2025-11-01 Epub Date: 2025-09-15 DOI: 10.1016/j.dss.2025.114540

Benjamin Gundersen , Saikishore Kalloori , Abhishek Srivastava

News recommender systems are decision support systems that exploit user-article interactions over a short duration of time to discover users’ interests and predict unseen news articles to generate a ranking of news articles that are relevant and interesting. In the news recommendation scenario, the relevance of articles decays quickly, and fresh articles are generated daily. Session based models are proposed using time-aware approaches to exploit interactions sequentially. Prior news recommender systems do not consider emotional information expressed in news articles within sessions for recommendations. Emotions play a key role in supporting decision-making and emotionally charged headlines can evoke curiosity or urgency, prompting users to click on certain articles. This paper presents an innovative decision support system for session based news recommendation, using expressed emotions from news articles, such as expressed in the title, abstract, and text, to improve user decision-making. We introduce a novel methodology that incorporates expressed emotions into three session based news recommendation models. Our results demonstrate that expressed emotion carries valuable information to improve session based news recommenders on various ranking metrics significantly and proved especially beneficial in scenarios with limited user interaction history, addressing the cold-start problem. The results show significant improvements in ranking metrics, emphasizing the utility of emotional features for dynamic decision-making support.

新闻推荐系统是一种决策支持系统，它利用用户与文章在短时间内的交互来发现用户的兴趣，并预测未见过的新闻文章，从而生成相关和有趣的新闻文章排名。在新闻推荐场景中，文章的相关性衰减很快，每天都有新鲜的文章产生。提出了基于会话的模型，使用时间感知方法来顺序地利用交互。以前的新闻推荐系统不考虑新闻文章中表达的情感信息。情绪在支持决策方面起着关键作用，充满情绪的标题可以唤起好奇心或紧迫感，促使用户点击某些文章。本文提出了一种创新的基于会话的新闻推荐决策支持系统，利用新闻文章中表达的情感，如标题、摘要和文本中表达的情感，来提高用户的决策。我们引入了一种新的方法，将表达的情感融入到三个基于会话的新闻推荐模型中。我们的研究结果表明，表达情感携带有价值的信息，可以显著提高基于会话的新闻推荐器在各种排名指标上的表现，并且在用户交互历史有限的情况下被证明特别有益，解决了冷启动问题。结果显示，排名指标有显著改善，强调了动态决策支持的情感特征的效用。

{"title":"Emotion aware session based news recommender systems","authors":"Benjamin Gundersen , Saikishore Kalloori , Abhishek Srivastava","doi":"10.1016/j.dss.2025.114540","DOIUrl":"10.1016/j.dss.2025.114540","url":null,"abstract":"<div><div>News recommender systems are decision support systems that exploit user-article interactions over a short duration of time to discover users’ interests and predict unseen news articles to generate a ranking of news articles that are relevant and interesting. In the news recommendation scenario, the relevance of articles decays quickly, and fresh articles are generated daily. Session based models are proposed using time-aware approaches to exploit interactions sequentially. Prior news recommender systems do not consider emotional information expressed in news articles within sessions for recommendations. Emotions play a key role in supporting decision-making and emotionally charged headlines can evoke curiosity or urgency, prompting users to click on certain articles. This paper presents an innovative decision support system for session based news recommendation, using expressed emotions from news articles, such as expressed in the title, abstract, and text, to improve user decision-making. We introduce a novel methodology that incorporates expressed emotions into three session based news recommendation models. Our results demonstrate that expressed emotion carries valuable information to improve session based news recommenders on various ranking metrics significantly and proved especially beneficial in scenarios with limited user interaction history, addressing the cold-start problem. The results show significant improvements in ranking metrics, emphasizing the utility of emotional features for dynamic decision-making support.</div></div>","PeriodicalId":55181,"journal":{"name":"Decision Support Systems","volume":"198 ","pages":"Article 114540"},"PeriodicalIF":6.8,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145120986","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}

引用次数: 0

Enhancing cybersecurity risk assessment using temporal knowledge graph-based explainable decision support system 基于时态知识图的可解释决策支持系统增强网络安全风险评估

IF 6.8 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Decision Support Systems

Pub Date : 2025-11-01 Epub Date: 2025-09-13 DOI: 10.1016/j.dss.2025.114526

Subhajit Bag , Sobhan Sarkar , Indranil Bose

Assessing cybersecurity policies is crucial for any organization to combat evolving cyber threats. The absence of a comprehensive dataset has prevented previous studies from analyzing the risk of organizations’ cybersecurity policies. Past studies have not considered temporal information in the policies. Analysis of cybersecurity policies using attention mechanism requires automated determination of optimal number of attention units which remains unaddressed. Moreover, absence of interpretation in cybersecurity studies creates a barrier to understanding policy vulnerabilities and developing targeted solutions. To address these challenges, we develop a decision support system which (i) enhances risk classification of organization’s cybersecurity policies, (ii) develops a comprehensive cybersecurity policy dataset from the websites of 190 companies, transformed into a knowledge graph to capture entity relationships among various policies, (iii) integrates temporal information into the knowledge graph by incorporating time stamps from event sequences in cyberattack information, (iv) develops Explainable Factor Analysis based Multi-Head Attention mechanism, which automates the determination of the optimal number of attention units and optimizes data allocation across attention units using factor analysis, and (v) utilizes attention heatmaps and shapley values for interpretability. Our cybersecurity policy dataset is used as a case study with four benchmark datasets for further validation. Results reveal that our model outperforms the other state-of-the-art, achieving an 87.78%

F_{1}

score, followed by robustness checking and statistical significance testing. Finally, Shapley values are used to interpret the model’s output to identify vulnerabilities within the organizational policies, providing crucial insights enabling decision-makers to enhance their cybersecurity policies and mitigate potential threats.

评估网络安全政策对于任何组织应对不断变化的网络威胁都至关重要。由于缺乏全面的数据集，以前的研究无法分析组织网络安全政策的风险。过去的研究没有考虑到政策中的时间信息。使用注意力机制分析网络安全策略需要自动确定最优的注意力单元数量，这一问题仍未得到解决。此外，网络安全研究中缺乏解释，这对理解政策漏洞和制定有针对性的解决方案造成了障碍。为了应对这些挑战，我们开发了一个决策支持系统，该系统(i)增强了组织网络安全政策的风险分类，（ii）从190家公司的网站开发了一个全面的网络安全政策数据集，转化为知识图谱，以捕获各种政策之间的实体关系，（iii）通过将网络攻击信息中事件序列的时间戳整合到知识图谱中，将时间信息集成到知识图谱中。（iv）开发基于可解释因子分析的多头注意机制，该机制自动确定注意单元的最佳数量，并使用因子分析优化注意单元之间的数据分配，以及(v)利用注意热图和shapley值来实现可解释性。我们的网络安全策略数据集被用作四个基准数据集的案例研究，以进一步验证。结果表明，我们的模型优于其他先进技术，达到87.78%的F1得分，随后进行稳健性检验和统计显著性检验。最后，Shapley值用于解释模型的输出，以识别组织策略中的漏洞，提供关键的见解，使决策者能够增强其网络安全策略并减轻潜在威胁。

{"title":"Enhancing cybersecurity risk assessment using temporal knowledge graph-based explainable decision support system","authors":"Subhajit Bag , Sobhan Sarkar , Indranil Bose","doi":"10.1016/j.dss.2025.114526","DOIUrl":"10.1016/j.dss.2025.114526","url":null,"abstract":"<div><div>Assessing cybersecurity policies is crucial for any organization to combat evolving cyber threats. The absence of a comprehensive dataset has prevented previous studies from analyzing the risk of organizations’ cybersecurity policies. Past studies have not considered temporal information in the policies. Analysis of cybersecurity policies using attention mechanism requires automated determination of optimal number of attention units which remains unaddressed. Moreover, absence of interpretation in cybersecurity studies creates a barrier to understanding policy vulnerabilities and developing targeted solutions. To address these challenges, we develop a decision support system which (i) enhances risk classification of organization’s cybersecurity policies, (ii) develops a comprehensive cybersecurity policy dataset from the websites of 190 companies, transformed into a knowledge graph to capture entity relationships among various policies, (iii) integrates temporal information into the knowledge graph by incorporating time stamps from event sequences in cyberattack information, (iv) develops Explainable Factor Analysis based Multi-Head Attention mechanism, which automates the determination of the optimal number of attention units and optimizes data allocation across attention units using factor analysis, and (v) utilizes attention heatmaps and shapley values for interpretability. Our cybersecurity policy dataset is used as a case study with four benchmark datasets for further validation. Results reveal that our model outperforms the other state-of-the-art, achieving an 87.78% <span><math><msub><mrow><mi>F</mi></mrow><mrow><mn>1</mn></mrow></msub></math></span> score, followed by robustness checking and statistical significance testing. Finally, Shapley values are used to interpret the model’s output to identify vulnerabilities within the organizational policies, providing crucial insights enabling decision-makers to enhance their cybersecurity policies and mitigate potential threats.</div></div>","PeriodicalId":55181,"journal":{"name":"Decision Support Systems","volume":"198 ","pages":"Article 114526"},"PeriodicalIF":6.8,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145097012","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}

引用次数: 0

Human-Robo-advisor collaboration in decision-making: Evidence from a multiphase mixed methods experimental study 人-机器人顾问在决策中的协作：来自多阶段混合方法实验研究的证据

IF 6.8 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Decision Support Systems

Pub Date : 2025-11-01 Epub Date: 2025-09-12 DOI: 10.1016/j.dss.2025.114541

Hasan Mahmud , Najmul Islam , Satish Krishnan

Robo-advisors (RAs) are cost-effective, bias-resistant alternatives to human financial advisors, yet adoption remains limited. While prior research has examined user interactions with RAs, less is known about how individuals interpret RA roles and integrate their advice into decision-making. To address this gap, this study employs a multiphase mixed methods design integrating a behavioral experiment (N = 334), thematic analysis, and follow-up quantitative testing. Findings suggest that people tend to rely on RAs, with reliance shaped by information about RA performance and the framing of advice as gains or losses. Thematic analysis reveals three RA roles in decision-making and four user types, each reflecting distinct patterns of advice integration. In addition, a 2 × 2 typology categorizes antecedents of acceptance into enablers and inhibitors at both the individual and algorithmic levels. By combining behavioral, interpretive, and confirmatory evidence, this study advances understanding of human–RA collaboration and provides actionable insights for designing more trustworthy and adaptive RA systems.

机器人财务顾问（RAs）是人类财务顾问的成本效益高、抗偏见的替代品，但采用仍然有限。虽然先前的研究已经检查了用户与RA的交互，但对于个人如何解释RA的角色并将其建议整合到决策中，我们知之甚少。为了弥补这一空白，本研究采用了多阶段混合方法设计，包括行为实验（N = 334）、主题分析和后续定量测试。研究结果表明，人们倾向于依赖RA，这种依赖是由RA表现的信息和建议的收益或损失框架决定的。专题分析揭示了RA在决策中的三种角色和四种用户类型，每种类型都反映了不同的建议集成模式。此外，一个2 × 2的类型学将接受的前因在个体和算法层面上分为促进因素和抑制因素。通过结合行为、解释和验证性证据，本研究促进了对人类RA协作的理解，并为设计更值得信赖和自适应的RA系统提供了可操作的见解。

{"title":"Human-Robo-advisor collaboration in decision-making: Evidence from a multiphase mixed methods experimental study","authors":"Hasan Mahmud , Najmul Islam , Satish Krishnan","doi":"10.1016/j.dss.2025.114541","DOIUrl":"10.1016/j.dss.2025.114541","url":null,"abstract":"<div><div>Robo-advisors (RAs) are cost-effective, bias-resistant alternatives to human financial advisors, yet adoption remains limited. While prior research has examined user interactions with RAs, less is known about how individuals interpret RA roles and integrate their advice into decision-making. To address this gap, this study employs a multiphase mixed methods design integrating a behavioral experiment (<em>N</em> = 334), thematic analysis, and follow-up quantitative testing. Findings suggest that people tend to rely on RAs, with reliance shaped by information about RA performance and the framing of advice as gains or losses. Thematic analysis reveals three RA roles in decision-making and four user types, each reflecting distinct patterns of advice integration. In addition, a 2 × 2 typology categorizes antecedents of acceptance into enablers and inhibitors at both the individual and algorithmic levels. By combining behavioral, interpretive, and confirmatory evidence, this study advances understanding of human–RA collaboration and provides actionable insights for designing more trustworthy and adaptive RA systems.</div></div>","PeriodicalId":55181,"journal":{"name":"Decision Support Systems","volume":"198 ","pages":"Article 114541"},"PeriodicalIF":6.8,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145221256","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}

引用次数: 0

Corporate credit scoring method based on unlabeled data and multi-source data 基于无标记数据和多源数据的企业信用评分方法

IF 6.8 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Decision Support Systems

Pub Date : 2025-11-01 Epub Date: 2025-09-16 DOI: 10.1016/j.dss.2025.114543

Yunhong Xu, Yitong Chen, Li Sun, Yu Chen

Unlabeled data and multi-source data provide unprecedented opportunities for the financial industry to improve credit scoring accuracy. When utilizing unlabeled data, existing credit scoring methods often suffer from unreliability issues due to improper clustering or the introduction of noise when predicting labels. When utilizing multi-source data, existing credit scoring methods based on federated learning frameworks fail to tailor models for different data distributions of different data sources due to the limitations of relying on a single global model. Moreover, recent studies have explored the individual value of unlabeled data and multi-source data, but they often fail to utilize both. To address these issues, we propose UMDCS (Unlabeled and Multi-Source data Driven Credit Scoring), a self-supervised credit scoring method that utilizes both unlabeled and multi-source data simultaneously. To utilize unlabeled data, we propose a novel sample masking function to generate pseudo-labels for unlabeled data and pre-train the encoder using the pretext tasks. To utilize multi-source data, we employ a horizontal federated learning framework to aggregate local encoders into a global model while preserving data privacy. The global encoder is concatenated with personalized predictors to form personalized credit scoring models for each data source. Five experiments and statistical significance tests show that UMDCS outperforms other baseline methods.

无标签数据和多源数据为金融业提高信用评分准确性提供了前所未有的机遇。当使用未标记的数据时，现有的信用评分方法往往由于聚类不当或在预测标签时引入噪声而存在不可靠性问题。现有的基于联邦学习框架的信用评分方法在使用多源数据时，由于依赖单一全局模型的限制，无法针对不同数据源的不同数据分布定制模型。此外，最近的研究已经探索了未标记数据和多源数据的个体价值，但往往未能充分利用两者。为了解决这些问题，我们提出了UMDCS（未标记和多源数据驱动信用评分），这是一种同时利用未标记和多源数据的自我监督信用评分方法。为了利用未标记数据，我们提出了一种新的样本掩蔽函数来为未标记数据生成伪标签，并使用借口任务对编码器进行预训练。为了利用多源数据，我们采用水平联邦学习框架将本地编码器聚合到全局模型中，同时保护数据隐私。全局编码器与个性化预测器相连接，形成每个数据源的个性化信用评分模型。五个实验和统计显著性检验表明，UMDCS优于其他基线方法。

{"title":"Corporate credit scoring method based on unlabeled data and multi-source data","authors":"Yunhong Xu, Yitong Chen, Li Sun, Yu Chen","doi":"10.1016/j.dss.2025.114543","DOIUrl":"10.1016/j.dss.2025.114543","url":null,"abstract":"<div><div>Unlabeled data and multi-source data provide unprecedented opportunities for the financial industry to improve credit scoring accuracy. When utilizing unlabeled data, existing credit scoring methods often suffer from unreliability issues due to improper clustering or the introduction of noise when predicting labels. When utilizing multi-source data, existing credit scoring methods based on federated learning frameworks fail to tailor models for different data distributions of different data sources due to the limitations of relying on a single global model. Moreover, recent studies have explored the individual value of unlabeled data and multi-source data, but they often fail to utilize both. To address these issues, we propose UMDCS (Unlabeled and Multi-Source data Driven Credit Scoring), a self-supervised credit scoring method that utilizes both unlabeled and multi-source data simultaneously. To utilize unlabeled data, we propose a novel sample masking function to generate pseudo-labels for unlabeled data and pre-train the encoder using the pretext tasks. To utilize multi-source data, we employ a horizontal federated learning framework to aggregate local encoders into a global model while preserving data privacy. The global encoder is concatenated with personalized predictors to form personalized credit scoring models for each data source. Five experiments and statistical significance tests show that UMDCS outperforms other baseline methods.</div></div>","PeriodicalId":55181,"journal":{"name":"Decision Support Systems","volume":"198 ","pages":"Article 114543"},"PeriodicalIF":6.8,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145120987","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}

引用次数: 0

A friend or a foe? The effect of generative artificial intelligence on creator contributions on original work sharing platforms 朋友还是敌人？生成式人工智能对原创作品分享平台创作者贡献的影响

IF 6.8 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Decision Support Systems

Pub Date : 2025-10-01 Epub Date: 2025-08-07 DOI: 10.1016/j.dss.2025.114513

Shan Liu , Wenxuan Hu , Baojun Gao

While generative artificial intelligence (GAI) is increasingly used to create content, it is often criticized for collecting and training private data and induces potential copy infringement issue. This dilemma leaves a question of whether GAI increases or decreases creators' work sharing. Drawn on protection motivation theory, this study examines how the launch of a GAI system affects creators' contributions on an original work sharing platform. We discover that GAI poses a threat to drawing-category creators, leading to a significant crowding-out effect on their contributions. Specifically, compared with that of non-drawing-category creators, the work sharing of drawing-category creators decreases by 19.64 % and 14.29 % within a short period after the launch and removal of the GAI system, respectively. We discover that creators' protective behavior is driven by GAI-related copyright infringement. Compared with creators without copyright protection, those with copyright protection are more inclined to cease contributions or even leave the platform. We further find that among copyright-protected creators, top creators, evidenced by their acquisition of a large number of supporters or platform honor titles, exhibit more pronounced responses to protect their works due to their higher coping efficacy. Notably, this threat reduces creators' sharing behavior or even lead to their exit from the platform. Nevertheless, such reduction is likely to gradually recover once the threat subsides. Overall, our findings have important implications for whether and how platform managers adopt GAI systems, especially in an original work sharing context.

虽然生成式人工智能（GAI）越来越多地用于创建内容，但它经常因收集和训练私人数据而受到批评，并引发潜在的复制侵权问题。这种困境带来了一个问题，即GAI是增加还是减少了创作者的作品分享。本研究运用保护动机理论，考察GAI制度的推出对创作者在原创作品分享平台上的贡献有何影响。我们发现GAI对绘画类创作者构成了威胁，导致他们的贡献受到明显的挤出效应。具体而言，与非绘图类创作者相比，在GAI系统上线和下线后的短时间内，绘图类创作者的作品分享率分别下降了19.64%和14.29%。我们发现创作者的保护行为是由与人工智能相关的版权侵权驱动的。与没有版权保护的创作者相比，有版权保护的创作者更倾向于停止创作甚至离开平台。我们进一步发现，在受版权保护的创作者中，获得大量支持者或平台荣誉称号的顶级创作者由于其更高的应对效能而表现出更明显的保护作品的反应。值得注意的是，这种威胁减少了创作者的分享行为，甚至导致他们退出平台。然而，一旦威胁消退，这种减少可能会逐渐恢复。总的来说，我们的发现对于平台管理者是否以及如何采用GAI系统具有重要意义，特别是在原始工作共享环境中。

{"title":"A friend or a foe? The effect of generative artificial intelligence on creator contributions on original work sharing platforms","authors":"Shan Liu , Wenxuan Hu , Baojun Gao","doi":"10.1016/j.dss.2025.114513","DOIUrl":"10.1016/j.dss.2025.114513","url":null,"abstract":"<div><div>While generative artificial intelligence (GAI) is increasingly used to create content, it is often criticized for collecting and training private data and induces potential copy infringement issue. This dilemma leaves a question of whether GAI increases or decreases creators' work sharing. Drawn on protection motivation theory, this study examines how the launch of a GAI system affects creators' contributions on an original work sharing platform. We discover that GAI poses a threat to drawing-category creators, leading to a significant crowding-out effect on their contributions. Specifically, compared with that of non-drawing-category creators, the work sharing of drawing-category creators decreases by 19.64 % and 14.29 % within a short period after the launch and removal of the GAI system, respectively. We discover that creators' protective behavior is driven by GAI-related copyright infringement. Compared with creators without copyright protection, those with copyright protection are more inclined to cease contributions or even leave the platform. We further find that among copyright-protected creators, top creators, evidenced by their acquisition of a large number of supporters or platform honor titles, exhibit more pronounced responses to protect their works due to their higher coping efficacy. Notably, this threat reduces creators' sharing behavior or even lead to their exit from the platform. Nevertheless, such reduction is likely to gradually recover once the threat subsides. Overall, our findings have important implications for whether and how platform managers adopt GAI systems, especially in an original work sharing context.</div></div>","PeriodicalId":55181,"journal":{"name":"Decision Support Systems","volume":"197 ","pages":"Article 114513"},"PeriodicalIF":6.8,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144828542","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}

引用次数: 0

Sparse-enhanced additive interaction neural network for interpretable credit decision 可解释信用决策的稀疏增强加性交互神经网络

IF 6.8 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Decision Support Systems

Pub Date : 2025-10-01 Epub Date: 2025-07-22 DOI: 10.1016/j.dss.2025.114507

Xingyu Lan , Hong Fan , Wanan Liu , Meng Xia , Kai Guo

Intelligent credit decision systems are crucial for financial institutions’ risk management, aiming to mitigate credit risk. While deep learning models offer high predictive accuracy, their opacity hinders decision support. Neural Additive Models (NAMs) offer feature-level interpretability but fail to capture complex interactions among credit risk factors. To enhance both accuracy and interpretability, we propose the Sparse-Enhanced Additive Interaction Neural Network (SAINTNet) for explainable credit scoring. SAINTNet advances NAM’s framework with dual-node additive modules and adaptive sparse feature selection, enabling autonomous feature learning. Leveraging entmax sparsity and optimized temperature settings, SAINTNet: (1) maintains interpretability, particularly for credit feature interactions; (2) achieves superior accuracy compared to black-box models. Experiments on four credit datasets demonstrate SAINTNet’s superior performance and systematic interpretability through global feature importance, local decision analysis, and interaction visualization, improving decision audits in high-risk credit scenarios.

智能信贷决策系统是金融机构风险管理的关键，其目的是降低信贷风险。虽然深度学习模型提供了很高的预测准确性，但它们的不透明性阻碍了决策支持。神经加性模型（NAMs）提供特征级的可解释性，但无法捕获信用风险因素之间复杂的相互作用。为了提高准确性和可解释性，我们提出了稀疏增强的加性交互神经网络（SAINTNet）用于可解释的信用评分。SAINTNet通过双节点加性模块和自适应稀疏特征选择改进了NAM框架，实现了自主特征学习。利用entmax稀疏性和优化的温度设置，SAINTNet:(1)保持可解释性，特别是对于信用特征交互；(2)与黑箱模型相比，精度更高。在四个信用数据集上进行的实验表明，SAINTNet通过全局特征重要性、局部决策分析和交互可视化等方法，具有卓越的性能和系统的可解释性，改善了高风险信用场景下的决策审计。

{"title":"Sparse-enhanced additive interaction neural network for interpretable credit decision","authors":"Xingyu Lan , Hong Fan , Wanan Liu , Meng Xia , Kai Guo","doi":"10.1016/j.dss.2025.114507","DOIUrl":"10.1016/j.dss.2025.114507","url":null,"abstract":"<div><div>Intelligent credit decision systems are crucial for financial institutions’ risk management, aiming to mitigate credit risk. While deep learning models offer high predictive accuracy, their opacity hinders decision support. Neural Additive Models (NAMs) offer feature-level interpretability but fail to capture complex interactions among credit risk factors. To enhance both accuracy and interpretability, we propose the Sparse-Enhanced Additive Interaction Neural Network (SAINTNet) for explainable credit scoring. SAINTNet advances NAM’s framework with dual-node additive modules and adaptive sparse feature selection, enabling autonomous feature learning. Leveraging entmax sparsity and optimized temperature settings, SAINTNet: (1) maintains interpretability, particularly for credit feature interactions; (2) achieves superior accuracy compared to black-box models. Experiments on four credit datasets demonstrate SAINTNet’s superior performance and systematic interpretability through global feature importance, local decision analysis, and interaction visualization, improving decision audits in high-risk credit scenarios.</div></div>","PeriodicalId":55181,"journal":{"name":"Decision Support Systems","volume":"197 ","pages":"Article 114507"},"PeriodicalIF":6.8,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144721917","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}

引用次数: 0

Boosting the underdogs: Unraveling how prevailing streamer visits drive revenue for emerging streamers on livestreaming entertainment platforms 推动弱者：揭示主流流媒体访问如何推动直播娱乐平台上新兴流媒体的收入

IF 6.8 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Decision Support Systems

Pub Date : 2025-10-01 Epub Date: 2025-07-25 DOI: 10.1016/j.dss.2025.114511

Huijing Guo , Xin Bao , Le Wang , Xin (Robert) Luo

Livestreaming entertainment (LSE) platforms have become increasingly popular for real-time social interaction. While high-status actors (prevailing streamers) attract large audiences, new streamers often struggle with visibility and earnings. This study examines how social capital transmission from high-status actors affect emerging streamers' live revenue, using Social Capital Theory and Arousal Theories as frameworks. We analyzed data from 52,010 emerging streamers over two weeks on a major LSE platform. The research shows that visits from established streamers significantly increase new streamers' revenue. This positive effect is notably stronger when new streamers have shown good past performance and belong to top guilds and visiting established streamers have strong performance records and actively interact during their visits. Our findings contribute to LSE platform research by highlighting the supportive role of established streamers. These insights can help platforms develop strategies to enhance platform vitality, diversify content, support emerging streamers' growth, and foster a more sustainable streaming ecosystem.

直播娱乐（LSE）平台在实时社交互动方面越来越受欢迎。虽然高地位的演员（流行的流媒体）吸引了大量的观众，但新的流媒体经常在知名度和收入方面挣扎。本研究以社会资本理论和激励理论为框架，探讨了社会资本传播对新兴流媒体直播收入的影响。我们在LSE的一个主要平台上分析了两周内来自52010个新兴流媒体的数据。研究表明，来自老牌主播的访问量显著增加了新主播的收入。当新的主播过去表现良好，并且属于顶级公会，并且访问的老牌主播有良好的表现记录并在访问期间积极互动时，这种积极的影响就会明显增强。我们的研究结果通过强调知名主播的支持作用，为伦敦政治经济学院的平台研究做出了贡献。这些见解可以帮助平台制定战略，增强平台活力，使内容多样化，支持新兴流媒体的发展，并培养一个更可持续的流媒体生态系统。

{"title":"Boosting the underdogs: Unraveling how prevailing streamer visits drive revenue for emerging streamers on livestreaming entertainment platforms","authors":"Huijing Guo , Xin Bao , Le Wang , Xin (Robert) Luo","doi":"10.1016/j.dss.2025.114511","DOIUrl":"10.1016/j.dss.2025.114511","url":null,"abstract":"<div><div>Livestreaming entertainment (LSE) platforms have become increasingly popular for real-time social interaction. While high-status actors (prevailing streamers) attract large audiences, new streamers often struggle with visibility and earnings. This study examines how social capital transmission from high-status actors affect emerging streamers' live revenue, using Social Capital Theory and Arousal Theories as frameworks. We analyzed data from 52,010 emerging streamers over two weeks on a major LSE platform. The research shows that visits from established streamers significantly increase new streamers' revenue. This positive effect is notably stronger when new streamers have shown good past performance and belong to top guilds and visiting established streamers have strong performance records and actively interact during their visits. Our findings contribute to LSE platform research by highlighting the supportive role of established streamers. These insights can help platforms develop strategies to enhance platform vitality, diversify content, support emerging streamers' growth, and foster a more sustainable streaming ecosystem.</div></div>","PeriodicalId":55181,"journal":{"name":"Decision Support Systems","volume":"197 ","pages":"Article 114511"},"PeriodicalIF":6.8,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144739453","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}

引用次数: 0

Blockchain-based token system for incentivizing peer review: A design science approach 基于区块链的激励同行评审的代币系统：一种设计科学方法

IF 6.8 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Decision Support Systems

Pub Date : 2025-10-01 Epub Date: 2025-07-31 DOI: 10.1016/j.dss.2025.114514

Chad Anderson , Pratiksha Shrestha , Suman Bhunia , Arthur Carvalho , Younghwa Lee

Peer review is an essential component of the evaluation and dissemination of new scientific knowledge. The peer review process can be viewed as a decision support framework relying on scholarly review systems, where decision-makers (editors) solicit input from experts (reviewers) to make editorial decisions on submitted manuscripts. Unfortunately, the challenges editors face in securing sufficient reviewers are well-documented, leading to prolonged review times and potentially diminished review quality. We explore and validate this trend through a literature review and interviews with scholars. We then employ a design science research methodology to design, develop, and evaluate potential incentive mechanisms to reverse that trend. In addition to proposing formal design principles that such mechanisms should follow, we suggest a concrete blockchain-based token system that enables editors to offer review incentives while enabling reviewers to flexibly utilize these incentives to meet their needs. We also explain how different types of tokens can be connected to practical submission and reward policies that journals may adopt. Our cost analysis, along with a survey-based field study and qualitative interviews with academics, highlight the effectiveness of our solution. Finally, we propose a formal design theory framework that designers of peer review systems can follow to create meaningful incentives to attract reviewers.

同行评议是评价和传播新科学知识的重要组成部分。同行评议过程可以被视为一个依赖学术评议系统的决策支持框架，其中决策者（编辑）征求专家（审稿人）的意见，以对提交的手稿做出编辑决定。不幸的是，编辑在确保足够的审稿人方面面临的挑战是充分记录的，这会导致审稿时间延长，并潜在地降低审稿质量。我们通过文献综述和对学者的访谈来探索和验证这一趋势。然后，我们采用设计科学研究方法来设计、开发和评估潜在的激励机制，以扭转这一趋势。除了提出这些机制应该遵循的正式设计原则外，我们还建议一个具体的基于区块链的令牌系统，使编辑能够提供审查激励，同时使审稿人能够灵活地利用这些激励来满足他们的需求。我们还解释了不同类型的代币如何与期刊可能采用的实际提交和奖励政策相关联。我们的成本分析，以及基于调查的实地研究和与学者的定性访谈，突出了我们解决方案的有效性。最后，我们提出了一个正式的设计理论框架，同行评议系统的设计者可以遵循这个框架来创建有意义的激励机制来吸引评议者。

{"title":"Blockchain-based token system for incentivizing peer review: A design science approach","authors":"Chad Anderson , Pratiksha Shrestha , Suman Bhunia , Arthur Carvalho , Younghwa Lee","doi":"10.1016/j.dss.2025.114514","DOIUrl":"10.1016/j.dss.2025.114514","url":null,"abstract":"<div><div>Peer review is an essential component of the evaluation and dissemination of new scientific knowledge. The peer review process can be viewed as a decision support framework relying on scholarly review systems, where decision-makers (editors) solicit input from experts (reviewers) to make editorial decisions on submitted manuscripts. Unfortunately, the challenges editors face in securing sufficient reviewers are well-documented, leading to prolonged review times and potentially diminished review quality. We explore and validate this trend through a literature review and interviews with scholars. We then employ a design science research methodology to design, develop, and evaluate potential incentive mechanisms to reverse that trend. In addition to proposing formal design principles that such mechanisms should follow, we suggest a concrete blockchain-based token system that enables editors to offer review incentives while enabling reviewers to flexibly utilize these incentives to meet their needs. We also explain how different types of tokens can be connected to practical submission and reward policies that journals may adopt. Our cost analysis, along with a survey-based field study and qualitative interviews with academics, highlight the effectiveness of our solution. Finally, we propose a formal design theory framework that designers of peer review systems can follow to create meaningful incentives to attract reviewers.</div></div>","PeriodicalId":55181,"journal":{"name":"Decision Support Systems","volume":"197 ","pages":"Article 114514"},"PeriodicalIF":6.8,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144757343","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}

引用次数: 0