Online Social Networks and Media最新文献_第4页

An explainable ensemble model for revealing the level of depression in social media by considering personality traits and sentiment polarity pattern 考虑人格特质和情绪极性模式的社交媒体抑郁水平可解释的集合模型

Q1 Social Sciences

Online Social Networks and Media

Pub Date : 2025-03-08 DOI: 10.1016/j.osnem.2025.100307

Gede Aditra Pradnyana , Wiwik Anggraeni , Eko Mulyanto Yuniarno , Mauridhi Hery Purnomo

Early detection of depression in mental health is crucial for better intervention. Social media has been extensively used to examine users’ behavior, motivating researchers to develop an automatic depression detection model. However, the accuracy and clarity of the reasons behind the detection results still need to be improved. Current research focuses primarily on syntactic and semantic information in user-posted texts, while other aspects of users’ psychological characteristics are often overlooked. Therefore, this study addresses the gap by proposing a novel model integrating personality traits and sentiment polarity patterns into an explainable ensemble model. Specifically, we developed two base learners for the averaged and meta-ensemble learning strategy. The first learner employed the Robustly Optimized BERT Pre-training Approach (RoBERTa). For the second learner, we combined the Random Forest and Bidirectional Long Short-Term Memory (RF-BiLSTM) methods to effectively handle the combination of personality traits and sequential information in sentiment polarity patterns. These additional features are obtained by performing domain adaptation for personality prediction and sentiment analysis using a lexicon-based model. Based on the experimental results, our ensemble model improved depression detection results by leveraging the strengths of each base learner. Our model advanced the state-of-the-art, outperforming existing models with an increase in accuracy and F1-score of 4.14% and 2.99%, respectively. The model successfully enhanced the interpretability of detection results, providing a more comprehensive understanding of the factors underlying depressive symptoms. This research highlights the potential of considering alternative additional features as a promising avenue for enhancing depression detection in social media.

在心理健康方面早期发现抑郁症对于更好的干预至关重要。社交媒体被广泛用于检查用户的行为，这促使研究人员开发了一种自动抑郁检测模型。然而，检测结果背后原因的准确性和清晰度仍有待提高。目前的研究主要集中在用户发布文本中的句法和语义信息，而用户心理特征的其他方面往往被忽视。因此，本研究提出了一种将人格特质和情感极性模式整合到一个可解释的集成模型中的新模型来解决这一空白。具体来说，我们为平均学习策略和元集成学习策略开发了两个基本学习器。第一个学习者采用鲁棒优化BERT预训练方法（RoBERTa）。对于第二个学习者，我们将随机森林和双向长短期记忆（RF-BiLSTM）方法结合起来，有效地处理了情感极性模式下人格特征和顺序信息的组合。这些额外的特征是通过使用基于词典的模型进行人格预测和情感分析的领域适应来获得的。在实验结果的基础上，我们的集成模型通过利用每个基学习器的优势来改善抑郁检测结果。我们的模型比现有的模型更先进，准确率和f1分数分别提高了4.14%和2.99%。该模型成功地提高了检测结果的可解释性，为抑郁症状的潜在因素提供了更全面的理解。这项研究强调了考虑替代附加功能作为增强社交媒体抑郁症检测的有希望的途径的潜力。

{"title":"An explainable ensemble model for revealing the level of depression in social media by considering personality traits and sentiment polarity pattern","authors":"Gede Aditra Pradnyana , Wiwik Anggraeni , Eko Mulyanto Yuniarno , Mauridhi Hery Purnomo","doi":"10.1016/j.osnem.2025.100307","DOIUrl":"10.1016/j.osnem.2025.100307","url":null,"abstract":"<div><div>Early detection of depression in mental health is crucial for better intervention. Social media has been extensively used to examine users’ behavior, motivating researchers to develop an automatic depression detection model. However, the accuracy and clarity of the reasons behind the detection results still need to be improved. Current research focuses primarily on syntactic and semantic information in user-posted texts, while other aspects of users’ psychological characteristics are often overlooked. Therefore, this study addresses the gap by proposing a novel model integrating personality traits and sentiment polarity patterns into an explainable ensemble model. Specifically, we developed two base learners for the averaged and meta-ensemble learning strategy. The first learner employed the Robustly Optimized BERT Pre-training Approach (RoBERTa). For the second learner, we combined the Random Forest and Bidirectional Long Short-Term Memory (RF-BiLSTM) methods to effectively handle the combination of personality traits and sequential information in sentiment polarity patterns. These additional features are obtained by performing domain adaptation for personality prediction and sentiment analysis using a lexicon-based model. Based on the experimental results, our ensemble model improved depression detection results by leveraging the strengths of each base learner. Our model advanced the state-of-the-art, outperforming existing models with an increase in accuracy and F1-score of 4.14% and 2.99%, respectively. The model successfully enhanced the interpretability of detection results, providing a more comprehensive understanding of the factors underlying depressive symptoms. This research highlights the potential of considering alternative additional features as a promising avenue for enhancing depression detection in social media.</div></div>","PeriodicalId":52228,"journal":{"name":"Online Social Networks and Media","volume":"46 ","pages":"Article 100307"},"PeriodicalIF":0.0,"publicationDate":"2025-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143579435","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Management of psychological emergency cases on social media: A hybrid approach combining knowledge graphs and graph neural networks 基于社交媒体的心理突发事件管理：知识图与图神经网络的混合方法

Q1 Social Sciences

Online Social Networks and Media

Pub Date : 2025-03-05 DOI: 10.1016/j.osnem.2025.100308

Mourad Ellouze , Sonda Rekik , Lamia Hadrich Belguith

The effects of psychological crises are evolving at an astounding rate nowadays, presenting a significant challenge for everyone involved in tracking these disorders. Therefore, we propose in this paper a hybrid approach based on linguistic processing and numerical techniques allowing to: (i) identify the presence of psychological emergencies among social network users by analyzing their textual production, (ii) determine the specific type of emergency case, (iii) elaborate a graph for each type of emergency, reflecting the different dimensions linked to the psychological emergency, allowing for a better diagnosis of the situation and providing an overall view of the crisis type, (iv) combine the separate graphs for each emergency to address the various semantic aspects. The work was accomplished using advanced language model techniques, knowledge graphs and neural network graphs. The combination of these techniques ensures that their advantages are leveraged while overcoming their limitations in terms of result generalization. The evaluation of different parts related to detecting the presence of psychological problems, predicting specific type of emergency cases, and detecting links between knowledge graphs was measured using the F-measure metric. The values derived from this measure, corresponding to the evaluation of these three tasks, are, respectively, 83%, 87% and 80%. For the evaluation of the elaboration of each graph related to specific type of emergency cases, this was accomplished using qualitative metric standards. The results obtained can be considered encouraging given the significant scale of our approach.

如今，心理危机的影响正以惊人的速度发展，对参与追踪这些疾病的每个人都提出了重大挑战。因此，我们在本文中提出了一种基于语言处理和数值技术的混合方法，允许：（一）通过分析社交网络用户的文本生成来确定心理紧急情况的存在；（二）确定紧急情况的具体类型；（三）为每种紧急情况制作一个图表，反映与心理紧急情况有关的不同维度，以便更好地诊断情况并提供危机类型的总体视图；（四）将每种紧急情况的单独图表结合起来，以解决各种语义方面的问题。这项工作是利用先进的语言模型技术、知识图和神经网络图来完成的。这些技术的组合确保了它们的优势被利用，同时克服了它们在结果泛化方面的局限性。对检测心理问题的存在、预测特定类型的紧急情况以及检测知识图之间的联系的不同部分的评估使用F-measure度量。从这个测量中得到的值，对应这三个任务的评价，分别是83%，87%和80%。为了评估与特定类型的紧急情况有关的每个图表的详细程度，采用了定性度量标准。考虑到我们方法的重大规模，所获得的结果可以被认为是令人鼓舞的。

{"title":"Management of psychological emergency cases on social media: A hybrid approach combining knowledge graphs and graph neural networks","authors":"Mourad Ellouze , Sonda Rekik , Lamia Hadrich Belguith","doi":"10.1016/j.osnem.2025.100308","DOIUrl":"10.1016/j.osnem.2025.100308","url":null,"abstract":"<div><div>The effects of psychological crises are evolving at an astounding rate nowadays, presenting a significant challenge for everyone involved in tracking these disorders. Therefore, we propose in this paper a hybrid approach based on linguistic processing and numerical techniques allowing to: (i) identify the presence of psychological emergencies among social network users by analyzing their textual production, (ii) determine the specific type of emergency case, (iii) elaborate a graph for each type of emergency, reflecting the different dimensions linked to the psychological emergency, allowing for a better diagnosis of the situation and providing an overall view of the crisis type, (iv) combine the separate graphs for each emergency to address the various semantic aspects. The work was accomplished using advanced language model techniques, knowledge graphs and neural network graphs. The combination of these techniques ensures that their advantages are leveraged while overcoming their limitations in terms of result generalization. The evaluation of different parts related to detecting the presence of psychological problems, predicting specific type of emergency cases, and detecting links between knowledge graphs was measured using the F-measure metric. The values derived from this measure, corresponding to the evaluation of these three tasks, are, respectively, 83%, 87% and 80%. For the evaluation of the elaboration of each graph related to specific type of emergency cases, this was accomplished using qualitative metric standards. The results obtained can be considered encouraging given the significant scale of our approach.</div></div>","PeriodicalId":52228,"journal":{"name":"Online Social Networks and Media","volume":"46 ","pages":"Article 100308"},"PeriodicalIF":0.0,"publicationDate":"2025-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143551494","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

DisTGranD: Granular event/sub-event classification for disaster response DisTGranD：灾难响应的细粒度事件/子事件分类

Q1 Social Sciences

Online Social Networks and Media

Pub Date : 2025-01-01 DOI: 10.1016/j.osnem.2024.100297

Ademola Adesokan , Sanjay Madria , Long Nguyen

Efficient crisis management relies on prompt and precise analysis of disaster data from various sources, including social media. The advantage of fine-grained, annotated, class-labeled data is the provision of a diversified range of information compared to high-level label datasets. In this study, we introduce a dataset richly annotated at a low level to more accurately classify crisis-related communication. To this end, we first present DisTGranD, an extensively annotated dataset of over 47,600 tweets related to earthquakes and hurricanes. The dataset uses the Automatic Content Extraction (ACE) standard to provide detailed classification into dual-layer annotation for events and sub-events and identify critical triggers and supporting arguments. The inter-annotator evaluation of DisTGranD demonstrated high agreement among annotators, with Fleiss Kappa scores of 0.90 and 0.93 for event and sub-event types, respectively. Moreover, a transformer-based embedded phrase extraction method showed XLNet achieving an impressive 96% intra-label similarity score for event type and 97% for sub-event type. We further proposed a novel deep learning classification model, RoBiCCus, which achieved

\geq 90 %

accuracy and F1-Score in the event and sub-event type classification tasks on our DisTGranD dataset and outperformed other models on publicly available disaster datasets. DisTGranD dataset represents a nuanced class-labeled framework for detecting and classifying disaster-related social media content, which can significantly aid decision-making in disaster response. This robust dataset enables deep-learning models to provide insightful, actionable data during crises. Our annotated dataset and code are publicly available on GitHub ¹.

有效的危机管理依赖于对包括社交媒体在内的各种来源的灾难数据进行及时、准确的分析。与高级标签数据集相比，细粒度、带注释、类标记的数据的优点是提供了多样化的信息范围。在本研究中，我们引入了一个低层次的丰富注释数据集，以更准确地对危机相关的通信进行分类。为此，我们首先展示了DisTGranD，这是一个广泛注释的数据集，包含超过47600条与地震和飓风相关的推文。该数据集使用自动内容提取（Automatic Content Extraction， ACE）标准，为事件和子事件提供双层注释的详细分类，并识别关键触发器和支持参数。disgrand的注释者间评价显示注释者之间的一致性很高，事件和子事件类型的Fleiss Kappa评分分别为0.90和0.93。此外，基于转换器的嵌入式短语提取方法表明，XLNet在事件类型和子事件类型上的标签内相似性得分分别达到了令人印象深刻的96%和97%。我们进一步提出了一种新的深度学习分类模型RoBiCCus，该模型在DisTGranD数据集上的事件和子事件类型分类任务中达到了≥90%的准确率和F1-Score，并且在公开可用的灾难数据集上优于其他模型。DisTGranD数据集代表了一个细微的类别标记框架，用于检测和分类与灾害相关的社交媒体内容，这可以显著地帮助灾害响应中的决策。这个强大的数据集使深度学习模型能够在危机期间提供有洞察力的、可操作的数据。我们的注释数据集和代码在GitHub 1上公开可用。

{"title":"DisTGranD: Granular event/sub-event classification for disaster response","authors":"Ademola Adesokan , Sanjay Madria , Long Nguyen","doi":"10.1016/j.osnem.2024.100297","DOIUrl":"10.1016/j.osnem.2024.100297","url":null,"abstract":"<div><div>Efficient crisis management relies on prompt and precise analysis of disaster data from various sources, including social media. The advantage of fine-grained, annotated, class-labeled data is the provision of a diversified range of information compared to high-level label datasets. In this study, we introduce a dataset richly annotated at a low level to more accurately classify crisis-related communication. To this end, we first present DisTGranD, an extensively annotated dataset of over 47,600 tweets related to earthquakes and hurricanes. The dataset uses the Automatic Content Extraction (ACE) standard to provide detailed classification into dual-layer annotation for events and sub-events and identify critical triggers and supporting arguments. The inter-annotator evaluation of DisTGranD demonstrated high agreement among annotators, with Fleiss Kappa scores of 0.90 and 0.93 for event and sub-event types, respectively. Moreover, a transformer-based embedded phrase extraction method showed XLNet achieving an impressive 96% intra-label similarity score for event type and 97% for sub-event type. We further proposed a novel deep learning classification model, RoBiCCus, which achieved <span><math><mrow><mo>≥</mo><mn>90</mn><mtext>%</mtext></mrow></math></span> accuracy and F1-Score in the event and sub-event type classification tasks on our DisTGranD dataset and outperformed other models on publicly available disaster datasets. DisTGranD dataset represents a nuanced class-labeled framework for detecting and classifying disaster-related social media content, which can significantly aid decision-making in disaster response. This robust dataset enables deep-learning models to provide insightful, actionable data during crises. Our annotated dataset and code are publicly available on GitHub <span><span><sup>1</sup></span></span>.</div></div>","PeriodicalId":52228,"journal":{"name":"Online Social Networks and Media","volume":"45 ","pages":"Article 100297"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095218","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Influencer self-disclosure practices on Instagram: A multi-country longitudinal study 影响者在Instagram上的自我披露实践：一项多国纵向研究

Q1 Social Sciences

Online Social Networks and Media

Pub Date : 2025-01-01 DOI: 10.1016/j.osnem.2024.100298

Thales Bertaglia , Catalina Goanta , Gerasimos Spanakis , Adriana Iamnitchi

This paper presents a longitudinal study of more than ten years of activity on Instagram consisting of over a million posts by 400 content creators from four countries: the US, Brazil, Netherlands and Germany. Our study shows differences in the professionalisation of content monetisation between countries, yet consistent patterns; significant differences in the frequency of posts yet similar user engagement trends; and significant differences in the disclosure of sponsored content in some countries, with a direct connection with national legislation. We analyse shifts in marketing strategies due to legislative and platform feature changes, focusing on how content creators adapt disclosure methods to different legal environments. We also analyse the impact of disclosures and sponsored posts on engagement and conclude that, although sponsored posts have lower engagement on average, properly disclosing ads does not reduce engagement further. Our observations stress the importance of disclosure compliance and can guide authorities in developing and monitoring them more effectively.

本文对来自美国、巴西、荷兰和德国四个国家的400名内容创作者在Instagram上10多年来的活动进行了纵向研究，其中包括100多万篇帖子。我们的研究显示，不同国家之间的内容货币化专业化存在差异，但模式是一致的；发布频率存在显著差异，但用户粘性趋势相似；而一些国家在赞助内容披露方面的显著差异，与国家立法有直接联系。我们分析了由于立法和平台功能变化而导致的营销策略变化，重点关注内容创作者如何根据不同的法律环境调整披露方法。我们还分析了披露和赞助帖子对参与度的影响，并得出结论，尽管赞助帖子的平均参与度较低，但适当披露广告不会进一步降低参与度。我们的观察结果强调了披露合规的重要性，并可以指导当局更有效地制定和监督披露合规。

{"title":"Influencer self-disclosure practices on Instagram: A multi-country longitudinal study","authors":"Thales Bertaglia , Catalina Goanta , Gerasimos Spanakis , Adriana Iamnitchi","doi":"10.1016/j.osnem.2024.100298","DOIUrl":"10.1016/j.osnem.2024.100298","url":null,"abstract":"<div><div>This paper presents a longitudinal study of more than ten years of activity on Instagram consisting of over a million posts by 400 content creators from four countries: the US, Brazil, Netherlands and Germany. Our study shows differences in the professionalisation of content monetisation between countries, yet consistent patterns; significant differences in the frequency of posts yet similar user engagement trends; and significant differences in the disclosure of sponsored content in some countries, with a direct connection with national legislation. We analyse shifts in marketing strategies due to legislative and platform feature changes, focusing on how content creators adapt disclosure methods to different legal environments. We also analyse the impact of disclosures and sponsored posts on engagement and conclude that, although sponsored posts have lower engagement on average, properly disclosing ads does not reduce engagement further. Our observations stress the importance of disclosure compliance and can guide authorities in developing and monitoring them more effectively.</div></div>","PeriodicalId":52228,"journal":{"name":"Online Social Networks and Media","volume":"45 ","pages":"Article 100298"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095211","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

BD2TSumm: A Benchmark Dataset for Abstractive Disaster Tweet Summarization bd2tsum：用于抽象灾难推文摘要的基准数据集

Q1 Social Sciences

Online Social Networks and Media

Pub Date : 2025-01-01 DOI: 10.1016/j.osnem.2024.100299

Piyush Kumar Garg , Roshni Chakraborty , Sourav Kumar Dandapat

Online social media platforms, such as Twitter, are mediums for valuable updates during disasters. However, the large scale of available information makes it difficult for humans to identify relevant information from the available information. An automatic summary of these tweets provides identification of relevant information easy and ensures a holistic overview of a disaster event to process the aid for disaster response. In literature, there are two types of abstractive disaster tweet summarization approaches based on the format of output summary: key-phrased-based (where summary is a set of key-phrases) and sentence-based (where summary is a paragraph consisting of sentences). Existing sentence-based abstractive approaches are either unsupervised or supervised. However, both types of approaches require a sizable amount of ground-truth summaries for training and/or evaluation such that they work on disaster events irrespective of type and location. The lack of abstractive disaster ground-truth summaries and guidelines for annotation motivates us to come up with a systematic procedure to create abstractive sentence ground-truth summaries of disaster events. Therefore, this paper presents a two-step systematic annotation procedure for sentence-based abstractive summary creation. Additionally, we release BD2TSumm, i.e., a benchmark ground-truth dataset for evaluating the sentence-based abstractive summarization approaches for disaster events. BD2TSumm consists of 15 ground-truth summaries belonging to 5 different continents and both natural and man-made disaster types. Furthermore, to ensure the high quality of the generated ground-truth summaries, we evaluate them qualitatively (using five metrics) and quantitatively (using two metrics). Finally, we compare 12 existing State-Of-The-Art (SOTA) abstractive summarization approaches on these ground-truth summaries using ROUGE-N F1-score.

在线社交媒体平台，如Twitter，是灾难期间有价值的更新媒介。然而，大量的可用信息使得人类很难从可用信息中识别出相关信息。这些推文的自动摘要可以很容易地识别相关信息，并确保对灾难事件有一个全面的概述，从而为灾难响应提供援助。在文献中，基于输出摘要的格式，有两种抽象的灾难推文摘要方法：基于关键短语（key-phrase -based，摘要是一组关键短语）和基于句子（sentence-based，摘要是由句子组成的段落）。现有基于句子的抽象方法要么是无监督的，要么是有监督的。然而，这两种方法都需要大量的基础事实总结来进行培训和/或评估，以便它们适用于灾害事件，而不考虑类型和地点。由于缺乏抽象的灾难基础真值摘要和注释指南，促使我们提出一种系统的程序来创建抽象的句子基础真值摘要。因此，本文提出了一种基于句子的抽象摘要生成的两步系统标注流程。此外，我们发布了BD2TSumm，即一个基准的真实数据集，用于评估基于句子的灾难事件抽象摘要方法。BD2TSumm由5个不同大洲的15个事实总结组成，包括自然灾害和人为灾害类型。此外，为了确保生成的基础事实摘要的高质量，我们定性地（使用五个指标）和定量地（使用两个指标）对它们进行评估。最后，我们使用ROUGE-N F1-score比较了12种现有的最先进的（SOTA）抽象摘要方法对这些基础真值摘要的影响。

{"title":"BD2TSumm: A Benchmark Dataset for Abstractive Disaster Tweet Summarization","authors":"Piyush Kumar Garg , Roshni Chakraborty , Sourav Kumar Dandapat","doi":"10.1016/j.osnem.2024.100299","DOIUrl":"10.1016/j.osnem.2024.100299","url":null,"abstract":"<div><div>Online social media platforms, such as Twitter, are mediums for valuable updates during disasters. However, the large scale of available information makes it difficult for humans to identify relevant information from the available information. An automatic summary of these tweets provides identification of relevant information easy and ensures a holistic overview of a disaster event to process the aid for disaster response. In literature, there are two types of abstractive disaster tweet summarization approaches based on the format of output summary: key-phrased-based (where summary is a set of key-phrases) and sentence-based (where summary is a paragraph consisting of sentences). Existing sentence-based abstractive approaches are either unsupervised or supervised. However, both types of approaches require a sizable amount of ground-truth summaries for training and/or evaluation such that they work on disaster events irrespective of type and location. The lack of abstractive disaster ground-truth summaries and guidelines for annotation motivates us to come up with a systematic procedure to create abstractive sentence ground-truth summaries of disaster events. Therefore, this paper presents a two-step systematic annotation procedure for sentence-based abstractive summary creation. Additionally, we release <em>BD2TSumm</em>, i.e., a benchmark ground-truth dataset for evaluating the sentence-based abstractive summarization approaches for disaster events. <em>BD2TSumm</em> consists of 15 ground-truth summaries belonging to 5 different continents and both natural and man-made disaster types. Furthermore, to ensure the high quality of the generated ground-truth summaries, we evaluate them qualitatively (using five metrics) and quantitatively (using two metrics). Finally, we compare 12 existing State-Of-The-Art (SOTA) abstractive summarization approaches on these ground-truth summaries using ROUGE-N F1-score.</div></div>","PeriodicalId":52228,"journal":{"name":"Online Social Networks and Media","volume":"45 ","pages":"Article 100299"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095219","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

How political symbols spread in online social networks: Using agent-based models to replicate the complex contagion of the yellow ribbon in Twitter 政治符号如何在在线社交网络中传播：使用基于主体的模型来复制Twitter中黄丝带的复杂传染

Q1 Social Sciences

Online Social Networks and Media

Pub Date : 2025-01-01 DOI: 10.1016/j.osnem.2025.100300

Francisco J. León-Medina

This paper analyzes the diffusion of the yellow ribbon in Twitter, a political symbol that represents the demand for the release of Catalan prisoners. We gathered data on potential users of the symbol in Twitter (users that publicly backed the cause), including their social network of friendships, and built an agent-based simulation to replicate the diffusion of the symbol in a digital twin version of the observed network. Our hypothesis was that complex contagion is the best explanation of the observed statistical relation between the proportion of adopting neighbors and the probability of adoption. Results show that the complex contagion model outperforms the simple contagion model and generates a better fit between the observed and the simulated pattern when the typical conditions of a complex contagion process are added to the baseline model, that is, when agents are affected by their reference group behavior rather than by the most influential nodes of the network, and when we identify a peripherical and densely connected network community and trigger the process from there. These results widen the set of behaviors whose diffusion can be explained as complex contagion to include adoption in low-risk/low-cost behaviors among people who would usually not resist adoption.

本文分析了黄色丝带在推特上的传播，这是一个政治符号，代表了对释放加泰罗尼亚囚犯的要求。我们收集了Twitter上该符号的潜在用户（公开支持该事业的用户）的数据，包括他们的社交网络，并建立了一个基于代理的模拟，以在观察到的网络的数字孪生版本中复制该符号的传播。我们的假设是，复杂传染是观察到的收养邻居比例与被收养概率之间的统计关系的最佳解释。结果表明，当基线模型中加入复杂传染过程的典型条件时，即当代理受其参考群体行为影响而不是受网络中最具影响力的节点影响时，当我们确定一个外围且紧密连接的网络社区并从那里触发该过程时，复杂传染模型优于简单传染模型，并在观察到的模式和模拟模式之间产生更好的拟合。这些结果扩大了行为的范围，这些行为的扩散可以解释为复杂的传染，包括在通常不会抵制采用的人群中采用低风险/低成本行为。

{"title":"How political symbols spread in online social networks: Using agent-based models to replicate the complex contagion of the yellow ribbon in Twitter","authors":"Francisco J. León-Medina","doi":"10.1016/j.osnem.2025.100300","DOIUrl":"10.1016/j.osnem.2025.100300","url":null,"abstract":"<div><div>This paper analyzes the diffusion of the yellow ribbon in Twitter, a political symbol that represents the demand for the release of Catalan prisoners. We gathered data on potential users of the symbol in Twitter (users that publicly backed the cause), including their social network of friendships, and built an agent-based simulation to replicate the diffusion of the symbol in a digital twin version of the observed network. Our hypothesis was that complex contagion is the best explanation of the observed statistical relation between the proportion of adopting neighbors and the probability of adoption. Results show that the complex contagion model outperforms the simple contagion model and generates a better fit between the observed and the simulated pattern when the typical conditions of a complex contagion process are added to the baseline model, that is, when agents are affected by their reference group behavior rather than by the most influential nodes of the network, and when we identify a peripherical and densely connected network community and trigger the process from there. These results widen the set of behaviors whose diffusion can be explained as complex contagion to include adoption in low-risk/low-cost behaviors among people who would usually <em>not</em> resist adoption.</div></div>","PeriodicalId":52228,"journal":{"name":"Online Social Networks and Media","volume":"45 ","pages":"Article 100300"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095301","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Why are you traveling? Inferring trip profiles from online reviews and domain-knowledge 你为什么要旅行？从在线评论和领域知识推断旅行概况

Q1 Social Sciences

Online Social Networks and Media

Pub Date : 2025-01-01 DOI: 10.1016/j.osnem.2024.100296

Lucas G.S. Félix, Washington Cunha, Claudio M.V. de Andrade, Marcos André Gonçalves, Jussara M. Almeida

This paper addresses the task of inferring trip profiles (TPs), which consists of determining the profile of travelers engaged in a particular trip given a set of possible categories. TPs may include working trips, leisure journeys with friends, or family vacations. Travelers with different TPs typically have varied plans regarding destinations and timing. TP inference may provide significant insights for numerous tourism-related services, such as geo-recommender systems and tour planning. We focus on TP inference using TripAdvisor, a prominent tourism-centric social media platform, as our data source. Our goal is to evaluate how effectively we can automatically discern the TP from a user review on this platform. A user review encompasses both textual feedback and domain-specific data (such as a user’s previous visits to the location), which are crucial for accurately characterizing the trip. To achieve this, we assess various feature sets (including text and domain-specific) and implement advanced machine learning models, such as neural Transformers and open-source Large Language Models (Llama 2, Bloom). We examine two variants of the TP inference task—binary and multi-class. Surprisingly, our findings reveal that combining domain-specific features with TF-IDF-based representation in an LGBM model performs as well as more complex Transformer and LLM models, while being much more efficient and interpretable.

本文讨论了推断旅行概况（TPs）的任务，该任务包括确定给定一组可能类别的特定旅行中从事的旅行者的概况。旅行计划可能包括工作旅行、与朋友的休闲旅行或家庭度假。有不同旅游计划的旅行者通常在目的地和时间方面有不同的计划。TP推理可以为许多与旅游相关的服务提供重要的见解，例如地理推荐系统和旅游规划。我们使用TripAdvisor（一个著名的以旅游为中心的社交媒体平台）作为我们的数据源，专注于TP推断。我们的目标是评估我们在这个平台上从用户评论中自动识别TP的有效性。用户评论包含文本反馈和特定于领域的数据（例如用户以前对该位置的访问），这对于准确描述旅行的特征至关重要。为了实现这一点，我们评估了各种特征集（包括文本和特定领域），并实现了先进的机器学习模型，如神经变形器和开源大型语言模型（Llama 2, Bloom）。我们研究了TP推理任务的两种变体——二进制和多类。令人惊讶的是，我们的研究结果表明，在LGBM模型中，将特定领域的特征与基于tf - idf的表示相结合，与更复杂的Transformer和LLM模型表现一样好，同时效率更高，可解释性更高。

{"title":"Why are you traveling? Inferring trip profiles from online reviews and domain-knowledge","authors":"Lucas G.S. Félix, Washington Cunha, Claudio M.V. de Andrade, Marcos André Gonçalves, Jussara M. Almeida","doi":"10.1016/j.osnem.2024.100296","DOIUrl":"10.1016/j.osnem.2024.100296","url":null,"abstract":"<div><div>This paper addresses the task of inferring trip profiles (TPs), which consists of determining the profile of travelers engaged in a particular trip given a set of possible categories. TPs may include working trips, leisure journeys with friends, or family vacations. Travelers with different TPs typically have varied plans regarding destinations and timing. TP inference may provide significant insights for numerous tourism-related services, such as geo-recommender systems and tour planning. We focus on TP inference using TripAdvisor, a prominent tourism-centric social media platform, as our data source. Our goal is to evaluate how effectively we can automatically discern the TP from a user review on this platform. A user review encompasses both textual feedback and domain-specific data (such as a user’s previous visits to the location), which are crucial for accurately characterizing the trip. To achieve this, we assess various feature sets (including text and domain-specific) and implement advanced machine learning models, such as neural Transformers and open-source Large Language Models (Llama 2, Bloom). We examine two variants of the TP inference task—binary and multi-class. Surprisingly, our findings reveal that combining domain-specific features with TF-IDF-based representation in an LGBM model performs as well as more complex Transformer and LLM models, while being much more efficient and interpretable.</div></div>","PeriodicalId":52228,"journal":{"name":"Online Social Networks and Media","volume":"45 ","pages":"Article 100296"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095300","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Harnessing prompt-based large language models for disaster monitoring and automated reporting from social media feedback 利用基于提示的大型语言模型，从社交媒体反馈中进行灾害监测和自动报告

Q1 Social Sciences

Online Social Networks and Media

Pub Date : 2024-11-25 DOI: 10.1016/j.osnem.2024.100295

Riccardo Cantini, Cristian Cosentino, Fabrizio Marozzo, Domenico Talia, Paolo Trunfio

In recent years, social media has emerged as one of the main platforms for real-time reporting of issues during disasters and catastrophic events. While great strides have been made in collecting such information, there remains an urgent need to improve user reports’ automation, aggregation, and organization to streamline various tasks, including rescue operations, resource allocation, and communication with the press. This paper introduces an innovative methodology that leverages the power of prompt-based Large Language Models (LLMs) to strengthen disaster response and management. By analyzing large volumes of user-generated content, our methodology identifies issues reported by citizens who have experienced a disastrous event, such as damaged buildings, broken gas pipelines, and flooding. It also localizes all posts containing references to geographic information in the text, allowing for aggregation of posts that occurred nearby. By leveraging these localized citizen-reported issues, the methodology generates insightful reports full of essential information for emergency services, news agencies, and other interested parties. Extensive experimentation on large datasets validates the accuracy and efficiency of our methodology in classifying posts, detecting sub-events, and producing real-time reports. These findings highlight the practical value of prompt-based LLMs in disaster response, emphasizing their flexibility and adaptability in delivering timely insights that support more effective interventions.

近年来，社交媒体已成为灾害和灾难事件发生期间实时报告问题的主要平台之一。虽然在收集此类信息方面取得了长足进步，但仍迫切需要改进用户报告的自动化、聚合和组织，以简化各种任务，包括救援行动、资源分配以及与新闻界的沟通。本文介绍了一种创新方法，该方法利用基于提示的大型语言模型（LLM）的力量来加强灾难响应和管理。通过分析大量用户生成的内容，我们的方法可识别经历过灾难性事件的市民所报告的问题，如受损的建筑物、破损的天然气管道和洪水。它还能将文本中包含地理信息参考的所有帖子本地化，从而汇总附近发生的帖子。通过利用这些本地化的公民报告问题，该方法可生成富有洞察力的报告，为应急服务、新闻机构和其他相关方提供重要信息。在大型数据集上进行的广泛实验验证了我们的方法在分类帖子、检测子事件和生成实时报告方面的准确性和效率。这些发现凸显了基于提示的 LLM 在灾难响应中的实用价值，强调了它们在提供及时见解以支持更有效干预方面的灵活性和适应性。

{"title":"Harnessing prompt-based large language models for disaster monitoring and automated reporting from social media feedback","authors":"Riccardo Cantini, Cristian Cosentino, Fabrizio Marozzo, Domenico Talia, Paolo Trunfio","doi":"10.1016/j.osnem.2024.100295","DOIUrl":"10.1016/j.osnem.2024.100295","url":null,"abstract":"<div><div>In recent years, social media has emerged as one of the main platforms for real-time reporting of issues during disasters and catastrophic events. While great strides have been made in collecting such information, there remains an urgent need to improve user reports’ automation, aggregation, and organization to streamline various tasks, including rescue operations, resource allocation, and communication with the press. This paper introduces an innovative methodology that leverages the power of prompt-based Large Language Models (LLMs) to strengthen disaster response and management. By analyzing large volumes of user-generated content, our methodology identifies issues reported by citizens who have experienced a disastrous event, such as damaged buildings, broken gas pipelines, and flooding. It also localizes all posts containing references to geographic information in the text, allowing for aggregation of posts that occurred nearby. By leveraging these localized citizen-reported issues, the methodology generates insightful reports full of essential information for emergency services, news agencies, and other interested parties. Extensive experimentation on large datasets validates the accuracy and efficiency of our methodology in classifying posts, detecting sub-events, and producing real-time reports. These findings highlight the practical value of prompt-based LLMs in disaster response, emphasizing their flexibility and adaptability in delivering timely insights that support more effective interventions.</div></div>","PeriodicalId":52228,"journal":{"name":"Online Social Networks and Media","volume":"45 ","pages":"Article 100295"},"PeriodicalIF":0.0,"publicationDate":"2024-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142702794","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

HaRNaT - A dynamic hashtag recommendation system using news HaRNaT - 利用新闻的动态标签推荐系统

Q1 Social Sciences

Online Social Networks and Media

Pub Date : 2024-11-23 DOI: 10.1016/j.osnem.2024.100294

Divya Gupta, Shampa Chakraverty

Microblogging platforms such as X and Mastadon have evolved into significant data sources, where the Hashtag Recommendation System (HRS) is being devised to automate the recommendation of hashtags for user queries. We propose a context-sensitive, Machine Learning based HRS named HaRNaT, that strategically leverages news articles to identify pertinent keywords and subjects related to a query. It interprets the fresh context of a query and tracks the evolving dynamics of hashtags to evaluate their relevance in the present context. In contrast to prior methods that primarily rely on microblog content for hashtag recommendation, HaRNaT mines contextually related microblogs and assesses the relevance of co-occurring hashtags with news information. To accomplish this, it evaluates hashtag features, including pertinence, popularity among users, and association with other hashtags. In performance evaluation of HaRNaT trained on these features demonstrates a macro-averaged precision of 84% with Naive Bayes and 80% with Logistic Regression. Compared to Hashtagify- a hashtag search engine, HaRNaT offers a dynamically evolving set of hashtags.

X 和 Mastadon 等微博平台已发展成为重要的数据源，其中的标签推荐系统（HRS）被设计用于为用户查询自动推荐标签。我们提出了一种基于机器学习的上下文敏感型 HRS，名为 HaRNaT，它能战略性地利用新闻文章来识别与查询相关的关键词和主题。它能解释查询的新上下文，并跟踪标签不断变化的动态，以评估其在当前上下文中的相关性。与之前主要依靠微博内容进行标签推荐的方法不同，HaRNaT 可挖掘与上下文相关的微博，并评估与新闻信息共同出现的标签的相关性。为此，它评估了标签的特征，包括相关性、在用户中的流行度以及与其他标签的关联性。根据这些特征对 HaRNaT 进行的性能评估表明，使用 Naive Bayes 算法的宏观平均精确度为 84%，使用 Logistic Regression 算法的宏观平均精确度为 80%。与 Hashtagify（一种标签搜索引擎）相比，HaRNaT 提供了一组动态演化的标签。

{"title":"HaRNaT - A dynamic hashtag recommendation system using news","authors":"Divya Gupta, Shampa Chakraverty","doi":"10.1016/j.osnem.2024.100294","DOIUrl":"10.1016/j.osnem.2024.100294","url":null,"abstract":"<div><div>Microblogging platforms such as <em>X</em> and <em>Mastadon</em> have evolved into significant data sources, where the Hashtag Recommendation System (HRS) is being devised to automate the recommendation of hashtags for user queries. We propose a context-sensitive, Machine Learning based HRS named <em>HaRNaT</em>, that strategically leverages news articles to identify pertinent keywords and subjects related to a query. It interprets the fresh context of a query and tracks the evolving dynamics of hashtags to evaluate their relevance in the present context. In contrast to prior methods that primarily rely on microblog content for hashtag recommendation, <em>HaRNaT</em> mines contextually related microblogs and assesses the relevance of co-occurring hashtags with news information. To accomplish this, it evaluates hashtag features, including pertinence, popularity among users, and association with other hashtags. In performance evaluation of <em>HaRNaT</em> trained on these features demonstrates a macro-averaged precision of 84% with Naive Bayes and 80% with Logistic Regression. Compared to <em>Hashtagify</em>- a hashtag search engine, <em>HaRNaT</em> offers a dynamically evolving set of hashtags.</div></div>","PeriodicalId":52228,"journal":{"name":"Online Social Networks and Media","volume":"45 ","pages":"Article 100294"},"PeriodicalIF":0.0,"publicationDate":"2024-11-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142702795","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

How does user-generated content on Social Media affect stock predictions? A case study on GameStop 社交媒体上的用户生成内容如何影响股票预测？GameStop 案例研究

Q1 Social Sciences

Online Social Networks and Media

Pub Date : 2024-11-01 DOI: 10.1016/j.osnem.2024.100293

Antonino Ferraro , Giancarlo Sperlì

One of the main challenges in the financial market concerns the forecasting of stock behavior, which plays a key role in supporting the financial decisions of investors. In recent years, the large amount of available financial data and the heterogeneous contextual information led researchers to investigate data-driven models using Artificial Intelligence (AI)-based approaches for forecasting stock prices. Recent methodologies focus mainly on analyzing participants from Reddit without considering other social media and how their combination affects the stock market, which remains an open challenge. In this paper, we combine financial data and textual user-generated information, which are provided as input to various deep learning models, to develop a stock forecasting system. The main novelties of the proposal concern the design of a multi-modal approach combining historical stock prices and sentiment scores extracted by different Online Social Networks (OSNs), also unveiling possible correlations about heterogeneous information evaluated during the GameStop squeeze. In particular, we have examined several AI-based models and investigated the impact of textual data inferred from well-known Online Social Networks (i.e., Reddit and Twitter) on stock market behavior by conducting a case study on GameStop. Although users’ dynamic opinions on social networks may have a detrimental impact on the stock prediction task, our investigation has demonstrated the usefulness of assessing user-generated content inferred from various OSNs on the market forecasting problem.

股票行为预测是金融市场面临的主要挑战之一，在支持投资者做出金融决策方面发挥着关键作用。近年来，大量可用的金融数据和异构的上下文信息促使研究人员利用基于人工智能（AI）的方法研究数据驱动模型，以预测股票价格。最近的方法主要集中于分析 Reddit 上的参与者，而没有考虑其他社交媒体以及它们的组合如何影响股市，这仍然是一个有待解决的难题。在本文中，我们结合了金融数据和用户生成的文本信息，将其作为各种深度学习模型的输入，开发了一个股票预测系统。该提案的主要新颖之处在于设计了一种多模式方法，将历史股票价格和不同在线社交网络（OSN）提取的情绪评分结合起来，同时揭示了 GameStop 挤压期间评估的异构信息可能存在的相关性。特别是，我们通过对 GameStop 的案例研究，检验了几种基于人工智能的模型，并调查了从知名在线社交网络（即 Reddit 和 Twitter）中推断出的文本数据对股市行为的影响。虽然用户在社交网络上的动态观点可能会对股票预测任务产生不利影响，但我们的调查证明了评估从各种在线社交网络推断出的用户生成内容对市场预测问题的有用性。

{"title":"How does user-generated content on Social Media affect stock predictions? A case study on GameStop","authors":"Antonino Ferraro , Giancarlo Sperlì","doi":"10.1016/j.osnem.2024.100293","DOIUrl":"10.1016/j.osnem.2024.100293","url":null,"abstract":"<div><div>One of the main challenges in the financial market concerns the forecasting of stock behavior, which plays a key role in supporting the financial decisions of investors. In recent years, the large amount of available financial data and the heterogeneous contextual information led researchers to investigate data-driven models using Artificial Intelligence (AI)-based approaches for forecasting stock prices. Recent methodologies focus mainly on analyzing participants from Reddit without considering other social media and how their combination affects the stock market, which remains an open challenge. In this paper, we combine financial data and textual user-generated information, which are provided as input to various deep learning models, to develop a stock forecasting system. The main novelties of the proposal concern the design of a multi-modal approach combining historical stock prices and sentiment scores extracted by different Online Social Networks (OSNs), also unveiling possible correlations about heterogeneous information evaluated during the GameStop squeeze. In particular, we have examined several AI-based models and investigated the impact of textual data inferred from well-known Online Social Networks (<em>i.e.</em>, Reddit and Twitter) on stock market behavior by conducting a case study on GameStop. Although users’ dynamic opinions on social networks may have a detrimental impact on the stock prediction task, our investigation has demonstrated the usefulness of assessing user-generated content inferred from various OSNs on the market forecasting problem.</div></div>","PeriodicalId":52228,"journal":{"name":"Online Social Networks and Media","volume":"43 ","pages":"Article 100293"},"PeriodicalIF":0.0,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142653576","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