Online Social Networks and Media最新文献

{"title":"Social search: Retrieving information in Online Social platforms – A survey","authors":"Maddalena Amendola ,&nbsp;Andrea Passarella ,&nbsp;Raffaele Perego","doi":"10.1016/j.osnem.2023.100254","DOIUrl":"https://doi.org/10.1016/j.osnem.2023.100254","url":null,"abstract":"<div><p><em>Social Search</em> research studies methodologies exploiting social information to better satisfy user information needs in Online Social Media while simplifying the search effort and consequently reducing the time spent and the computational resources utilized. Starting from previous studies, in this work, we analyze the current state of the art of the Social Search area, proposing a new taxonomy and highlighting current limitations and open research directions. We divide the Social Search area into three subcategories, where the social aspect plays a pivotal role: <em>Social Question&amp;Answering</em>, <em>Social Content Search</em>, and <em>Social Collaborative Search</em>. For each subcategory, we present the key concepts and selected representative approaches in the literature in greater detail. We found that, up to now, a large body of studies model users’ preferences and their relations by simply combining social features made available by social platforms. It paves the way for significant research to exploit more structured information about users’ social profiles and behaviours (as they can be inferred from data available on social platforms) to optimize their information needs further.</p></div>","PeriodicalId":52228,"journal":{"name":"Online Social Networks and Media","volume":"36 ","pages":"Article 100254"},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49888616","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}

{"title":"Session-based cyberbullying detection in social media: A survey","authors":"Peiling Yi,&nbsp;Arkaitz Zubiaga","doi":"10.1016/j.osnem.2023.100250","DOIUrl":"https://doi.org/10.1016/j.osnem.2023.100250","url":null,"abstract":"<div><p>Cyberbullying is a pervasive problem in online social media, where a bully abuses a victim through a social media session. By investigating cyberbullying perpetrated through social media sessions, recent research has looked into mining patterns and features for modelling and understanding the two defining characteristics of cyberbullying: repetitive behaviour and power imbalance. In this survey paper, we define a framework that encapsulates four different steps session-based cyberbullying detection should go through, and discuss the multiple challenges that differ from single text-based cyberbullying detection. Based on this framework, we provide a comprehensive overview of session-based cyberbullying detection in social media, delving into existing efforts from a data and methodological perspective. Our review leads us to proposing evidence-based criteria for a set of best practices to create session-based cyberbullying datasets. In addition, we perform benchmark experiments comparing the performance of state-of-the-art session-based cyberbullying detection models as well as large pre-trained language models across two different datasets. Through our review, we also put forth a set of open challenges as future research directions.</p></div>","PeriodicalId":52228,"journal":{"name":"Online Social Networks and Media","volume":"36 ","pages":"Article 100250"},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49888996","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}

{"title":"Turning captchas against humanity: Captcha-based attacks in online social media","authors":"Mauro Conti,&nbsp;Luca Pajola,&nbsp;Pier Paolo Tricomi","doi":"10.1016/j.osnem.2023.100252","DOIUrl":"https://doi.org/10.1016/j.osnem.2023.100252","url":null,"abstract":"<div><p>Nowadays, people generate and share massive amounts of content on online platforms (e.g., social networks, blogs). In 2021, the 1.9 billion daily active Facebook users posted around 150 thousand photos every minute. Content moderators constantly monitor these online platforms to prevent the spreading of inappropriate content (e.g., hate speech, nudity images). Based on deep learning (DL) advances, Automatic Content Moderators (ACM) help human moderators handle high data volume. Despite their advantages, attackers can exploit weaknesses of DL components (e.g., preprocessing, model) to affect their performance. Therefore, an attacker can leverage such techniques to spread inappropriate content by evading ACM.</p><p>In this work, we analyzed 4600 potentially toxic Instagram posts, and we discovered that 44% of them adopt obfuscations that might undermine ACM. As these posts are reminiscent of captchas (i.e., not understandable by automated mechanisms), we coin this threat as Captcha Attack (<span><math><mrow><mi>C</mi><mi>A</mi><mi>P</mi><mi>A</mi></mrow></math></span>). Our contributions start by proposing a <span><math><mrow><mi>C</mi><mi>A</mi><mi>P</mi><mi>A</mi></mrow></math></span> taxonomy to better understand how ACM is vulnerable to obfuscation attacks. We then focus on the broad sub-category of <span><math><mrow><mi>C</mi><mi>A</mi><mi>P</mi><mi>A</mi></mrow></math></span> using textual Captcha Challenges, namely <span>CC-CAPA</span>, and we empirically demonstrate that it evades real-world ACM (i.e., Amazon, Google, Microsoft) with 100% accuracy. Our investigation revealed that ACM failures are caused by the OCR text extraction phase. The training of OCRs to withstand such obfuscation is therefore crucial, but huge amounts of data are required. Thus, we investigate methods to identify <span>CC-CAPA</span> samples from large sets of data (originated by three OSN – Pinterest, Twitter, Yahoo-Flickr), and we empirically demonstrate that supervised techniques identify target styles of samples almost perfectly. Unsupervised solutions, on the other hand, represent a solid methodology for inspecting uncommon data to detect new obfuscation techniques.</p></div>","PeriodicalId":52228,"journal":{"name":"Online Social Networks and Media","volume":"36 ","pages":"Article 100252"},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49888997","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}

{"title":"The role of social media on the evolution of companies: A Twitter analysis of Streaming Service Providers","authors":"Marco Arazzi ,&nbsp;Marco Ferretti ,&nbsp;Serena Nicolazzo ,&nbsp;Antonino Nocera","doi":"10.1016/j.osnem.2023.100251","DOIUrl":"https://doi.org/10.1016/j.osnem.2023.100251","url":null,"abstract":"<div><p><span>In recent years, Social Networks and, in particular, Twitter have proved to be a fertile ground for those scholars and companies interested in exploring the effectiveness of brand marketing communications. This is even more true when it comes to TV Streaming Service Providers, such as Netflix or Amazon. For these types of companies, Twitter represents not only a valuable source of data for business intelligence, but also a connected and co-viewing platform and outage detection system. In this paper, we carry out our analysis by exploring and comparing, through disparate </span>machine learning techniques<span> and natural language processing solutions, the behavior of several Twitter accounts corresponding to different Streaming Service Providers by considering their possible stage in the Technology Adoption Life Cycle. Interestingly, such an analysis allows for the identification of the most suitable strategies that can be carried out on Twitter by Streaming Service Providers to improve the user involvement on the basis of their current stage. To the best of our knowledge, a complete analysis able to depict Twitter strategies of success for Streaming Service Providers does not exist in current literature yet.</span></p></div>","PeriodicalId":52228,"journal":{"name":"Online Social Networks and Media","volume":"36 ","pages":"Article 100251"},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49888994","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}

{"title":"Interpretable fake news detection with topic and deep variational models","authors":"Marjan Hosseini ,&nbsp;Alireza Javadian Sabet ,&nbsp;Suining He ,&nbsp;Derek Aguiar","doi":"10.1016/j.osnem.2023.100249","DOIUrl":"https://doi.org/10.1016/j.osnem.2023.100249","url":null,"abstract":"<div><p><span>The growing societal dependence on social media and user generated content for news and information has increased the influence of unreliable sources and fake content, which muddles public discourse and lessens trust in the media. Validating the credibility of such information is a difficult task that is susceptible to confirmation bias, leading to the development of algorithmic techniques to distinguish between fake and real news. However, most existing methods are challenging to interpret, making it difficult to establish trust in predictions, and make assumptions that are unrealistic in many real-world scenarios, e.g., the availability of audiovisual features or provenance. In this work, we focus on fake news detection of textual content using interpretable features and methods. In particular, we have developed a deep probabilistic model that integrates a dense representation of textual news using a variational </span>autoencoder<span> and bi-directional Long Short-Term Memory (LSTM) networks with semantic topic-related features inferred from a Bayesian admixture model. Extensive experimental studies with 3 real-world datasets demonstrate that our model achieves comparable performance to state-of-the-art competing models while facilitating model interpretability<span> from the learned topics. Finally, we have conducted model ablation studies to justify the effectiveness and accuracy of integrating neural embeddings and topic features both quantitatively by evaluating performance and qualitatively through separability in lower dimensional embeddings.</span></span></p></div>","PeriodicalId":52228,"journal":{"name":"Online Social Networks and Media","volume":"36 ","pages":"Article 100249"},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49888995","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}

{"title":"Multi-contextual learning in disinformation research: A review of challenges, approaches, and opportunities","authors":"Bhaskarjyoti Das,&nbsp;Sudarshan T‏S‏B‏","doi":"10.1016/j.osnem.2023.100247","DOIUrl":"https://doi.org/10.1016/j.osnem.2023.100247","url":null,"abstract":"<div><p>Though a fair amount of research is being done to address disinformation in online social media, it has so far managed to stay ahead of the researchers’ learning curves forcing the publishers to rely on manual effort to a large extent. The root cause lies in the complex multi-contextual nature of the problem. The way a disinformation propagates on the social graph depends on multiple contexts i.e., content of the original news, credibility of the news source, poster of the message referring the news, message content, recipients of message with their social as well as psychological backgrounds, the role played by the available knowledge, and the temporal as well as the propagation pattern while the message becomes viral on the social graph. This article reviews each of these contexts to define the multi-contextual learning problem and summarizes the work done using each of them. Multi-contextual learning gets exacerbated by few other challenges. This article also reviews the approaches adopted so far to tackle each of these challenges along with an exhaustive review of the multi-contextual learning strategies adopted so far. The multi-contextuality aspect as well as the related challenges are horizontal in nature across the three primary verticals of disinformation i.e., fake news, rumor, and propaganda. Existing review articles primarily tackle one of these verticals in isolation with one or few of the above mentioned contexts. Also the related challenges have not seen any focused review so far. This article seeks to address these gaps by offering a comprehensive systemic view across this domain and concludes with a list of future research directions.</p></div>","PeriodicalId":52228,"journal":{"name":"Online Social Networks and Media","volume":"34 ","pages":"Article 100247"},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49906484","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}

{"title":"Erratum to <Patterns of democracy? Social network analysis of parliamentary Twitter networks in 12 countries’> <[Online Social Networks and Media, 24 (2023) /100154]>","authors":"Stiene Praet ,&nbsp;David Martens ,&nbsp;Peter Van Aelst","doi":"10.1016/j.osnem.2023.100246","DOIUrl":"https://doi.org/10.1016/j.osnem.2023.100246","url":null,"abstract":"","PeriodicalId":52228,"journal":{"name":"Online Social Networks and Media","volume":"34 ","pages":"Article 100246"},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49906483","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}

{"title":"A nontrivial interplay between triadic closure, preferential, and anti-preferential attachment: New insights from online data","authors":"Ivan V. Kozitsin ,&nbsp;Alexander V. Gubanov ,&nbsp;Eduard R. Sayfulin ,&nbsp;Vyacheslav L. Goiko","doi":"10.1016/j.osnem.2023.100248","DOIUrl":"https://doi.org/10.1016/j.osnem.2023.100248","url":null,"abstract":"<div><p>This paper presents an analysis of a temporal network that describes the social connections of a large-scale (∼30,000) sample of online social network<span> users, inhabitants of a fixed city. We tested how the main network formation determinants—transitivity, preferential attachment, and social selection—contribute to network evolution. We obtained that tie appearing and tie removing events are governed by different combinations of mechanisms: whereas the structure of the network is responsible for the formation of new ties, nodal nonstructural characteristics “decide” whether a tie will continue to exist. Next, our findings show that only one network formation mechanism, gender selectivity, has a significant impact on both tie appearing and tie removing processes. What is interesting, the effect of gender selectivity is most notable for low-degree vertices. Besides this, our analysis revealed that opinion selectivity appears to be a noticeable (but not very important) factor only in the case of tie removing, whereas its contribution to tie appearing is elusive. Our findings suggest that nodes’ activity is a crucial factor of network evolution—the majority of tie removing events can be explained by the age-based activity mechanism. Finally, we report that transitivity and preferential attachment do govern network evolution. However, there are two important details: (i) their zone of influence is restricted primarily by tie appearing and (ii) the preferential attachment mechanism is replaced by the anti-preferential attachment rule if the number of common peers is greater than zero.</span></p></div>","PeriodicalId":52228,"journal":{"name":"Online Social Networks and Media","volume":"34 ","pages":"Article 100248"},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49906485","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}

{"title":"Negativity spreads faster: A large-scale multilingual twitter analysis on the role of sentiment in political communication","authors":"Dimosthenis Antypas,&nbsp;Alun Preece,&nbsp;Jose Camacho-Collados","doi":"10.1016/j.osnem.2023.100242","DOIUrl":"https://doi.org/10.1016/j.osnem.2023.100242","url":null,"abstract":"<div><p>Social media has become extremely influential when it comes to policy making in modern societies, especially in the western world, where platforms such as Twitter allow users to follow politicians, thus making citizens more involved in political discussion. In the same vein, politicians use Twitter to express their opinions, debate among others on current topics and promote their political agendas aiming to influence voter behaviour. In this paper, we attempt to analyse tweets of politicians from three European countries and explore the virality of their tweets. Previous studies have shown that tweets conveying negative sentiment are likely to be retweeted more frequently. By utilising state-of-the-art pre-trained language models, we performed sentiment analysis on hundreds of thousands of tweets collected from members of parliament in Greece, Spain and the United Kingdom, including devolved administrations. We achieved this by systematically exploring and analysing the differences between influential and less popular tweets. Our analysis indicates that politicians’ negatively charged tweets spread more widely, especially in more recent times, and highlights interesting differences between political parties as well as between politicians and the general population.</p></div>","PeriodicalId":52228,"journal":{"name":"Online Social Networks and Media","volume":"33 ","pages":"Article 100242"},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49870346","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}

{"title":"Deep active learning for misinformation detection using geometric deep learning","authors":"Giorgio Barnabò ,&nbsp;Federico Siciliano ,&nbsp;Carlos Castillo ,&nbsp;Stefano Leonardi ,&nbsp;Preslav Nakov ,&nbsp;Giovanni Da San Martino ,&nbsp;Fabrizio Silvestri","doi":"10.1016/j.osnem.2023.100244","DOIUrl":"https://doi.org/10.1016/j.osnem.2023.100244","url":null,"abstract":"<div><p><span><span>Human fact-checkers currently represent a key component of any semi-automatic misinformation </span>detection pipeline<span><span>. While current state-of-the-art systems are mostly based on geometric deep-learning models, these architectures still need human-labeled data to be trained and updated — due to shifting topic distributions and adversarial attacks. Most research on automatic misinformation detection, however, neither considers time budget constraints on the number of pieces of news that can be manually fact-checked, nor tries to reduce the burden of fact-checking on – mostly pro bono – </span>annotators and journalists. The first contribution of this work is a thorough analysis of active learning (AL) strategies applied to </span></span>Graph Neural Networks (GNN) for misinformation detection. Then, based on this analysis, we propose Deep Error Sampling (DES) — a new deep active learning architecture that, when coupled with uncertainty sampling, performs equally or better than the most common AL strategies and the only existing active learning procedure specifically targeting fake news detection. Overall, our experimental results on two benchmark datasets show that all AL strategies outperform random sampling, allowing – on average – to achieve a 2% increase in AUC for the same percentage of third-party fact-checked news and to save up to 25% of labeling effort for a desired level of classification performance. As for DES, while it does not always clearly outperform other strategies, it still reduces variance in the performance between rounds, resulting in a more reliable method. To the best of our knowledge, we are the first to comprehensively study active learning in the context of misinformation detection and to show its potential to reduce the burden of third-party fact-checking without compromising classification performance.</p></div>","PeriodicalId":52228,"journal":{"name":"Online Social Networks and Media","volume":"33 ","pages":"Article 100244"},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49893561","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}