ABSTRACT Recent research on personalized retrieval technology has been actively conducted to meet the needs of users for seeking adequate information. To refine the retrieval, researchers are considering user behavior patterns in a variety of ways. In this study, we use eye‐tracking metadata to predict users' levels of comprehension as textual evidence for IR processes. Furthermore, we incorporated eye‐tracking metadata on the Automated Readability Index (ARI), a readability assessment tool of an English text. Our research is largely divided into two tasks: i) comprehension evaluation task (CET) and ii) comprehension‐based retrieval task (CRT). In the CET task, for predicting the comprehension level, we applied various regression models. Among them, the Voting regressor demonstrated the highest performance with a Spearman's 𝜌 of 0.68. In the CRT task, we incorporated the level of comprehension predicted in the CET task and ARI score into the ranking results. We derived a sentenceBERT to find the relevant text for a query and the Normalized Discounted Cumulative Gain (nDCG) for evaluating the CRT task. The nDCG score for Comprehension Level only and that with ARI together were 0.65 and 0.78, respectively. Thus, applying ARI resulted in a higher nDCG score compared to comprehension level only.
{"title":"Reading Comprehension in Information Retrieval (<scp>RCIR</scp>) for Personalized Results","authors":"Yumi Kim, Heesop Kim","doi":"10.1002/pra2.929","DOIUrl":"https://doi.org/10.1002/pra2.929","url":null,"abstract":"ABSTRACT Recent research on personalized retrieval technology has been actively conducted to meet the needs of users for seeking adequate information. To refine the retrieval, researchers are considering user behavior patterns in a variety of ways. In this study, we use eye‐tracking metadata to predict users' levels of comprehension as textual evidence for IR processes. Furthermore, we incorporated eye‐tracking metadata on the Automated Readability Index (ARI), a readability assessment tool of an English text. Our research is largely divided into two tasks: i) comprehension evaluation task (CET) and ii) comprehension‐based retrieval task (CRT). In the CET task, for predicting the comprehension level, we applied various regression models. Among them, the Voting regressor demonstrated the highest performance with a Spearman's 𝜌 of 0.68. In the CRT task, we incorporated the level of comprehension predicted in the CET task and ARI score into the ranking results. We derived a sentenceBERT to find the relevant text for a query and the Normalized Discounted Cumulative Gain (nDCG) for evaluating the CRT task. The nDCG score for Comprehension Level only and that with ARI together were 0.65 and 0.78, respectively. Thus, applying ARI resulted in a higher nDCG score compared to comprehension level only.","PeriodicalId":37833,"journal":{"name":"Proceedings of the Association for Information Science and Technology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136010295","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}
Haotian Hu, Alex Jie Yang, Sanhong Deng, Dongbo Wang, Min Song, Si Shen
ABSTRACT Drug–Drug Interaction (DDI) may affect the activity and efficacy of drugs, potentially leading to diminished therapeutic effect or even serious side effects. Therefore, automatic recognition of drug entities and relations involved in DDI is of great significance for pharmaceutical and medical care. In this paper, we propose a generative DDI triplets extraction framework based on Large Language Models (LLMs). We comprehensively apply various training methods, such as In‐context learning, Instruction‐tuning, and Task‐tuning, to investigate the biomedical information extraction capabilities of GPT‐3, OPT, and LLaMA. We also introduce Low‐Rank Adaptation (LoRA) technology to significantly reduce trainable parameters. The proposed method achieves satisfactory results in DDI triplet extraction, and demonstrates strong generalization ability on similar corpus.
{"title":"A Generative <scp>Drug–Drug</scp> Interaction Triplets Extraction Framework Based on Large Language Models","authors":"Haotian Hu, Alex Jie Yang, Sanhong Deng, Dongbo Wang, Min Song, Si Shen","doi":"10.1002/pra2.918","DOIUrl":"https://doi.org/10.1002/pra2.918","url":null,"abstract":"ABSTRACT Drug–Drug Interaction (DDI) may affect the activity and efficacy of drugs, potentially leading to diminished therapeutic effect or even serious side effects. Therefore, automatic recognition of drug entities and relations involved in DDI is of great significance for pharmaceutical and medical care. In this paper, we propose a generative DDI triplets extraction framework based on Large Language Models (LLMs). We comprehensively apply various training methods, such as In‐context learning, Instruction‐tuning, and Task‐tuning, to investigate the biomedical information extraction capabilities of GPT‐3, OPT, and LLaMA. We also introduce Low‐Rank Adaptation (LoRA) technology to significantly reduce trainable parameters. The proposed method achieves satisfactory results in DDI triplet extraction, and demonstrates strong generalization ability on similar corpus.","PeriodicalId":37833,"journal":{"name":"Proceedings of the Association for Information Science and Technology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136010986","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}
Hongyi Qin, Xiaojing Cai, Weikang Yuan, Cui Huang, Siqi Luo
ABSTRACT This study employs a scientometric approach to shed light on the evolving intellectual structure of AI in healthcare (AIH) research. The results substantiate the multi‐layered nature of knowledge production within the AIH domain, comprising the foundation, technology, and application layers. The application layer has witnessed a notable expansion in both its scope and depth, encompassing diverse areas including medical image analysis, data analysis and mining, decision support systems, and intelligence assistance. Additionally, a significant shift has occurred in its knowledge production process, wherein the conventional reliance on empiricism has been augmented by the incorporation of datafied innovation. This process of datafication has enriched the empirical underpinnings of AIH research, fostering a more comprehensive and evidence‐based approach to knowledge production.
{"title":"Scientific Knowledge Production and Artificial Intelligence for Healthcare: A Scientometric View","authors":"Hongyi Qin, Xiaojing Cai, Weikang Yuan, Cui Huang, Siqi Luo","doi":"10.1002/pra2.959","DOIUrl":"https://doi.org/10.1002/pra2.959","url":null,"abstract":"ABSTRACT This study employs a scientometric approach to shed light on the evolving intellectual structure of AI in healthcare (AIH) research. The results substantiate the multi‐layered nature of knowledge production within the AIH domain, comprising the foundation, technology, and application layers. The application layer has witnessed a notable expansion in both its scope and depth, encompassing diverse areas including medical image analysis, data analysis and mining, decision support systems, and intelligence assistance. Additionally, a significant shift has occurred in its knowledge production process, wherein the conventional reliance on empiricism has been augmented by the incorporation of datafied innovation. This process of datafication has enriched the empirical underpinnings of AIH research, fostering a more comprehensive and evidence‐based approach to knowledge production.","PeriodicalId":37833,"journal":{"name":"Proceedings of the Association for Information Science and Technology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136010998","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}
Jinhao Li, Yuxiang (Chris) Zhao, Yan Zhang, Xujie Ye
ABSTRACT Vicarious learning (VL), a concept widely explored in organizational behavior, has also received attention in social media research in recent years. Compared to live streaming e‐commerce and online communities, less VL has been studied based on danmaku, an instant commentary in video sites. We consider the characteristics of human information interaction in danmaku contexts and focus on exploring the core elements of coactive vicarious learning (CVL), where five core elements are identified. We will explore the influencing factors through content analysis and semi‐structured interviews. And a preliminary conceptual framework of CVL experience in danmaku contexts will try to be brought out for future empirical investigation.
{"title":"Coactive Vicarious Learning in Danmaku Contexts: A New Perspective of Informal Learning","authors":"Jinhao Li, Yuxiang (Chris) Zhao, Yan Zhang, Xujie Ye","doi":"10.1002/pra2.934","DOIUrl":"https://doi.org/10.1002/pra2.934","url":null,"abstract":"ABSTRACT Vicarious learning (VL), a concept widely explored in organizational behavior, has also received attention in social media research in recent years. Compared to live streaming e‐commerce and online communities, less VL has been studied based on danmaku, an instant commentary in video sites. We consider the characteristics of human information interaction in danmaku contexts and focus on exploring the core elements of coactive vicarious learning (CVL), where five core elements are identified. We will explore the influencing factors through content analysis and semi‐structured interviews. And a preliminary conceptual framework of CVL experience in danmaku contexts will try to be brought out for future empirical investigation.","PeriodicalId":37833,"journal":{"name":"Proceedings of the Association for Information Science and Technology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136011393","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}
ABSTRACT In the rapid flow of digital transformation, kiosks have naturally become part of our daily lives. In this study, we targeted visually impaired college students in their 20s, the primary user group of kiosks. We evaluated the usability of a self‐service certificate issuance and fast‐food restaurant kiosks. Based on Nielsen's five usability evaluation criteria, we presented experimental tasks to the visually impaired college students, monitored their performance, and conducted interviews to assess usability. Through this process, we aimed to understand the usage difficulties of visually impaired people when using kiosks and identify their specific requirements. Furthermore, we aimed to provide insights into improving the accessibility and usability of kiosks for this population and offer practical implications for developing kiosk education programs.
{"title":"Usability Evaluation of Kiosks for Visually Impaired College Students","authors":"Yumi Kim, Kyounghoon Kim, Jongwook Lee","doi":"10.1002/pra2.930","DOIUrl":"https://doi.org/10.1002/pra2.930","url":null,"abstract":"ABSTRACT In the rapid flow of digital transformation, kiosks have naturally become part of our daily lives. In this study, we targeted visually impaired college students in their 20s, the primary user group of kiosks. We evaluated the usability of a self‐service certificate issuance and fast‐food restaurant kiosks. Based on Nielsen's five usability evaluation criteria, we presented experimental tasks to the visually impaired college students, monitored their performance, and conducted interviews to assess usability. Through this process, we aimed to understand the usage difficulties of visually impaired people when using kiosks and identify their specific requirements. Furthermore, we aimed to provide insights into improving the accessibility and usability of kiosks for this population and offer practical implications for developing kiosk education programs.","PeriodicalId":37833,"journal":{"name":"Proceedings of the Association for Information Science and Technology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136011589","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}
ABSTRACT Online nonsuicidal self‐injury communities commonly create and share information on harm reduction strategies and exchange social support on social media platforms, including the short‐form video sharing platform TikTok. While TikTok's Community Guidelines permit users to share personal experiences with mental health topics, TikTok explicitly bans content depicting, promoting, normalizing, or glorifying activities that could lead to self‐harm. As such, TikTok may moderate user‐generated content, leading to exclusion and marginalization in this digital space. Through semi‐structured interviews with eight TikTok users with a history of nonsuicidal self‐injury, this pilot study explores how users experience TikTok's algorithm to create and engage with content on nonsuicidal self‐injury. Findings demonstrate that users understand how to circumnavigate TikTok's algorithm through algospeak (i.e., codewords or turns of phrases) and signaling to maintain visibility on the platform. Further, findings emphasize that users actively engage in self‐surveillance and self‐censorship to create a safe online community. In turn, content moderation can ultimately hinder progress toward the destigmatization of nonsuicidal self‐injury and restrict social support exchanged within online nonsuicidal self‐injury communities.
{"title":"Nonsuicidal <scp>Self‐Injury</scp> and Content Moderation on <scp>TikTok</scp>","authors":"Valerie Vera","doi":"10.1002/pra2.979","DOIUrl":"https://doi.org/10.1002/pra2.979","url":null,"abstract":"ABSTRACT Online nonsuicidal self‐injury communities commonly create and share information on harm reduction strategies and exchange social support on social media platforms, including the short‐form video sharing platform TikTok. While TikTok's Community Guidelines permit users to share personal experiences with mental health topics, TikTok explicitly bans content depicting, promoting, normalizing, or glorifying activities that could lead to self‐harm. As such, TikTok may moderate user‐generated content, leading to exclusion and marginalization in this digital space. Through semi‐structured interviews with eight TikTok users with a history of nonsuicidal self‐injury, this pilot study explores how users experience TikTok's algorithm to create and engage with content on nonsuicidal self‐injury. Findings demonstrate that users understand how to circumnavigate TikTok's algorithm through algospeak (i.e., codewords or turns of phrases) and signaling to maintain visibility on the platform. Further, findings emphasize that users actively engage in self‐surveillance and self‐censorship to create a safe online community. In turn, content moderation can ultimately hinder progress toward the destigmatization of nonsuicidal self‐injury and restrict social support exchanged within online nonsuicidal self‐injury communities.","PeriodicalId":37833,"journal":{"name":"Proceedings of the Association for Information Science and Technology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136009428","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}
Yiran Duan, Christy Khoury, Alexander O. Smith, Una Joh, Jeff Hemsley
ABSTRACT In January 2023, heavy California flooding prompted users to capture and share video footage of their impacted surroundings. This preliminary study uses this crisis event to compare commenting behavior across three video content formats: YouTube videos, YouTube shorts, and TikTok videos. Using network and regression analysis to study 45 videos across these three formats, we find that users commented and replied to others more on YouTube than TikTok despite TikTok videos having more views than YouTube videos. Additionally, we find the most vibrant comment behavior under YouTube shorts. This work provokes additional research to understand the exact ways in which platform design and affordances can influence crisis communication around a specific event.
{"title":"Comparing Crisis Communication on <scp>TikTok</scp> and <scp>YouTube</scp>: A Case Study of the 2023 California Floods","authors":"Yiran Duan, Christy Khoury, Alexander O. Smith, Una Joh, Jeff Hemsley","doi":"10.1002/pra2.908","DOIUrl":"https://doi.org/10.1002/pra2.908","url":null,"abstract":"ABSTRACT In January 2023, heavy California flooding prompted users to capture and share video footage of their impacted surroundings. This preliminary study uses this crisis event to compare commenting behavior across three video content formats: YouTube videos, YouTube shorts, and TikTok videos. Using network and regression analysis to study 45 videos across these three formats, we find that users commented and replied to others more on YouTube than TikTok despite TikTok videos having more views than YouTube videos. Additionally, we find the most vibrant comment behavior under YouTube shorts. This work provokes additional research to understand the exact ways in which platform design and affordances can influence crisis communication around a specific event.","PeriodicalId":37833,"journal":{"name":"Proceedings of the Association for Information Science and Technology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136009429","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}
Nosheen Warraich, Nadia Caidi, Bharat Mehra, Cansu Ekmekcioglu, Irfan Ali
ABSTRACT Academics argue that the COVID‐19 pandemic has limited students' ability to learn, with significant digital inequities occurring between students from the global North and the global South. Students and academics from developing nations encountered particular challenges and difficulties with the move toward online styles of learning. Much like their colleagues from developed countries, they were unprepared for this predicament, but on top of the crisis context, deeper issues were having to do with digital inequalities and disparities that were exacerbated by the inadequate digital infrastructure (smart devices/gadgets, internet access, and speed) and online interaction abilities and practices. The goal of this panel is to address the pressing issue of digital inclusion in online education, specifically the broader challenge of ensuring that online education is accessible to all. As information researchers continue to work towards enhancing online learning, it is crucial to address the disparities in the sharing of information and knowledge and to bridge the gaps that exist across communities and nations. The panelists (three of whom work in developed countries and two in developing countries) will relate their experiences and viewpoints thus bringing their knowledge to bear in examining the concepts of digital inequality and digital inclusion. The rest of the session will be devoted to discussions and brainstorming with attendees around these issues, with special attention being given to perspectives that seek to bridge the disparities and promote inclusion in education.
{"title":"Digital Inequalities to Digital Inclusion in Online Learning: Viewpoints of <scp>LIS</scp> Educators Seeking to Bridge the Disparities","authors":"Nosheen Warraich, Nadia Caidi, Bharat Mehra, Cansu Ekmekcioglu, Irfan Ali","doi":"10.1002/pra2.876","DOIUrl":"https://doi.org/10.1002/pra2.876","url":null,"abstract":"ABSTRACT Academics argue that the COVID‐19 pandemic has limited students' ability to learn, with significant digital inequities occurring between students from the global North and the global South. Students and academics from developing nations encountered particular challenges and difficulties with the move toward online styles of learning. Much like their colleagues from developed countries, they were unprepared for this predicament, but on top of the crisis context, deeper issues were having to do with digital inequalities and disparities that were exacerbated by the inadequate digital infrastructure (smart devices/gadgets, internet access, and speed) and online interaction abilities and practices. The goal of this panel is to address the pressing issue of digital inclusion in online education, specifically the broader challenge of ensuring that online education is accessible to all. As information researchers continue to work towards enhancing online learning, it is crucial to address the disparities in the sharing of information and knowledge and to bridge the gaps that exist across communities and nations. The panelists (three of whom work in developed countries and two in developing countries) will relate their experiences and viewpoints thus bringing their knowledge to bear in examining the concepts of digital inequality and digital inclusion. The rest of the session will be devoted to discussions and brainstorming with attendees around these issues, with special attention being given to perspectives that seek to bridge the disparities and promote inclusion in education.","PeriodicalId":37833,"journal":{"name":"Proceedings of the Association for Information Science and Technology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136009430","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}
Kornelija Petr Balog, Sanjica Faletar, Tomislav Jakopec
ABSTRACT Digital technology has a great potential for assisting older people in their everyday tasks and general well‐being. However, older adults are relatively slow to adopt the new technology and one of the obstacles may be their negative perception or perceived uselessness of the technology. The paper presents preliminary findings from a study into the attitudes toward digital technology and its perceived usefulness among the older adults in the city of Osijek, Croatia. Results show that majority of respondents have positive attitudes toward digital technology and majority perceives it as very useful. The study identified a number of factors, such as age, gender, education and quality of life that affect the respondents’ varying attitudes toward digital technology and its perceived usefulness. The research findings can help policy makers and local institutions such as libraries in designing digital literacy courses and provision of support to older adults.
{"title":"Older Adults’ Attitudes toward Digital Technology and Perceptions of Its Usefulness: Example of the City of Osijek, Croatia","authors":"Kornelija Petr Balog, Sanjica Faletar, Tomislav Jakopec","doi":"10.1002/pra2.840","DOIUrl":"https://doi.org/10.1002/pra2.840","url":null,"abstract":"ABSTRACT Digital technology has a great potential for assisting older people in their everyday tasks and general well‐being. However, older adults are relatively slow to adopt the new technology and one of the obstacles may be their negative perception or perceived uselessness of the technology. The paper presents preliminary findings from a study into the attitudes toward digital technology and its perceived usefulness among the older adults in the city of Osijek, Croatia. Results show that majority of respondents have positive attitudes toward digital technology and majority perceives it as very useful. The study identified a number of factors, such as age, gender, education and quality of life that affect the respondents’ varying attitudes toward digital technology and its perceived usefulness. The research findings can help policy makers and local institutions such as libraries in designing digital literacy courses and provision of support to older adults.","PeriodicalId":37833,"journal":{"name":"Proceedings of the Association for Information Science and Technology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136009551","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}
ABSTRACT Polarization of views (known as ideological polarization) is one of the greatest societal challenges of our time, potentially sewing distrust and hate among individuals and communities and threatening to undermine the fabric of democracy. Divisive issues such as abortion and gun control are ever‐present and can drive issue polarization, and even affective polarization—a disdain for ‘the other side,’ which can further divide society. Social media has been flagged as a breeding ground for polarized views, with private groups and personalized algorithms facilitating self‐creation of echo chambers that may lead to polarization. While there is prior research on the technological influences on view strengthening, scant Human‐centered research exists and most of it has focused on view change in general, rather than view strengthening specifically. To investigate peoples' experiences of view strengthening on social media, we interviewed 10 people who recently strengthened their views on important topics. While some took steps to avoid creating echo chambers (e.g., by seeking out opposing views), others intentionally created them to allow their views to strengthen without interference. These findings have important implications for designing social media platforms that support careful and conscious view strengthening while mitigating against the risk of information manipulation.
{"title":"Stronger Than Yesterday: Investigating Peoples' Experiences of View Strengthening on Social Media","authors":"Sabrina Beall, Stephann Makri, Dana McKay","doi":"10.1002/pra2.767","DOIUrl":"https://doi.org/10.1002/pra2.767","url":null,"abstract":"ABSTRACT Polarization of views (known as ideological polarization) is one of the greatest societal challenges of our time, potentially sewing distrust and hate among individuals and communities and threatening to undermine the fabric of democracy. Divisive issues such as abortion and gun control are ever‐present and can drive issue polarization, and even affective polarization—a disdain for ‘the other side,’ which can further divide society. Social media has been flagged as a breeding ground for polarized views, with private groups and personalized algorithms facilitating self‐creation of echo chambers that may lead to polarization. While there is prior research on the technological influences on view strengthening, scant Human‐centered research exists and most of it has focused on view change in general, rather than view strengthening specifically. To investigate peoples' experiences of view strengthening on social media, we interviewed 10 people who recently strengthened their views on important topics. While some took steps to avoid creating echo chambers (e.g., by seeking out opposing views), others intentionally created them to allow their views to strengthen without interference. These findings have important implications for designing social media platforms that support careful and conscious view strengthening while mitigating against the risk of information manipulation.","PeriodicalId":37833,"journal":{"name":"Proceedings of the Association for Information Science and Technology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136009557","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：