A. Munk, M. Abildgaard, Andreas Birkbak, M. Petersen
The paper discusses three ways of appropriating Instagram for social research through the case of obesity. We draw on the notion obesogenic environments, in which obesity is understood as related to a wide range of cultural, social and physical factors. Together with a group of obesity researchers and cultural analysts we explored a dataset of 82,449 Instagram posts tagged with location from the five most and the five least overweight local authorities in the UK. The geo-located posts were studied through three distinct approaches to the data; each drawing on their own set of interdependent conceptualizations of what constitutes obesogenic environments, Instagram and cultural analysis respectively. The first approach values Instagram as a repository of images that can be coded and counted; while the second asks about the everyday practices of Instagram users. In a third approach we view Instagram itself as an analytical tool that produces a media-specific version of phenomena such as obesity. Following this third appropriation, we conclude that to unlock Instagram's potential for social research it must be considered as more than a collection of user-tagged images, but as an analytical context in its own right.
{"title":"(Re-)Appropriating Instagram for Social Research: Three Methods for Studying Obesogenic Environments","authors":"A. Munk, M. Abildgaard, Andreas Birkbak, M. Petersen","doi":"10.1145/2930971.2930991","DOIUrl":"https://doi.org/10.1145/2930971.2930991","url":null,"abstract":"The paper discusses three ways of appropriating Instagram for social research through the case of obesity. We draw on the notion obesogenic environments, in which obesity is understood as related to a wide range of cultural, social and physical factors. Together with a group of obesity researchers and cultural analysts we explored a dataset of 82,449 Instagram posts tagged with location from the five most and the five least overweight local authorities in the UK. The geo-located posts were studied through three distinct approaches to the data; each drawing on their own set of interdependent conceptualizations of what constitutes obesogenic environments, Instagram and cultural analysis respectively. The first approach values Instagram as a repository of images that can be coded and counted; while the second asks about the everyday practices of Instagram users. In a third approach we view Instagram itself as an analytical tool that produces a media-specific version of phenomena such as obesity. Following this third appropriation, we conclude that to unlock Instagram's potential for social research it must be considered as more than a collection of user-tagged images, but as an analytical context in its own right.","PeriodicalId":227482,"journal":{"name":"Proceedings of the 7th 2016 International Conference on Social Media & Society","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122764820","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}
K. Glasgow, Jessica Vitak, Y. Tausczik, Clayton Fink
In the aftermath of a traumatic mass casualty event, a community's resources are strained, while its needs for tangible, emotional, and informational support are elevated. Social media may serve to bridge the distance between the locally affected community and those outside who are willing to offer support. This exploratory study uses Twitter as a lens for examining gratitude for support in the aftermath of disaster. We examine how social media may provide new opportunities for support to be exchanged and networks to be formed in the aftermath of a traumatic event. By analyzing tweets originating from Newtown, Connecticut after the school shooting, we identify and describe six categories of support exchanged through Twitter, including two categories (symbolic and role-based) that have not been extensively discussed in the social support literature -- but are valued by the community. Each type of support network shows distinct structural characteristics and temporal variance.
{"title":"Grieving in the 21st Century: Social Media's Role in Facilitating Supportive Exchanges Following Community-Level Traumatic Events","authors":"K. Glasgow, Jessica Vitak, Y. Tausczik, Clayton Fink","doi":"10.1145/2930971.2930975","DOIUrl":"https://doi.org/10.1145/2930971.2930975","url":null,"abstract":"In the aftermath of a traumatic mass casualty event, a community's resources are strained, while its needs for tangible, emotional, and informational support are elevated. Social media may serve to bridge the distance between the locally affected community and those outside who are willing to offer support. This exploratory study uses Twitter as a lens for examining gratitude for support in the aftermath of disaster. We examine how social media may provide new opportunities for support to be exchanged and networks to be formed in the aftermath of a traumatic event. By analyzing tweets originating from Newtown, Connecticut after the school shooting, we identify and describe six categories of support exchanged through Twitter, including two categories (symbolic and role-based) that have not been extensively discussed in the social support literature -- but are valued by the community. Each type of support network shows distinct structural characteristics and temporal variance.","PeriodicalId":227482,"journal":{"name":"Proceedings of the 7th 2016 International Conference on Social Media & Society","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122769317","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}
Using Facebook Groups to connect otherwise anonymous people that live in a single urban development is a relatively new phenomenon. Within residential developments there are a number of common management and performance issues experienced by many isolated inhabitants that are identified through building performance evaluation studies. Facebook is a ubiquitous social network tool and powerful communication platform, particularly popular among young adults. This paper explores the use of Facebook in relation to management and performance issues in two cases of Facebook Group usage within residential communities in the UK. Data was collected through longitudinal digital and physical visits to the residential communities and to the Facebook Group sites. Findings are presented in relation to home learning, site/neighbourhood and self-organising initiatives. We propose that weak-tie residential communities can develop collective efficacy and work together for the overall good of the residential development through communicating on a Facebook Group. This helps to improve the physical environment, facilitating further collective action. There is a clear overlap between social media narrative and the physical experience of daily life, which can help to empower residents.
{"title":"Exploring the Relationship Between a 'Facebook Group' and Face-to-Face Interactions in 'Weak-Tie' Residential Communities","authors":"M. Baborska-Narożny, E. Stirling, F. Stevenson","doi":"10.1145/2930971.2930989","DOIUrl":"https://doi.org/10.1145/2930971.2930989","url":null,"abstract":"Using Facebook Groups to connect otherwise anonymous people that live in a single urban development is a relatively new phenomenon. Within residential developments there are a number of common management and performance issues experienced by many isolated inhabitants that are identified through building performance evaluation studies. Facebook is a ubiquitous social network tool and powerful communication platform, particularly popular among young adults. This paper explores the use of Facebook in relation to management and performance issues in two cases of Facebook Group usage within residential communities in the UK. Data was collected through longitudinal digital and physical visits to the residential communities and to the Facebook Group sites. Findings are presented in relation to home learning, site/neighbourhood and self-organising initiatives. We propose that weak-tie residential communities can develop collective efficacy and work together for the overall good of the residential development through communicating on a Facebook Group. This helps to improve the physical environment, facilitating further collective action. There is a clear overlap between social media narrative and the physical experience of daily life, which can help to empower residents.","PeriodicalId":227482,"journal":{"name":"Proceedings of the 7th 2016 International Conference on Social Media & Society","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116941517","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}
The rise of social networking mobile applications (apps) (e.g, Grindr, Jack'd) created specifically for men who have sex with men (MSM) has generated public health concerns and conflicting studies about the impact on risky sexual behaviors. This study seeks to gain a more precise understanding of why and how MSM are using social networking mobile apps, and informs the theoretical debate concerning the impact of social networking technology on sexual risk behaviors. A questionnaire survey was conducted, both online and offline, with young MSM app-users in Hong Kong to examine their apps use (frequency, history, and exposure of own face and body) and recent sexual partnering via apps (total sexual partners [TSP] and condomless sex partners [CSP]) in relation to gratifications sought and sexual sensation seeking. The results indicated that finding sexual partners was not a high priority for using MSM apps; surveillance, relationship, and diversion motives were more important while social motive shared similar importance. App-use frequency, sex motive, and sexual sensation seeking predicted more TSP while surveillance motive predicted fewer TSP. None of these variables, however, directly predicted CSP. Sexual sensation seeking in interaction with sex or diversion motive predicted both TSP and CSP. Despite lacking significant association with sex motive or sexual sensation seeking, app-use frequency was a stronger independent predictor of TSP. While frequent app use may facilitate more app-met sexual partners, this study found no evidence indicating that app use promotes riskier sexual behavior with those partners.
{"title":"The Roles of Sensation Seeking and Gratifications Sought in Social Networking Apps Use and Attendant Sexual Behaviors","authors":"T. E. D. Yeo, Y. Ng","doi":"10.1145/2930971.2930990","DOIUrl":"https://doi.org/10.1145/2930971.2930990","url":null,"abstract":"The rise of social networking mobile applications (apps) (e.g, Grindr, Jack'd) created specifically for men who have sex with men (MSM) has generated public health concerns and conflicting studies about the impact on risky sexual behaviors. This study seeks to gain a more precise understanding of why and how MSM are using social networking mobile apps, and informs the theoretical debate concerning the impact of social networking technology on sexual risk behaviors. A questionnaire survey was conducted, both online and offline, with young MSM app-users in Hong Kong to examine their apps use (frequency, history, and exposure of own face and body) and recent sexual partnering via apps (total sexual partners [TSP] and condomless sex partners [CSP]) in relation to gratifications sought and sexual sensation seeking. The results indicated that finding sexual partners was not a high priority for using MSM apps; surveillance, relationship, and diversion motives were more important while social motive shared similar importance. App-use frequency, sex motive, and sexual sensation seeking predicted more TSP while surveillance motive predicted fewer TSP. None of these variables, however, directly predicted CSP. Sexual sensation seeking in interaction with sex or diversion motive predicted both TSP and CSP. Despite lacking significant association with sex motive or sexual sensation seeking, app-use frequency was a stronger independent predictor of TSP. While frequent app use may facilitate more app-met sexual partners, this study found no evidence indicating that app use promotes riskier sexual behavior with those partners.","PeriodicalId":227482,"journal":{"name":"Proceedings of the 7th 2016 International Conference on Social Media & Society","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114272154","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}
This paper investigates the characteristics of rumor-related tweets that would attract retransmission. Drawing on the uses and gratifications (U & G) and influential users' theories, it proposes a rumor retransmission model which comprises variables associated to the source and the message of the tweets. From a total of 5,885 rumor-related tweets about the death of the founding Prime Minister of Singapore Lee Kuan Yew collected, 800 original tweets without a "RT" prefix were selected for analysis. It was found that the experience and connectivity of the source are correlated to retransmission. The age of the account and number of followers have positive relationships with retransmission, while the number of tweets posted and number of friends have negative relationships. For characteristics of the message, attractiveness and expressions of sense of belonging are positively related to retransmission but the use of directed messages and medium-specific features, namely hashtags and URLs have negative relationships with retransmission. Messages with low level of emotions trigger more retweets than those without emotional expressions and those that are highly emotional. Results could suggest that in inauspicious contexts such as the rumored death of a political figure, Twitter users appear to favor less complex content and seem to be discerning and rational in making retransmission decisions.
本文研究了与谣言相关的推文吸引转发的特征。借鉴使用与满足理论(uses and gratifications, U & G)和有影响力的用户理论，提出了一个谣言转发模型，该模型包含与推文来源和消息相关的变量。从总共收集到的5885条关于新加坡开国总理李光耀去世的与谣言有关的推文中，选择了800条没有“RT”前缀的原创推文进行分析。研究发现，源的经验和连通性与重传有关。账号年龄和关注者数量与转发呈正相关关系，发文数和好友数与转发呈负相关关系。就信息的特征而言，吸引力和归属感的表达与转发呈正相关，而定向信息的使用和媒介特定特征，即标签和url与转发呈负相关。与没有情绪表达和高度情绪表达的信息相比，低情绪表达的信息会引发更多的转发。结果可能表明，在不吉利的情况下，比如一位政治人物去世的传闻，Twitter用户似乎更喜欢不那么复杂的内容，在做出转发决定时似乎更有洞察力和理性。
{"title":"The Retransmission of Rumor-related Tweets: Characteristics of Source and Message","authors":"A. Chua, Cheng-Ying Tee, A. Pang, Ee-Peng Lim","doi":"10.1145/2930971.2930994","DOIUrl":"https://doi.org/10.1145/2930971.2930994","url":null,"abstract":"This paper investigates the characteristics of rumor-related tweets that would attract retransmission. Drawing on the uses and gratifications (U & G) and influential users' theories, it proposes a rumor retransmission model which comprises variables associated to the source and the message of the tweets. From a total of 5,885 rumor-related tweets about the death of the founding Prime Minister of Singapore Lee Kuan Yew collected, 800 original tweets without a \"RT\" prefix were selected for analysis. It was found that the experience and connectivity of the source are correlated to retransmission. The age of the account and number of followers have positive relationships with retransmission, while the number of tweets posted and number of friends have negative relationships. For characteristics of the message, attractiveness and expressions of sense of belonging are positively related to retransmission but the use of directed messages and medium-specific features, namely hashtags and URLs have negative relationships with retransmission. Messages with low level of emotions trigger more retweets than those without emotional expressions and those that are highly emotional. Results could suggest that in inauspicious contexts such as the rumored death of a political figure, Twitter users appear to favor less complex content and seem to be discerning and rational in making retransmission decisions.","PeriodicalId":227482,"journal":{"name":"Proceedings of the 7th 2016 International Conference on Social Media & Society","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125814194","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}
In this paper we study identity-based motivation in Wikipedia as a drive for editors to act congruently with their cultural identity values by contributing with content related to them. To assess its influence, we developed a computational method to identify articles related to the cultural identities associated to a language and applied it to 40 Wikipedia language editions. The results show that about a quarter of each Wikipedia language edition is dedicated to represent the corresponding cultural identities. The topical coverage of these articles reflects that geography, biographies, and culture are the most common themes, although each language shows its idiosyncrasy and other topics are also present. The majority of these articles remain exclusive to each language, which is consistent with the idea that a Cultural Identity is defined in relation to others; as entangled and separated. An analysis of how this content is shared among language editions reveals special links between cultures. The approach and findings presented in this study can help to foster participation and inter-cultural enrichment of Wikipedias. The datasets produced in this study are made available for further research.
{"title":"Cultural Identities in Wikipedias","authors":"Marc Miquel-Ribé, David Laniado","doi":"10.1145/2930971.2930996","DOIUrl":"https://doi.org/10.1145/2930971.2930996","url":null,"abstract":"In this paper we study identity-based motivation in Wikipedia as a drive for editors to act congruently with their cultural identity values by contributing with content related to them. To assess its influence, we developed a computational method to identify articles related to the cultural identities associated to a language and applied it to 40 Wikipedia language editions. The results show that about a quarter of each Wikipedia language edition is dedicated to represent the corresponding cultural identities. The topical coverage of these articles reflects that geography, biographies, and culture are the most common themes, although each language shows its idiosyncrasy and other topics are also present. The majority of these articles remain exclusive to each language, which is consistent with the idea that a Cultural Identity is defined in relation to others; as entangled and separated. An analysis of how this content is shared among language editions reveals special links between cultures. The approach and findings presented in this study can help to foster participation and inter-cultural enrichment of Wikipedias. The datasets produced in this study are made available for further research.","PeriodicalId":227482,"journal":{"name":"Proceedings of the 7th 2016 International Conference on Social Media & Society","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125442037","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}
Recent advances in technology have changed the way people use Internet. Customers use online platforms to socialize with friends and family, interact with new people and to gather latest information from all around the world. People coming together in online platforms give rise to virtual communities. In this paper we highlight the importance of online brand communities and compare community leaders across brands. Understanding the notion of influencers is important for marketers as these leaders help to create a reliable brand community that can resonate with consumers' desire to build brand loyalty and devotion. Using three different brands of jeans we examined online communities formed within Twitter, using Sysomos. The results showed that there is not much overlap with influencers across brands. However, influencers that are common across brands are very similar in structure and communication strategies.
{"title":"Exploring the Similarities of Influencers in Online Brand Communities","authors":"Tasmina Afroze","doi":"10.1145/2930971.2930981","DOIUrl":"https://doi.org/10.1145/2930971.2930981","url":null,"abstract":"Recent advances in technology have changed the way people use Internet. Customers use online platforms to socialize with friends and family, interact with new people and to gather latest information from all around the world. People coming together in online platforms give rise to virtual communities. In this paper we highlight the importance of online brand communities and compare community leaders across brands. Understanding the notion of influencers is important for marketers as these leaders help to create a reliable brand community that can resonate with consumers' desire to build brand loyalty and devotion. Using three different brands of jeans we examined online communities formed within Twitter, using Sysomos. The results showed that there is not much overlap with influencers across brands. However, influencers that are common across brands are very similar in structure and communication strategies.","PeriodicalId":227482,"journal":{"name":"Proceedings of the 7th 2016 International Conference on Social Media & Society","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123624110","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}
Online third party marketplaces link buyers and sellers by providing a neutral platform for exchange. However, this requires buyers to assess the quality of goods without being able to handle or sample them. Recent research has proposed extending the warranting principle, an emerging theory of online interpersonal impression formation, to the judgements consumer make about products online. The warranting principle holds that information that is more difficult to manufacture or manipulate, such as an individualized photo of an auction item, should be more informative to a consumers judgement of the product than information which is easy to falsify or manipulate. While this theoretical extension assumes that all online goods are capable of being assessed in the same way, some goods (i.e. experience goods) are more difficult to assess in online marketplaces because they must be experienced before their true quality and characteristics can be known. The current study examines 2,401 completed auctions from eBay revealing that the relationship between warranting cues and price discounts is moderated by the type of good being sold. The theoretical contributions and limitations of the study are discussed.
{"title":"Testing the Application of Warranting Theory to Online Third Party Marketplaces: The Effects of Information Uniqueness and Product Type","authors":"W. S. Sanders, Gopi Chand Nutakki, O. Nasraoui","doi":"10.1145/2930971.2930988","DOIUrl":"https://doi.org/10.1145/2930971.2930988","url":null,"abstract":"Online third party marketplaces link buyers and sellers by providing a neutral platform for exchange. However, this requires buyers to assess the quality of goods without being able to handle or sample them. Recent research has proposed extending the warranting principle, an emerging theory of online interpersonal impression formation, to the judgements consumer make about products online. The warranting principle holds that information that is more difficult to manufacture or manipulate, such as an individualized photo of an auction item, should be more informative to a consumers judgement of the product than information which is easy to falsify or manipulate. While this theoretical extension assumes that all online goods are capable of being assessed in the same way, some goods (i.e. experience goods) are more difficult to assess in online marketplaces because they must be experienced before their true quality and characteristics can be known. The current study examines 2,401 completed auctions from eBay revealing that the relationship between warranting cues and price discounts is moderated by the type of good being sold. The theoretical contributions and limitations of the study are discussed.","PeriodicalId":227482,"journal":{"name":"Proceedings of the 7th 2016 International Conference on Social Media & Society","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123094215","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}
Christopher J. Zimmerman, René Madsen, Henrik Hammer Eliassen, Ravikiran Vatrapu
This paper applies urban informatics methods and techniques on big data generated from the concentrated environment of the second largest music festival in the world, Roskilde Festival. First, we explain how to utilize relevant dimensions from human geography theories towards mapping a 'Geography of Importance'. Second, we elaborate on methods deployed for collecting both mobile GPS and social media traces that the smart phone generates in physical spaces. Third, we compare and contrast the automatically geocoded presence in space and at events with the intentionally socially tagged consumption of these spaces and events as place-based experiences. In doing so, these two layers of space-based movements and place-based experiences reveal the appropriation of affordances and choices of aesthetic appreciation by the crowd at large of what is subjectively and relatively meaningful, actionable, and valuable.
{"title":"Space vs. Place: Comparing Space-based Movements and Place-based Experiences at the Roskilde Festival 2015","authors":"Christopher J. Zimmerman, René Madsen, Henrik Hammer Eliassen, Ravikiran Vatrapu","doi":"10.1145/2930971.2930983","DOIUrl":"https://doi.org/10.1145/2930971.2930983","url":null,"abstract":"This paper applies urban informatics methods and techniques on big data generated from the concentrated environment of the second largest music festival in the world, Roskilde Festival. First, we explain how to utilize relevant dimensions from human geography theories towards mapping a 'Geography of Importance'. Second, we elaborate on methods deployed for collecting both mobile GPS and social media traces that the smart phone generates in physical spaces. Third, we compare and contrast the automatically geocoded presence in space and at events with the intentionally socially tagged consumption of these spaces and events as place-based experiences. In doing so, these two layers of space-based movements and place-based experiences reveal the appropriation of affordances and choices of aesthetic appreciation by the crowd at large of what is subjectively and relatively meaningful, actionable, and valuable.","PeriodicalId":227482,"journal":{"name":"Proceedings of the 7th 2016 International Conference on Social Media & Society","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134116951","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}
Vassilis Plachouras, Jochen L. Leidner, Andrew G. Garrow
When a drug that is sold exhibits side effects, a well functioning ecosystem of pharmaceutical drug suppliers includes responsive regulators and pharmaceutical companies. Existing systems for monitoring adverse drug events, such as the Federal Adverse Events Reporting System (FAERS) in the US, have shown limited effectiveness due to the lack of incentives for healthcare professionals and patients. While social media present opportunities to mine information about adverse events in near real-time, there are still important questions to be answered in order to understand their impact on pharmacovigilance. First, it is not known how many relevant social media posts occur per day on platforms like Twitter, i.e., whether there is "enough signal" for a post-market pharmacovigilance program based on Twitter mining. Second, it is not known what other topics are discussed by users in posts mentioning pharmaceutical drugs. In this paper, we outline how social media can be used as a human sensor for drug use monitoring. We introduce a large-scale, near real-time system for computational pharmacovigilance, and use our system to estimate the order of magnitude of the volume of daily self-reported pharmaceutical drug side effect tweets. The processing pipeline comprises a set of cascaded filters, followed by a supervised machine learning classifier. The cascaded filters quickly reduce the volume to a manageable sub-stream, from which a Support Vector Machine (SVM) based classifier identifies adverse events based on a rich set of features taking into account surface-textual properties, as well as domain knowledge about drugs, side effects and the Twitter medium. Using a dataset of 10,000 manually annotated tweets, a SVM classifier achieves F1=60.4% and AUC=0.894. The yield of the classifier for a drug universe comprising 2,600 keywords is 721 tweets per day. We also investigate what other topics are discussed in the posts mentioning pharmaceutical drugs. We conclude by suggesting an ecosystem where regulators and pharmaceutical companies utilize social media to obtain feedback about consequences of pharmaceutical drug use.
{"title":"Quantifying Self-Reported Adverse Drug Events on Twitter: Signal and Topic Analysis","authors":"Vassilis Plachouras, Jochen L. Leidner, Andrew G. Garrow","doi":"10.1145/2930971.2930977","DOIUrl":"https://doi.org/10.1145/2930971.2930977","url":null,"abstract":"When a drug that is sold exhibits side effects, a well functioning ecosystem of pharmaceutical drug suppliers includes responsive regulators and pharmaceutical companies. Existing systems for monitoring adverse drug events, such as the Federal Adverse Events Reporting System (FAERS) in the US, have shown limited effectiveness due to the lack of incentives for healthcare professionals and patients. While social media present opportunities to mine information about adverse events in near real-time, there are still important questions to be answered in order to understand their impact on pharmacovigilance. First, it is not known how many relevant social media posts occur per day on platforms like Twitter, i.e., whether there is \"enough signal\" for a post-market pharmacovigilance program based on Twitter mining. Second, it is not known what other topics are discussed by users in posts mentioning pharmaceutical drugs. In this paper, we outline how social media can be used as a human sensor for drug use monitoring. We introduce a large-scale, near real-time system for computational pharmacovigilance, and use our system to estimate the order of magnitude of the volume of daily self-reported pharmaceutical drug side effect tweets. The processing pipeline comprises a set of cascaded filters, followed by a supervised machine learning classifier. The cascaded filters quickly reduce the volume to a manageable sub-stream, from which a Support Vector Machine (SVM) based classifier identifies adverse events based on a rich set of features taking into account surface-textual properties, as well as domain knowledge about drugs, side effects and the Twitter medium. Using a dataset of 10,000 manually annotated tweets, a SVM classifier achieves F1=60.4% and AUC=0.894. The yield of the classifier for a drug universe comprising 2,600 keywords is 721 tweets per day. We also investigate what other topics are discussed in the posts mentioning pharmaceutical drugs. We conclude by suggesting an ecosystem where regulators and pharmaceutical companies utilize social media to obtain feedback about consequences of pharmaceutical drug use.","PeriodicalId":227482,"journal":{"name":"Proceedings of the 7th 2016 International Conference on Social Media & Society","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129110181","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：