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":"39 1","pages":"0"},"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}
This paper focuses on the importance of non-public social media spaces in contemporary democratic participation at the grassroots level, based on case studies of citizen-led, community and activist groups. The research pilots the concept of participation spaces to reify online and offline contexts where people participate in democracy. Participation spaces include social media presences, websites, blogs, email, paper media, and physical spaces. This approach enables the parallel study of diverse spaces (more or less public; on and offline). Participation spaces were investigated across three local groups, through interviews and participant observation; then modelled as Socio-Technical Interaction Networks (STINs) [1]. This research provides an alternative and richer picture of social media use, within eParticipation, to studies solely based on public Internet content, such as data sets of tweets. In the participation spaces studies most communication takes place in non-public contexts, such as closed Facebook groups, email, and face-to-face meetings. Non-public social media spaces are particularly effective in supporting collaboration between people from diverse social groups. These spaces can be understood as boundary objects [2] and play strong roles in democracy.
{"title":"Non-public eParticipation in Social Media Spaces","authors":"Ella Taylor-Smith, Colin F. Smith","doi":"10.1145/2930971.2930974","DOIUrl":"https://doi.org/10.1145/2930971.2930974","url":null,"abstract":"This paper focuses on the importance of non-public social media spaces in contemporary democratic participation at the grassroots level, based on case studies of citizen-led, community and activist groups. The research pilots the concept of participation spaces to reify online and offline contexts where people participate in democracy. Participation spaces include social media presences, websites, blogs, email, paper media, and physical spaces. This approach enables the parallel study of diverse spaces (more or less public; on and offline). Participation spaces were investigated across three local groups, through interviews and participant observation; then modelled as Socio-Technical Interaction Networks (STINs) [1]. This research provides an alternative and richer picture of social media use, within eParticipation, to studies solely based on public Internet content, such as data sets of tweets. In the participation spaces studies most communication takes place in non-public contexts, such as closed Facebook groups, email, and face-to-face meetings. Non-public social media spaces are particularly effective in supporting collaboration between people from diverse social groups. These spaces can be understood as boundary objects [2] and play strong roles in democracy.","PeriodicalId":227482,"journal":{"name":"Proceedings of the 7th 2016 International Conference on Social Media & Society","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128415688","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}
Sociological studies on the Internet have often examined digital inequalities. These studies show how Internet access, skills, uses and outcomes vary between different population segments. However, we know more about social inequalities in general Internet use than in social media use. Especially, we lack differentiated statistical evidence of the social profiles of distinct social media platforms. To address this issue, we use a large survey data set in the United Kingdom and investigate the social structuration of six major social media platforms. We find that age and socio-economic status are driving forces of several -- but not all -- of these platforms. Aggregating platform adoption into a general measure of social media use blurs some of the subtleties of more fine-grained indicators, namely platform uses and specific activities, such as status updating and commenting.
{"title":"The Social Structuration of Six Major Social Media Platforms in the United Kingdom: Facebook, LinkedIn, Twitter, Instagram, Google+ and Pinterest","authors":"Grant Blank, C. Lutz","doi":"10.1145/2930971.2930979","DOIUrl":"https://doi.org/10.1145/2930971.2930979","url":null,"abstract":"Sociological studies on the Internet have often examined digital inequalities. These studies show how Internet access, skills, uses and outcomes vary between different population segments. However, we know more about social inequalities in general Internet use than in social media use. Especially, we lack differentiated statistical evidence of the social profiles of distinct social media platforms. To address this issue, we use a large survey data set in the United Kingdom and investigate the social structuration of six major social media platforms. We find that age and socio-economic status are driving forces of several -- but not all -- of these platforms. Aggregating platform adoption into a general measure of social media use blurs some of the subtleties of more fine-grained indicators, namely platform uses and specific activities, such as status updating and commenting.","PeriodicalId":227482,"journal":{"name":"Proceedings of the 7th 2016 International Conference on Social Media & Society","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123413030","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}
Mobile dating apps with geolocative function have gained popularity for fostering social, romantic and sexual connections between proximate strangers. Focusing on the experience of social time, this paper sheds light on users' experience on two popular gay mobile dating apps, namely Grindr and Jack'd. Based on in-depth interviews and focus-group discussions with 74 young gay men in Hong Kong, this paper identifies that the tempo and sequence produced by the specific affordances of apps are important to understanding users' experience. Specifically, accelerated tempo of interactions facilitated by constant connectivity, ubiquitous computing, geolocative function, and the apps' messaging system was seen to entail instantaneous and ephemeral relationships. The interface design, which foregrounds profile photos and backgrounds textual self-descriptions, structures the sequence of browsing and screening in a way that prioritizes physical appearance. This design feature was perceived to privilege users seeking casual hook-ups. These findings suggest that the temporality of browsing and exchange on apps is incongruous with the temporal norms prescribing formation of friendship and long-term romance. The violation of these normative expectations affects the perceived quality and satisfaction of app use, resulting in users' frustrations.
{"title":"Relationships form so quickly that you won't cherish them: Mobile Dating Apps and the Culture of Instantaneous Relationships","authors":"T. E. D. Yeo, Tsz Hin Fung","doi":"10.1145/2930971.2930973","DOIUrl":"https://doi.org/10.1145/2930971.2930973","url":null,"abstract":"Mobile dating apps with geolocative function have gained popularity for fostering social, romantic and sexual connections between proximate strangers. Focusing on the experience of social time, this paper sheds light on users' experience on two popular gay mobile dating apps, namely Grindr and Jack'd. Based on in-depth interviews and focus-group discussions with 74 young gay men in Hong Kong, this paper identifies that the tempo and sequence produced by the specific affordances of apps are important to understanding users' experience. Specifically, accelerated tempo of interactions facilitated by constant connectivity, ubiquitous computing, geolocative function, and the apps' messaging system was seen to entail instantaneous and ephemeral relationships. The interface design, which foregrounds profile photos and backgrounds textual self-descriptions, structures the sequence of browsing and screening in a way that prioritizes physical appearance. This design feature was perceived to privilege users seeking casual hook-ups. These findings suggest that the temporality of browsing and exchange on apps is incongruous with the temporal norms prescribing formation of friendship and long-term romance. The violation of these normative expectations affects the perceived quality and satisfaction of app use, resulting in users' frustrations.","PeriodicalId":227482,"journal":{"name":"Proceedings of the 7th 2016 International Conference on Social Media & Society","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126128583","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}
George Panteras, Xu Lu, A. Croitoru, A. Crooks, A. Stefanidis
Social media platforms have become extremely popular during the past few years, presenting an alternate, and often preferred, avenue for information dissemination within massive global communities. Such user-generated multimedia content is emerging as a critical source of information for a variety of applications, and particularly during times of crisis. In order to fully explore this potential, there is a need to better assess, and improve when possible, the accuracy of such information. This paper addresses this issue by focusing in particular on user-contributed image tagging in Flickr. We use as case study a natural disaster event (wildfire), and assess the reliability of user-generated tags. Furthermore, we compare these data to the results of a content-based annotation approach in order to assess the potential performance of an alternative, user-independent, automated approach to annotate such imagery. Our results show that Flickr user annotations can be considered quite reliable (at the level of ~50%), and that using a spatially distributed training dataset for our content-based image retrieval (CBIR) annotation process improves the performance of the content-based image labeling (to the level of ~75%).
{"title":"Accuracy Of User-Contributed Image Tagging In Flickr: A Natural Disaster Case Study","authors":"George Panteras, Xu Lu, A. Croitoru, A. Crooks, A. Stefanidis","doi":"10.1145/2930971.2930986","DOIUrl":"https://doi.org/10.1145/2930971.2930986","url":null,"abstract":"Social media platforms have become extremely popular during the past few years, presenting an alternate, and often preferred, avenue for information dissemination within massive global communities. Such user-generated multimedia content is emerging as a critical source of information for a variety of applications, and particularly during times of crisis. In order to fully explore this potential, there is a need to better assess, and improve when possible, the accuracy of such information. This paper addresses this issue by focusing in particular on user-contributed image tagging in Flickr. We use as case study a natural disaster event (wildfire), and assess the reliability of user-generated tags. Furthermore, we compare these data to the results of a content-based annotation approach in order to assess the potential performance of an alternative, user-independent, automated approach to annotate such imagery. Our results show that Flickr user annotations can be considered quite reliable (at the level of ~50%), and that using a spatially distributed training dataset for our content-based image retrieval (CBIR) annotation process improves the performance of the content-based image labeling (to the level of ~75%).","PeriodicalId":227482,"journal":{"name":"Proceedings of the 7th 2016 International Conference on Social Media & Society","volume":"240 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133137228","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：