Proceedings of the 30th Australasian Conference on Information Systems (ACIS), 2019
Information systems (IS) are widely used in organisations to improve business performance. The st... more Information systems (IS) are widely used in organisations to improve business performance. The steady progression in improving technologies like artificial intelligence (AI) and the need of securing future success of organisations lead to new requirements for IS. This research in progress firstly introduces the term AI-based services (AIBS) describing AI as a component enriching IS aiming at collaborating with employees and assisting in the execution of work-related tasks. The study derives requirements from ten expert interviews to successful design AIBS following Design Science Research (DSR). For a successful deployment of AIBS in organisations the D&M IS Success Model will be considered to validated requirements within three major dimensions of quality: Information Quality, System Quality, and Service Quality. Amongst others, preliminary findings propose that AIBS must be preferably authentic. Further discussion and research on AIBS is forced, thus, providing first insights on the deployment of AIBS in organisations.
Social media have become forums of discussions on political and societal debates in which individ... more Social media have become forums of discussions on political and societal debates in which individual users may forward information or influence others. While prior studies either employed network analyses or surveys to identify opinion leaders and their characteristics, the present investigation combines these two approaches to address the relationship between observable and self-perceived influence. For this purpose, a retweet network of Twitter communication on the Brexit debate (N = 15,018) was analyzed in relation to a survey on motives and personality traits that was filled out by a subsample of active users (N = 98). Results showed that users’ eigenvector centrality (as a measure of influence in the network) was significantly related to their political interest and their number of followers, but not to self-perceived opinion leadership. According to a comparison of self-assessment and network position, those with stronger motivations to distribute relevant information tended to overestimate their influence in the network. Implications for the identification of opinion leaders are discussed.
Im Kontext des IT-Service-Managements stellt ein wesentlicher Faktor zur möglichst effizienten un... more Im Kontext des IT-Service-Managements stellt ein wesentlicher Faktor zur möglichst effizienten und effektiven Entstörung gemeldeter Probleme das Incident Management dar. Aufgenommene Störungen werden dabei identifiziert, protokolliert und kategorisiert, so dass nachgelagerte Strukturen, wie Second- und Third-Level-Support, Beeinträchtigungen unmittelbar beheben können. In dem vorliegenden Beitrag wird ein Proof of Concept vorgestellt, der den Einsatz künstlicher Intelligenz auf das Incident Management in Unternehmen untersucht. Speziell wird hierbei dargelegt, wie ein entsprechendes System zum Einsatz kommt, um die Kategorisierung einer Störung in einem Praxisunternehmen zu beschleunigen. Als Datengrundlage hierfür dienen historische Tickets, die im Rahmen des Incident Managements erfasst worden sind. Ziel ist es, automatisiert die betroffene Applikation bzw. zuständige Gruppe zu ermitteln. Das Resultat dieser Machbarkeitsstudie ist ein neuronales Netzwerk, mit dem eine übergreifende Wahrscheinlichkeit von 94 % korrekt zugeordneter Kategorien erreicht wird. Auf Basis der vorliegenden Ergebnisse werden Herausforderungen während der Umsetzung dargestellt. Potentielle Einsatzgebiete und denkbare Weiterentwicklungen, speziell in Bezug auf die Kollaboration zwischen MitarbeiterInnen des Services Desks und künstlicher Intelligenz, werden abgeleitet und diskutiert. Zudem wird präsentiert, warum der Einsatz von künstlicher Intelligenz, sowohl im IT-Service-Management, als auch in anderen unternehmensrelevanten Vorgängen, sinnvoll erscheint.
Information systems such as social media strongly influencepublic opinion formation. Additionally... more Information systems such as social media strongly influencepublic opinion formation. Additionally, communication on the Internet is shaped by individuals and organisations with variousaims. This environment has given rise to phenomena such as manipulated content, fake news, and social bots. To examine the influence of manipulated opinions, we draw on the spiral of silence theory and complex adaptive systems. We translate empirical evidence of individual behaviour into an agent-based model and show that the model results in the emergence of a consensus on the collective level. In contrast to most previous approaches, this model explicitly represents interactions as a network. The most central actor in the network determines the final consensus 60–70% of the time. We then use the model to examine the influence of manipulative actors such as social bots on public opinion formation. The results indicate that, in a highly polarised setting, depending on their network position and the overall network density, bot participation by as little as 2–4% of a communication network can be sufficient to tip over the opinion climate in two out of three cases. These findings demonstrate a mechanism by which bots could shape the norms adopted by social media users.
International Conference on Information Systems Development (ISD), 2018
Learning and understanding what happened before, during, and after a crisis is extrem... more Learning and understanding what happened before, during, and after a crisis is extremely important for the improvement of the response process. For this purpose, social media has become an important communication medium used by both the affected persons and the emergency management services (EMSs). However, in different crises, different information may be needed, and the information shared in social media varies in its usefulness: It could be highly critical or completely irrelevant to the rescue operation. Supplying the best possible up-to-date information is crucial to the EMS, whose actions based on that information may save lives and resources. This paper studies a particular use case of extreme weather in Norway and identifies the information needs, the problem faced by EMSs, and how they use social media. It, further, pinpoints what different social media analysis platforms can provide in this type of crisis. The results of the research are criteria that social media analysis should follow to address EMSs' concerns. The output of this work can be used to more precisely describe social media communication for crises and to design more efficient platforms for information retrieval from social media.
International Conference on Information Systems (ICIS), 2018
Warning messages are being discussed as a possible mechanism to contain the circulation of false... more Warning messages are being discussed as a possible mechanism to contain the circulation of false information on social media. Their effectiveness for this purpose, however, is unclear. This article describes a survey experiment carried out to test two designs of warning messages: a simple one identical to the one used by Facebook, and a more complex one informed by recent research. We find no evidence that either design is clearly superior to not showing a warning message. This result has serious implications for brands and politicians, who might find false information about them spreading uncontrollably, as well as for managers of social media platforms, who are struggling to find effective means of controlling the diffusion of misinformation.
International Conference on Information Systems (ICIS), 2018
Like many other industries, the media industry has been profoundly affected by social... more Like many other industries, the media industry has been profoundly affected by social media. The public discuss the news on services such as Twitter, and some celebrity tweets are picked up by journalists and become news stories themselves. Information systems research into information diffusion has largely neglected these cross-platform diffusion patterns. To address this gap, the presented research in progress examines indicators of information spill-over between Twitter and ten German news websites, by detecting references in the form of URLs and mentions on both sides. Furthermore, the paper presents automatic and manual methods to identify two categories of spill-over, reference spill-over and content spill-over. Preliminary findings reveal differences between news outlets in how frequently they are referenced by Twitter users, and in how often they reference Twitter in their own stories.
Wikipedia is an important source of information in today’s world. Yet, the lack of gender diversi... more Wikipedia is an important source of information in today’s world. Yet, the lack of gender diversity in its community has been shown to affect the topics covered. Each Wikipedia article has a talk page that volunteer editors use to discuss proposed changes. Research on the gender bias has focused on article contribution and topic coverage, but not talk page activity. It has been suggested that the conflicts that take place in talk pages are especially intimidating for women, but this assertion has not been quantified yet. To fill this gap, we collected a dataset of all comments on Wikipedia talk pages, enriching it with gender information available from users who have chosen to disclose their gender on their user profiles or settings. Among the users active in talk pages, 49,387 indicated that they are male while only 5,996 indicated that they are female. The comments of these users make up for 4 million comments, approximately one quarter of all comments on Wikipedia. In addition, we observed that female participation varies by topic, reflecting traditional gender stereotypes: compared to Science, Technology, Engineering and Mathematics (STEM) topics, women were more active in categories such as Gender studies or Feminism. Results also indicate that a post on a talk page is 2.4% less likely to be replied to if the author is female. Likewise, reply probability varies from topic to topic. These results provide quantitative support for a gender bias in Wikipedia talk pages, and serve as a basis for discussing why overall female participation is low.
Proceedings of the Twenty-Sixth European Conference on Information Systems (ECIS), 2018
Recently, automated communication on social media has seen increased attention. Social bots, soci... more Recently, automated communication on social media has seen increased attention. Social bots, social media accounts controlled by algorithms that mimic human behaviour, have been found to attempt to influence users in several political contexts. However, their use in a commercial context, e.g. to boost sales of a product by aggressively promoting it with thousands of messages, has so far been neglected. To address this shortcoming, this paper examines the case of the social media music platform SoundCloud. We gathered a dataset of six months of activity, comprising 15,850,069 tracks and 12,125,095 comments. We then calculated a comment uniqueness score for highly active accounts to assess the variability of their comments. First analyses show that some accounts post suspiciously repetitive comments. These accounts also frequently repost existing content, but contribute little original content. An analysis of the commenting network further underlines that these accounts differ clearly from regular users. We conclude that the comment uniqueness metric can be used as an indicator to distinguish bots from humans, and that a considerable proportion of SoundCloud comments are likely to emanate from bots or semi-automated accounts. The implications of these findings and future plans are discussed.
In the context of events that involve public voting, such as televised competitions or elections,... more In the context of events that involve public voting, such as televised competitions or elections, it has increasingly been recognized that communication data from social media is related to the outcome. Existing studies mainly analyse the number of messages and their sentiment, yet the role of different data collection periods has not been examined sufficiently. We collected Twitter data in 2015 and 2016 to examine the relationship between the audience voting of the Eurovision Song Contest and predictors based on quantity and emotions, and compared the results of using data from before and during the event. We found that the choice of time period greatly affected the results obtained. Data collected prior to the event exhibited a much stronger association with the final ranking than data collected during the event. In addition, the model based on pre-event data in 2015 showed considerable accuracy in predicting the 2016 results, illustrating the usefulness of social media data for predicting the outcomes of events outside social media.
When Donald Trump won the Republican nomination and subsequently beat Hillary Clinton in the pres... more When Donald Trump won the Republican nomination and subsequently beat Hillary Clinton in the presidential elections, his success came as a surprise to most observers. This research contributes to understanding the dynamics of this unusual campaign, in which social media played a prominent role. We collected 6,099 tweets by both nominees during the presidential primaries and identified the 21 most frequently discussed issues through computer-assisted content analysis. Secondly, we used time series analysis to investigate whether the candidates influenced each other's political agendas. Most tweets by the candidates were found not to be about policy but about parties, other politicians, and the media. Of the political issues that were discussed, the most prominent ones were employment, family, minorities and terrorism. For tweets about minorities, we found possible evidence of agenda setting. We conclude that social media are mainly being used to reach out to supporters, instead of interacting with the opponent.
Proceedings of the 51st Hawaii International Conference on System Sciences, Jan 3, 2018
During a crisis, authorities need to effectively disseminate information. We address the problem ... more During a crisis, authorities need to effectively disseminate information. We address the problem of deciding how crisis-related information should be published on Facebook to reach as many people as possible. We examine three recent terrorist attacks in Berlin, London and Stockholm. Our specific focus lies with official Facebook pages by municipalities and emergency service agencies. We collected posts about the events, along with the number of shares, likes and emotional reactions to them. In a regression analysis, several variables were examined that capture decisions on which information to publish and how. Posts containing condolences were found to result in three times as many emotional reactions as other posts, all other variables held constant. Images and videos positively affected the number of reactions by factors of 2.2 and 3.9, respectively, while text length negatively affected the number of shares. These results will help in the development of effective guidelines.
International Journal of Information Management, 2018
Since an ever-increasing part of the population makes use of social media in their day-today live... more Since an ever-increasing part of the population makes use of social media in their day-today lives, social media data is being analysed in many different disciplines. The social media analytics process involves four distinct steps, data discovery, collection, preparation, and analysis. While there is a great deal of literature on the challenges and difficulties involving specific data analysis methods, there hardly exists research on the stages of data discovery, collection, and preparation. To address this gap, we conducted an extended and structured literature analysis through which we identified challenges addressed and solutions proposed. The literature search revealed that the volume of data was most often cited as a challenge by researchers. In contrast, other categories have received less attention. Based on the results of the literature search, we discuss the most important challenges for researchers and present potential solutions. The findings are used to extend an existing framework on social media analytics. The article provides benefits for researchers and practitioners who wish to collect and analyse social media data.
Australasian Conference on Information Systems, 2017
So-called 'social bots' have garnered a lot of attention lately. Previous research showed that th... more So-called 'social bots' have garnered a lot of attention lately. Previous research showed that they attempted to influence political events such as the Brexit referendum and the US presidential elections. It remains, however, somewhat unclear what exactly can be understood by the term 'social bot'. This paper addresses the need to better understand the intentions of bots on social media and to develop a shared understanding of how 'social' bots differ from other types of bots. We thus describe a systematic review of publications that researched bot accounts on social media. Based on the results of this literature review, we propose a scheme for categorising bot accounts on social media sites. Our scheme groups bot accounts by two dimensions – Imitation of human behaviour and Intent.
The last few years have seen more and more employees using their personal mobile devices for work... more The last few years have seen more and more employees using their personal mobile devices for work-related tasks. This phenomenon is part of a broader trend known as IT consumerization. Enterprises and employees have recognized that they might profit from these developments and implemented “Bring Your Own Device” (BYOD) policies, but they also have to face new challenges. This study investigates which types of employees adopt BYOD and how they benefit from it. To address these questions, the authors conducted a survey with 219 participants. Participants were classified into adopter types based on the Diffusion of Innovation Theory. The results indicate that early adopters and the early majority use their personal smartphones more often for work-related tasks than laggards, and that innovators and early adopters more often receive work-related email on their personal smartphones than other adopter types. It is concluded that DOI can successfully be applied to explain BYOD adoption behavior. Differentiated management strategies have to be applied in order to address the whole workforce.
Wikipedia offers researchers unique insights into the collaboration and communication patterns of... more Wikipedia offers researchers unique insights into the collaboration and communication patterns of a large self-regulating community of editors. The main medium of direct communication between editors of an article is the article's talk page. However, a talk page file is unstructured and therefore difficult to analyse automatically. A few parsers exist that enable its transformation into a structured data format. However, they are rarely open source, support only a limited subset of the talk page syntax - resulting in the loss of content - and usually support only one export format. Together with this article we offer a very fast, lightweight, open source parser with support for various output formats. In a preliminary evaluation it achieved a high accuracy. The parser uses a grammar-based approach - offering a transparent implementation and easy extensibility.
Proceedings of NLP4CMC III: 3rd Workshop on Natural Language Processing for Computer-Mediated Communication, 2016
Some users of social media are spreading racist, sexist, and otherwise hateful content. For the p... more Some users of social media are spreading racist, sexist, and otherwise hateful content. For the purpose of training a hate speech detection system, the reliability of the annotations is crucial, but there is no universally agreed-upon definition. We collected potentially hateful messages and asked two groups of internet users to determine whether they were hate speech or not, whether they should be banned or not and to rate their degree of offensiveness. One of the groups was shown a definition prior to completing the survey. We aimed to assess whether hate speech can be annotated reliably, and the extent to which existing definitions are in accordance with subjective ratings. Our results indicate that showing users a definition caused them to partially align their own opinion with the definition but did not improve reliability, which was very low overall. We conclude that the presence of hate speech should perhaps not be considered a binary yes-or-no decision, and raters need more detailed instructions for the annotation.
Proceedings of the 30th Australasian Conference on Information Systems (ACIS), 2019
Information systems (IS) are widely used in organisations to improve business performance. The st... more Information systems (IS) are widely used in organisations to improve business performance. The steady progression in improving technologies like artificial intelligence (AI) and the need of securing future success of organisations lead to new requirements for IS. This research in progress firstly introduces the term AI-based services (AIBS) describing AI as a component enriching IS aiming at collaborating with employees and assisting in the execution of work-related tasks. The study derives requirements from ten expert interviews to successful design AIBS following Design Science Research (DSR). For a successful deployment of AIBS in organisations the D&M IS Success Model will be considered to validated requirements within three major dimensions of quality: Information Quality, System Quality, and Service Quality. Amongst others, preliminary findings propose that AIBS must be preferably authentic. Further discussion and research on AIBS is forced, thus, providing first insights on the deployment of AIBS in organisations.
Social media have become forums of discussions on political and societal debates in which individ... more Social media have become forums of discussions on political and societal debates in which individual users may forward information or influence others. While prior studies either employed network analyses or surveys to identify opinion leaders and their characteristics, the present investigation combines these two approaches to address the relationship between observable and self-perceived influence. For this purpose, a retweet network of Twitter communication on the Brexit debate (N = 15,018) was analyzed in relation to a survey on motives and personality traits that was filled out by a subsample of active users (N = 98). Results showed that users’ eigenvector centrality (as a measure of influence in the network) was significantly related to their political interest and their number of followers, but not to self-perceived opinion leadership. According to a comparison of self-assessment and network position, those with stronger motivations to distribute relevant information tended to overestimate their influence in the network. Implications for the identification of opinion leaders are discussed.
Im Kontext des IT-Service-Managements stellt ein wesentlicher Faktor zur möglichst effizienten un... more Im Kontext des IT-Service-Managements stellt ein wesentlicher Faktor zur möglichst effizienten und effektiven Entstörung gemeldeter Probleme das Incident Management dar. Aufgenommene Störungen werden dabei identifiziert, protokolliert und kategorisiert, so dass nachgelagerte Strukturen, wie Second- und Third-Level-Support, Beeinträchtigungen unmittelbar beheben können. In dem vorliegenden Beitrag wird ein Proof of Concept vorgestellt, der den Einsatz künstlicher Intelligenz auf das Incident Management in Unternehmen untersucht. Speziell wird hierbei dargelegt, wie ein entsprechendes System zum Einsatz kommt, um die Kategorisierung einer Störung in einem Praxisunternehmen zu beschleunigen. Als Datengrundlage hierfür dienen historische Tickets, die im Rahmen des Incident Managements erfasst worden sind. Ziel ist es, automatisiert die betroffene Applikation bzw. zuständige Gruppe zu ermitteln. Das Resultat dieser Machbarkeitsstudie ist ein neuronales Netzwerk, mit dem eine übergreifende Wahrscheinlichkeit von 94 % korrekt zugeordneter Kategorien erreicht wird. Auf Basis der vorliegenden Ergebnisse werden Herausforderungen während der Umsetzung dargestellt. Potentielle Einsatzgebiete und denkbare Weiterentwicklungen, speziell in Bezug auf die Kollaboration zwischen MitarbeiterInnen des Services Desks und künstlicher Intelligenz, werden abgeleitet und diskutiert. Zudem wird präsentiert, warum der Einsatz von künstlicher Intelligenz, sowohl im IT-Service-Management, als auch in anderen unternehmensrelevanten Vorgängen, sinnvoll erscheint.
Information systems such as social media strongly influencepublic opinion formation. Additionally... more Information systems such as social media strongly influencepublic opinion formation. Additionally, communication on the Internet is shaped by individuals and organisations with variousaims. This environment has given rise to phenomena such as manipulated content, fake news, and social bots. To examine the influence of manipulated opinions, we draw on the spiral of silence theory and complex adaptive systems. We translate empirical evidence of individual behaviour into an agent-based model and show that the model results in the emergence of a consensus on the collective level. In contrast to most previous approaches, this model explicitly represents interactions as a network. The most central actor in the network determines the final consensus 60–70% of the time. We then use the model to examine the influence of manipulative actors such as social bots on public opinion formation. The results indicate that, in a highly polarised setting, depending on their network position and the overall network density, bot participation by as little as 2–4% of a communication network can be sufficient to tip over the opinion climate in two out of three cases. These findings demonstrate a mechanism by which bots could shape the norms adopted by social media users.
International Conference on Information Systems Development (ISD), 2018
Learning and understanding what happened before, during, and after a crisis is extrem... more Learning and understanding what happened before, during, and after a crisis is extremely important for the improvement of the response process. For this purpose, social media has become an important communication medium used by both the affected persons and the emergency management services (EMSs). However, in different crises, different information may be needed, and the information shared in social media varies in its usefulness: It could be highly critical or completely irrelevant to the rescue operation. Supplying the best possible up-to-date information is crucial to the EMS, whose actions based on that information may save lives and resources. This paper studies a particular use case of extreme weather in Norway and identifies the information needs, the problem faced by EMSs, and how they use social media. It, further, pinpoints what different social media analysis platforms can provide in this type of crisis. The results of the research are criteria that social media analysis should follow to address EMSs' concerns. The output of this work can be used to more precisely describe social media communication for crises and to design more efficient platforms for information retrieval from social media.
International Conference on Information Systems (ICIS), 2018
Warning messages are being discussed as a possible mechanism to contain the circulation of false... more Warning messages are being discussed as a possible mechanism to contain the circulation of false information on social media. Their effectiveness for this purpose, however, is unclear. This article describes a survey experiment carried out to test two designs of warning messages: a simple one identical to the one used by Facebook, and a more complex one informed by recent research. We find no evidence that either design is clearly superior to not showing a warning message. This result has serious implications for brands and politicians, who might find false information about them spreading uncontrollably, as well as for managers of social media platforms, who are struggling to find effective means of controlling the diffusion of misinformation.
International Conference on Information Systems (ICIS), 2018
Like many other industries, the media industry has been profoundly affected by social... more Like many other industries, the media industry has been profoundly affected by social media. The public discuss the news on services such as Twitter, and some celebrity tweets are picked up by journalists and become news stories themselves. Information systems research into information diffusion has largely neglected these cross-platform diffusion patterns. To address this gap, the presented research in progress examines indicators of information spill-over between Twitter and ten German news websites, by detecting references in the form of URLs and mentions on both sides. Furthermore, the paper presents automatic and manual methods to identify two categories of spill-over, reference spill-over and content spill-over. Preliminary findings reveal differences between news outlets in how frequently they are referenced by Twitter users, and in how often they reference Twitter in their own stories.
Wikipedia is an important source of information in today’s world. Yet, the lack of gender diversi... more Wikipedia is an important source of information in today’s world. Yet, the lack of gender diversity in its community has been shown to affect the topics covered. Each Wikipedia article has a talk page that volunteer editors use to discuss proposed changes. Research on the gender bias has focused on article contribution and topic coverage, but not talk page activity. It has been suggested that the conflicts that take place in talk pages are especially intimidating for women, but this assertion has not been quantified yet. To fill this gap, we collected a dataset of all comments on Wikipedia talk pages, enriching it with gender information available from users who have chosen to disclose their gender on their user profiles or settings. Among the users active in talk pages, 49,387 indicated that they are male while only 5,996 indicated that they are female. The comments of these users make up for 4 million comments, approximately one quarter of all comments on Wikipedia. In addition, we observed that female participation varies by topic, reflecting traditional gender stereotypes: compared to Science, Technology, Engineering and Mathematics (STEM) topics, women were more active in categories such as Gender studies or Feminism. Results also indicate that a post on a talk page is 2.4% less likely to be replied to if the author is female. Likewise, reply probability varies from topic to topic. These results provide quantitative support for a gender bias in Wikipedia talk pages, and serve as a basis for discussing why overall female participation is low.
Proceedings of the Twenty-Sixth European Conference on Information Systems (ECIS), 2018
Recently, automated communication on social media has seen increased attention. Social bots, soci... more Recently, automated communication on social media has seen increased attention. Social bots, social media accounts controlled by algorithms that mimic human behaviour, have been found to attempt to influence users in several political contexts. However, their use in a commercial context, e.g. to boost sales of a product by aggressively promoting it with thousands of messages, has so far been neglected. To address this shortcoming, this paper examines the case of the social media music platform SoundCloud. We gathered a dataset of six months of activity, comprising 15,850,069 tracks and 12,125,095 comments. We then calculated a comment uniqueness score for highly active accounts to assess the variability of their comments. First analyses show that some accounts post suspiciously repetitive comments. These accounts also frequently repost existing content, but contribute little original content. An analysis of the commenting network further underlines that these accounts differ clearly from regular users. We conclude that the comment uniqueness metric can be used as an indicator to distinguish bots from humans, and that a considerable proportion of SoundCloud comments are likely to emanate from bots or semi-automated accounts. The implications of these findings and future plans are discussed.
In the context of events that involve public voting, such as televised competitions or elections,... more In the context of events that involve public voting, such as televised competitions or elections, it has increasingly been recognized that communication data from social media is related to the outcome. Existing studies mainly analyse the number of messages and their sentiment, yet the role of different data collection periods has not been examined sufficiently. We collected Twitter data in 2015 and 2016 to examine the relationship between the audience voting of the Eurovision Song Contest and predictors based on quantity and emotions, and compared the results of using data from before and during the event. We found that the choice of time period greatly affected the results obtained. Data collected prior to the event exhibited a much stronger association with the final ranking than data collected during the event. In addition, the model based on pre-event data in 2015 showed considerable accuracy in predicting the 2016 results, illustrating the usefulness of social media data for predicting the outcomes of events outside social media.
When Donald Trump won the Republican nomination and subsequently beat Hillary Clinton in the pres... more When Donald Trump won the Republican nomination and subsequently beat Hillary Clinton in the presidential elections, his success came as a surprise to most observers. This research contributes to understanding the dynamics of this unusual campaign, in which social media played a prominent role. We collected 6,099 tweets by both nominees during the presidential primaries and identified the 21 most frequently discussed issues through computer-assisted content analysis. Secondly, we used time series analysis to investigate whether the candidates influenced each other's political agendas. Most tweets by the candidates were found not to be about policy but about parties, other politicians, and the media. Of the political issues that were discussed, the most prominent ones were employment, family, minorities and terrorism. For tweets about minorities, we found possible evidence of agenda setting. We conclude that social media are mainly being used to reach out to supporters, instead of interacting with the opponent.
Proceedings of the 51st Hawaii International Conference on System Sciences, Jan 3, 2018
During a crisis, authorities need to effectively disseminate information. We address the problem ... more During a crisis, authorities need to effectively disseminate information. We address the problem of deciding how crisis-related information should be published on Facebook to reach as many people as possible. We examine three recent terrorist attacks in Berlin, London and Stockholm. Our specific focus lies with official Facebook pages by municipalities and emergency service agencies. We collected posts about the events, along with the number of shares, likes and emotional reactions to them. In a regression analysis, several variables were examined that capture decisions on which information to publish and how. Posts containing condolences were found to result in three times as many emotional reactions as other posts, all other variables held constant. Images and videos positively affected the number of reactions by factors of 2.2 and 3.9, respectively, while text length negatively affected the number of shares. These results will help in the development of effective guidelines.
International Journal of Information Management, 2018
Since an ever-increasing part of the population makes use of social media in their day-today live... more Since an ever-increasing part of the population makes use of social media in their day-today lives, social media data is being analysed in many different disciplines. The social media analytics process involves four distinct steps, data discovery, collection, preparation, and analysis. While there is a great deal of literature on the challenges and difficulties involving specific data analysis methods, there hardly exists research on the stages of data discovery, collection, and preparation. To address this gap, we conducted an extended and structured literature analysis through which we identified challenges addressed and solutions proposed. The literature search revealed that the volume of data was most often cited as a challenge by researchers. In contrast, other categories have received less attention. Based on the results of the literature search, we discuss the most important challenges for researchers and present potential solutions. The findings are used to extend an existing framework on social media analytics. The article provides benefits for researchers and practitioners who wish to collect and analyse social media data.
Australasian Conference on Information Systems, 2017
So-called 'social bots' have garnered a lot of attention lately. Previous research showed that th... more So-called 'social bots' have garnered a lot of attention lately. Previous research showed that they attempted to influence political events such as the Brexit referendum and the US presidential elections. It remains, however, somewhat unclear what exactly can be understood by the term 'social bot'. This paper addresses the need to better understand the intentions of bots on social media and to develop a shared understanding of how 'social' bots differ from other types of bots. We thus describe a systematic review of publications that researched bot accounts on social media. Based on the results of this literature review, we propose a scheme for categorising bot accounts on social media sites. Our scheme groups bot accounts by two dimensions – Imitation of human behaviour and Intent.
The last few years have seen more and more employees using their personal mobile devices for work... more The last few years have seen more and more employees using their personal mobile devices for work-related tasks. This phenomenon is part of a broader trend known as IT consumerization. Enterprises and employees have recognized that they might profit from these developments and implemented “Bring Your Own Device” (BYOD) policies, but they also have to face new challenges. This study investigates which types of employees adopt BYOD and how they benefit from it. To address these questions, the authors conducted a survey with 219 participants. Participants were classified into adopter types based on the Diffusion of Innovation Theory. The results indicate that early adopters and the early majority use their personal smartphones more often for work-related tasks than laggards, and that innovators and early adopters more often receive work-related email on their personal smartphones than other adopter types. It is concluded that DOI can successfully be applied to explain BYOD adoption behavior. Differentiated management strategies have to be applied in order to address the whole workforce.
Wikipedia offers researchers unique insights into the collaboration and communication patterns of... more Wikipedia offers researchers unique insights into the collaboration and communication patterns of a large self-regulating community of editors. The main medium of direct communication between editors of an article is the article's talk page. However, a talk page file is unstructured and therefore difficult to analyse automatically. A few parsers exist that enable its transformation into a structured data format. However, they are rarely open source, support only a limited subset of the talk page syntax - resulting in the loss of content - and usually support only one export format. Together with this article we offer a very fast, lightweight, open source parser with support for various output formats. In a preliminary evaluation it achieved a high accuracy. The parser uses a grammar-based approach - offering a transparent implementation and easy extensibility.
Proceedings of NLP4CMC III: 3rd Workshop on Natural Language Processing for Computer-Mediated Communication, 2016
Some users of social media are spreading racist, sexist, and otherwise hateful content. For the p... more Some users of social media are spreading racist, sexist, and otherwise hateful content. For the purpose of training a hate speech detection system, the reliability of the annotations is crucial, but there is no universally agreed-upon definition. We collected potentially hateful messages and asked two groups of internet users to determine whether they were hate speech or not, whether they should be banned or not and to rate their degree of offensiveness. One of the groups was shown a definition prior to completing the survey. We aimed to assess whether hate speech can be annotated reliably, and the extent to which existing definitions are in accordance with subjective ratings. Our results indicate that showing users a definition caused them to partially align their own opinion with the definition but did not improve reliability, which was very low overall. We conclude that the presence of hate speech should perhaps not be considered a binary yes-or-no decision, and raters need more detailed instructions for the annotation.
Uploads
Papers by Björn Ross