An empirical evaluation of text representation schemes to filter the social media stream

Author(s):  
Sandip Modha ◽  
Prasenjit Majumder ◽  
Thomas Mandl
2019 ◽  
Vol 35 (2) ◽  
pp. 284-305 ◽  
Author(s):  
Riyad Eid ◽  
Ziad Abdelmoety ◽  
Gomaa Agag

Purpose The social media have enabled companies to reach out to global markets and provided them with the opportunity to customize their strategies and offerings in an unprecedented way. Given the scant empirical evaluation of social media use in the small- and medium-sized enterprises (SMEs) business-to-business (B-to-B) context, this paper aims to offer a comprehensive description of the antecedents and consequences of social media use in international B-to-B SMEs and the way in which this use affects their export performance. Design/methodology/approach This study uses a sample of 277 British B-to-B SMEs and uses positivist research with a quantitative approach, adopting a survey strategy through questionnaires and structural equation modeling. Findings The results reveal that the use of social media influences export performance through the quality of international business contacts – understanding customers’ views and preferences, brand awareness and knowledge of the competition in various international markets. This study contributes to the emerging literature on B-to-B SMEs digital marketing by determining the mechanism through which B-to-B SMEs may benefit from using the social media in their efforts to export. Originality/value Despite the promising potential of the social media, especially for export-oriented companies, very limited attention has hitherto been paid to the relationship between the use of social media and export performance. This study attempts to fill the gap by investigating the extent to which actual use of social media impacts on the performance of exporting firms.


2021 ◽  
Vol 11 (9) ◽  
pp. 3872
Author(s):  
Jose Moreno Ortega ◽  
Juan Bernabé-Moreno

The massive impact caused by the COVID-19 pandemic has left no one indifferent, becoming an unprecedented challenge. The use of protections such as sanitary masks has become increasingly common, restrictions in our daily lives, such as social distancing or confinements, have had serious consequences on the economy and our welfare state. Although the measures imposed throughout the world follow the same pattern, they have been applied with different criteria depending on the country. Over extended periods of time, people tend to change their perception of an event and its magnitude, or in other words, they stop being so concerned despite the seriousness of the matter. In this paper, we introduce a new metric to quantify the degree of emotional concern of people being affected by a topic, and we confirm how populations from different countries follow this trend of downplaying the effect of the pandemic and reach a state of indifference. To do this, we propose a method to analyze the social media stream over time extracting the different emotional states from the Russel Circumplex plane and computing the shifting created by the tragic event—the pandemic. We complete this metric by incorporating searching behavior to reflect not only push contents but also pull inquiries. The resulting metric establishes a relationship between the pandemic and the emotional response by defining the degree of Emotional Concern. Although the method can be applied to any location with a significant and varied amount of geo-localized social media streams, the scope of this paper covers the most representative cities in Europe.


2021 ◽  
Vol 11 (12) ◽  
pp. 5489
Author(s):  
Ibrahim Riza Hallac ◽  
Betul Ay ◽  
Galip Aydin

Gathering useful insights from social media data has gained great interest over the recent years. User representation can be a key task in mining publicly available user-generated rich content offered by the social media platforms. The way to automatically create meaningful observations about users of a social network is to obtain real-valued vectors for the users with user embedding representation learning models. In this study, we presented one of the most comprehensive studies in the literature in terms of learning high-quality social media user representations by leveraging state-of-the-art text representation approaches. We proposed a novel doc2vec-based representation method, which can encode both textual and non-textual information of a social media user into a low dimensional vector. In addition, various experiments were performed for investigating the performance of text representation techniques and concepts including word2vec, doc2vec, Glove, NumberBatch, FastText, BERT, ELMO, and TF-IDF. We also shared a new social media dataset comprising data from 500 manually selected Twitter users of five predefined groups. The dataset contains different activity data such as comment, retweet, like, location, as well as the actual tweets composed by the users.


Author(s):  
Kathleen Hall Jamieson

Cyberwar examines the ways in which Russian interventions not only affected the behaviors of key players but altered the 2016 presidential campaign’s media and social media landscape. After laying out a theory of influence that explains how Russian activities could have produced effects, Jamieson documents the hackers and trolls’ influence on the topics in the news, the questions in the presidential debates, and the social media stream. Drawing on her analysis of messages crafted and amplified by Russian operatives, changes that Russian-hacked content elicited in news and the debates, the scholarly work of other researchers, and Annenberg surveys, she concludes that it is plausible to believe that Russian machinations helped elect Donald J. Trump the 45th president of the United States.


2017 ◽  
Vol 16 (1) ◽  
pp. 12-24 ◽  
Author(s):  
Nicole Behringer ◽  
Kai Sassenberg ◽  
Annika Scholl

Abstract. Knowledge exchange via social media is crucial for organizational success. Yet, many employees only read others’ contributions without actively contributing their knowledge. We thus examined predictors of the willingness to contribute knowledge. Applying social identity theory and expectancy theory to knowledge exchange, we investigated the interplay of users’ identification with their organization and perceived usefulness of a social media tool. In two studies, identification facilitated users’ willingness to contribute knowledge – provided that the social media tool seemed useful (vs. not-useful). Interestingly, identification also raised the importance of acquiring knowledge collectively, which could in turn compensate for low usefulness of the tool. Hence, considering both social and media factors is crucial to enhance employees’ willingness to share knowledge via social media.


Author(s):  
Tomas Brusell

When modern technology permeates every corner of life, there are ignited more and more hopes among the disabled to be compensated for the loss of mobility and participation in normal life, and with Information and Communication Technologies (ICT), Exoskeleton Technologies and truly hands free technologies (HMI), it's possible for the disabled to be included in the social and pedagogic spheres, especially via computers and smartphones with social media apps and digital instruments for Augmented Reality (AR) .In this paper a nouvel HMI technology is presented with relevance for the inclusion of disabled in every day life with specific focus on the future development of "smart cities" and "smart homes".


Sign in / Sign up

Export Citation Format

Share Document