Análisis de contenidos en social media: clasificación de mensajes e identificación de influyentes en el Banco Central Europeo (BCE)

Este proyecto pretende avanzar en el estado del arte de la investigación acerca de los Social Media en las áreas involucradas en aspectos como la detección de regularidades en entornos de los Social Media, la creación de modelos de crecimiento, evolución y propagación o, más en general, el descubrimiento de fenómenos interesantes en torno a la dinámica de los Social Media. Estas áreas abarcarán: La obtención de información mediante la minería del conjunto de datos procedentes de los Social Media, compuestos por contenidos, gente [datos de uso] e interacciones entre los mismos [redes sociales]. La construcción de conocimiento (modelos, métricas, etc.) a partir de la información obtenida mediante minería. El avance en la capacidad de acción en los Social Media mediante la simulación y experimentación con el conocimiento obtenido.

A review on political analysis and social media

En los países democráticos, conocer la intención de voto de los ciudadanos y las valoraciones de los principales partidos y líderes políticos es de gran interés tanto para los propios partidos como para los medios de comunicación y el público en general. Para ello se han utilizado tradicionalmente costosas encuestas personales. El auge de las redes sociales, principalmente Twitter, permite pensar en ellas como una alternativa barata a las encuestas. En este trabajo, revisamos la bibliografía científica más relevante en este ámbito, poniendo especial énfasis en el caso español.

From the initial interest in ‘Why do people often communicate in very "strange" ways on Social Media?’ to a research project on CMC

This presentation is about the inside story of the PhD project El malagueño real, mental y virtual. Configuración de los significados sociales de una variedad urbana in Hispanic Linguistics. That is, the production and perception of the Spanish spoken in the city of Malaga and used on the social network sites Facebook and Tuenti by users from Malaga is analysed. Actually, the southern Spanish variety in question is quite distinct from the national standard in terms of its phonetic features, its prestige, and the attitudes to it. Thus, the project started with the initial interest in «Why do people often communicate in very “strange” ways on social media» which then slightly changed to the final research interest in «What do the different non-standard variants mean in virtual (and real) malagueño?». This long – sometimes hazardous, yet mostly fun – process is exposed in more detail by looking at the research questions, the methods and results. Lastly, the presentation concludes with some lessons learnt and an outlook on possibilities and necessities for further investigation.

"That's what social media is all about - engagement" : Hur märkesinnehavare använder visuell storytelling på Instagram för att skapa märkesgemenskaper

Utvecklingen av sociala medier såsom t.ex. Facebook och Instagram har inneburit nya möjligheter att kommunicera sin livsstil och sina personliga val, som överskrider såväl kulturella som geografiska gränser. Nya gemenskaper skapas. Detta får betydelse för hur konsumenter kommunicerar med varandra samt hur märkesinnehavarna kommunicerar med sina kunder. Syftet med min studie är att undersöka hur en märkesinnehavare inom mode använder sociala medier, i detta fall Instagram, i sin kommunikation med kunderna för att skapa engagemang kring sitt varumärke. Aktuell forskning visar att värde i varumärket skapas genom en ömsesidig kommunikation mellan märkesinnehavaren och konsumenterna. Kommunikationen sker både mellan märkesinnehavare och konsumenter samt mellan konsumenterna i en märkesgemenskap. Storytelling är ett verksamt medel i denna meningsskapande process. En semiotisk analys användes för att analysera bilderna och hur de bidrar till märkets storytelling. Resultatet visar att Instagram ger märkesinnehavare och konsumenter möjlighet att gemensamt skapa värde. De tekniska funktionerna i applikationen gör det lätt för märkesinnehavare att interagera med sina konsumenter och skapa en dynamisk märkesgemenskap. På Sabo Skirts instagramkonto är märkesinnehavaren och följarna tillsammans märkets författare. Följarna använder kläderna och ger dem betydelse i sin egen vardag samtidigt som märkesinnehavaren skapar möjligheter för dem att synas på ett instagramkonto med 1,5 miljoner följare. Detta kan tolkas med hjälp av teorin uses and gratifications som säger att användningen av media alltid är målinriktad. Människor väljer aktivt media efter vilka deras behov är och hur väl användningen tillfredsställer dessa behov.

Digital Activism: How Social Media Prevalence has Impacted Modern Activism

Thesis (Master's)--University of Washington, 2016-06

BEHIND THE COMPUTER SCREEN: WHAT IRB PROFESSIONALS REALLY THINK ABOUT SOCIAL MEDIA RESEARCH

Thesis (Master's)--University of Washington, 2016-06

Corporate Social Media: How Two-Way Disclosure Channels Influence Investors

Thesis (Ph.D.)--University of Washington, 2016-06

social me: social media literacy for middle school classrooms

Thesis (Master's)--University of Washington, 2016-06

Autochthonous heritage languages and social media:writing and bilingual practices in Low German on Facebook

This article analyses how speakers of an autochthonous heritage language (AHL) make use of digital media, through the example of Low German, a regional language used by a decreasing number of speakers mainly in northern Germany. The focus of the analysis is on Web 2.0 and its interactive potential for individual speakers. The study therefore examines linguistic practices on the social network site Facebook, with special emphasis on language choice, bilingual practices and writing in the autochthonous heritage language. The findings suggest that social network sites such as Facebook have the potential to provide new mediatized spaces for speakers of an AHL that can instigate sociolinguistic change.

Revolution in sales:The impact of social media and related technology on the selling environment

Over the years several articles have tracked the impact of technology on various aspects of the sales domain. However, the advent of social media and technologies related to social media has gone largely unnoticed in the literature. This article first provides brief attention to changing aspects of technology within the sales environment, leading to the identification of social media as a dominant new selling tool. A qualitative approach (focus groups) is employed to explore the breadth of current technology usage by sales managers and salespeople. Analysis of the data, collected in the United States and the United Kingdom, reveals six major themes: connectivity, relationships, selling tools, generational, global, and sales/marketing interface. Results provide evidence of a revolution in the buyer-seller relationship that includes some unanticipated consequences both for sales organization performance and needed future research contributions.

Use of social media to enable medical teams to work virtually (and better)

Refers to: Sophie Scott The researcher of the future…makes the most of social media The Lancet, Volume 381, Supplement 1, 27 February 2013, Pages S5-S6

A weakly-supervised Bayesian model for violence detection from social media

Social streams have proven to be the mostup-to-date and inclusive information on cur-rent events. In this paper we propose a novelprobabilistic modelling framework, called violence detection model (VDM), which enables the identification of text containing violent content and extraction of violence-related topics over social media data. The proposed VDM model does not require any labeled corpora for training, instead, it only needs the in-corporation of word prior knowledge which captures whether a word indicates violence or not. We propose a novel approach of deriving word prior knowledge using the relative entropy measurement of words based on the in-tuition that low entropy words are indicative of semantically coherent topics and therefore more informative, while high entropy words indicates words whose usage is more topical diverse and therefore less informative. Our proposed VDM model has been evaluated on the TREC Microblog 2011 dataset to identify topics related to violence. Experimental results show that deriving word priors using our proposed relative entropy method is more effective than the widely-used information gain method. Moreover, VDM gives higher violence classification results and produces more coherent violence-related topics compared toa few competitive baselines.

Risk communication and social media:are we ready?

Oral presentation