Twitter - Professional Tool

Abstract:

Anticipating repliers in online conversations is a fundamental challenge for computer mediated communication systems which aim to make textual, audio and/or video communication as natural as face to face communication. The massive amounts of data that social media generates has facilitated the study of online conversations on a scale unimaginable a few years ago. In this work we use data from Twitter to explore the predictability of repliers, and investigate the factors which influence who will reply to a message. Our results suggest that social factors, which describe the strength of relations between users, are more useful than topical factors. This indicates that Twitter users’ reply behavior is more impacted by social relations than by topics. Finally, we show that a binary classification model, which differentiates between users who will and users who will not reply to a certain message, may achieve an F1-score of 0:74 when using social features.

http://claudiawagner.info/publications/websci2013_conversationAnalysis.pdf

Abstract:

Traditional media outlets are known to report political news in a biased way, and it matters because it could affect the political beliefs of the audience and can ultimately alter voting behavior. Hence tracking bias in everyday news and building a platform where people can receive balanced news information is important. In this work, we proposed a model that maps the news media outlets along an one dimensional dichotomous political spectrum using the co-subscriptions relationships
inferred by Twitter links. By analyzing 7 million Twitter links, we have shown that the political dichotomy naturally arises on Twitter when we only consider direct media
subscription. Furthermore, we demonstrated a real-time Twitter-based application that visualizes an ideological map of various media sources.

http://www.arcomem.eu/wp-content/uploads/2012/07/Twitter-follow-links-reveal-bicameral-landscape-of-newspapers.pdf

By Dan Pontefract

"In terms of increasing the engagement level and culture of an organisation, micro-blogging can provide several benefits, including:

- Greater understanding of what is actually going on in the organisation across teams or projects;
- Personalizing and demystifying the aura of senior leaders;

- Seeking opinion and exploring options before decisions are made;
- Driving engagement and the feeling that everyone’s opinion is important;
- Increasing social status of employees in business related matters;
- Providing information that is timely through non-formal use;
- Building trust amongst senior leaders and front-line workers in open, transparent ways;

- Weak chance of disrupting already established workflows"

"Abstract:

A particular challenge in the area of social media analysis is how to nd communities within a larger network of social interactions.
Here a community may be a group of microblogging users who post content on a coherent topic, or who are associated with a speci c event or news story. Twitter provides the ability to curate users into lists, corresponding to meaningful topics or themes. Here we describe an approach for crowdsourcing the list building e orts of many di erent Twitter users, in order to identify topical communities. This approach involves the use of ensemble community nding to produce stable groupings of user lists, and by extension, individual Twitter users. We examine this approach in the context of a case study surrounding the detection of communities on Twitter relating to the London 2012 Olympics."

http://arxiv.org/pdf/1207.0017.pdf

A study made by Theresa B. Clarke and C. Leigh Nelson

Purpose of the Study.

This study explores various outcomes associated with the incorporation of Twitter in the marketing classroom. To determine if Twitter use is a beneficial pursuit for marketing educators, we investigated classroom community, pedagogical effectiveness, and learning outcomes based on Twitter use and non-use
within a required marketing course.

Method/Design and Sample.

For comparative purposes, a quasi-experimental design was employed across two semesters of the same undergraduate integrated marketing communications course. One semester (48 students) employed heavy Twitter use by both students and the instructor; the other semester had no Twitter use (36 students).

Results.

Independent sample t-tests (p < .05) were conducted to test the hypotheses. The course using Twitter had a significantly higher sense of classroom community and perception of pedagogical effectiveness. While there was no difference in perceived learning across the two groups, the group using Twitter outperformed the
non-Twitter group on actual learning.

Value to Marketing Educators.

This study extends the small, but growing, body of knowledge on the use of Twitter in the higher education marketing classroom. Findings and recommendations add value to marketing educators by helping them make more informed decisions regarding whether or not to use Twitter in their courses.

http://www.mmaglobal.org/JAME-Archive/JAME_Vol20_3/Twitter%20Use%20in%20Undergraduate%20Marketing%20Courses.pdf

Abstract:

Twitter introduced user lists in late 2009, allowing users to be grouped according to meaningful topics or themes. Lists have since been adopted by media outlets as a means of organising content around news stories. Thus the curation of these lists is important - they should contain the key information gatekeepers and present a balanced perspective on a story. Here we address this list curation process from a
recommender systems perspective. We propose a variety of criteria for generating user list recommendations, based on content analysis, network analysis, and the "crowdsourcing" of existing user lists. We demonstrate that these types of criteria are often only successful for datasets with certain characteristics. To resolve this issue, we propose the aggregation of these dif ferent "views" of a news story on Twitter to produce more accurate user recommendations to support the curation process.

http://arxiv.org/pdf/1206.1728.pdf