REVEAL aims to advance the necessary technologies for making a higher level analysis of social media possible. The project will enable users to reveal hidden ‘modalities’ such as reputation, influence or credibility of information.
Meet some of our project partners from the kick-off meeting who tell us more about their expectations of REVEAL.
The world of media and communication is currently experiencing enormous disruptions: from one-way communication and word of mouth exchanges, we have moved to bi- or multi directional communication patterns. No longer can selected few (e.g. media organizations) act as gatekeepers, deciding what is communicated to whom and what not. Individuals now have the opportunity to access information directly from primary sources, through a channel we label e’-word of mouth’, or what we commonly call ‘Social Media’. However, a key problem: It takes a lot of effort to distinguish useful information from the ‘noise’ (e.g. useless or misleading information). This challenge has become the focus of various research efforts. REVEAL aims to discover higher level concepts hidden within information. This is what we aim for: We will reveal much more than bare content. Further to discovering what is being said, we will determine how trustworthy that information is. We will predict contributor impact and how much or to what extent all this affects reputation or influence. The core of our work is to reveal hidden modalities for the benefit of a better understanding and utilization of the Social Media world.
JOURNALISM SCENARIO – Event Coverage
In Social Media we have hidden communities with members vividly interacting with content and with each other. We need to be able to discover and analyse the structure of these communities through automatic inference of community-type relations between members by continuous analysis of their social interactions. For this we will need to employ novel techniques in Social Media graph analysis, privacy preserving data analytics, crowdsourcing through gamification and psychology-based sentiment analysis.
ORGANISE the information
The volume of the content in Social Media is huge and the useless information usually outweighs the useful parts (obviously usefulness also depends on what one is looking for). For this reason we will need to deploy intelligent methods for focused analysis of all types of content, strongly based on event recognition and entity extraction techniques, so that information can be clustered around automatically extracted and dynamically evolving topics. New techniques like computational stylometry along with forensic image and video analysis will be needed.
DEDUCE contextual relationships
The context in Social Media is crucial for the deep understanding of its relevance. Context-based content analysis will be needed for the deduction of causal and spatio-temporal relationships that may prove the existence or absence of connections between social media content and actual events and/or entities. Techniques for discovering the provenance of information are considered vital for tracking the content back to its origins.
PORTRAY information in an intuitive way
Finding all this complex information will be useless without an effective presentation layer that will help users understand and further utilise such higher level concepts. Advances in media content delivery will be necessary through simple and appealing interfaces that guide users through the hidden relations (context) which connect the retrieved content and may allow the reader to process more information. The crowd-sourcing mechanism will thus be strengthened and will allow receiving the importance, originality and validity of actual content and related comments. Further, different visualizations based on device capabilities will be offered.
PROTECT personal data
Quality, diversity and accuracy of information have traditionally been important policy goals in media legislations. In Social Media users are producing and disseminating enormous amounts of information about themselves or others. Hence, a major challenge for media legislation is how to deal with the protection of personal data (both of direct participants and third parties who are referenced, mentioned, portrayed etc.). In the scope of the project we will apply a personal data regulatory framework for safeguarding fundamental rights, freedom of expression, cultural diversity, accuracy and credibility of information, privacy etc.
JOURNALISM Scenario 1 – Event Coverage
As during the 2011 revolution, reports and comments in social networking sites about what was happening in these days from late November 2012 onwards were manifold. And once again, the challenge for journalists and the media was to report on developments and occurrences in the entire country, sense the mood of the different groups and viewpoints, verify what was being reported from various parts of Egypt, analyse the situation, and filter out lies, propaganda, or comments that served particular aims and interests – in order to draw their own conclusions and report the truth about the situation.
JOURNALISM Scenario 2 – Content Production
A journalist is assigned the task to produce an online documentary. The plan is to blend textual information with audio-visual content and material from social networks. The latter may include integrated picture galleries, audio snippets, user-generated videos and photos and a wealth of “social commentary” in the form of posts and tweets.
ENTERPRISE use case
TECHcommunity is the largest online community for Software AG. It is the central point for all kinds of interests regarding Software AG. This can be aggregated by the REVEAL platform, organised, summarised and evaluated, and presented to SAG employees who use it to improve the company’s processes or products.