Every week, we provide interesting links from around the web that discuss issues regarding the verification processes of content from social media. Here is our list for this week.
Free Image Verification Data Corpus
Last week, we discussed current challenges of image verification, as there are quite a few examples of fake images that are disseminated via social networks (very interesting discourse here – in German). It is obvious that verifying multimedia content is key in the journalistic process. As a result, REVEAL has decided to create a repository of “fake and real images that were disseminated via Twitter in the context of different events” and make that publicly available for the research community (on Github). That way, researchers can experiment with various algorithms that can be trained to detect images during different events. “Our hope is that this dataset will become a valuable resource for researchers working in the area of media verification. One aim of making it available for free is to aid in the reproducibility of research results and the development of new approaches. We welcome ideas and contributions to grow and improve this dataset.” The dataset is primarily maintained by CERTH researcher Christina Boididou (follow her on Twitter @CMpoi). Try it out, we are looking forward to your feedback.
Source: REVEAL project
SocialSensor App to discover newsworthy trends
A project closely related to REVEAL is called SocialSensor. Its objective is to “develop a new framework for enabling real-time multimedia indexing and search in the Social Web”. Vy recently, the project launched an app for the iPhone that “automatically surfaces the most ‘newsworthy’ and fast moving trends. It aggregates pictures, hashtags, posts from multiple networks. The user can filter results by time, network, location, popularity, recency or by the trust of those in the network.” You can download the for free here and start browsing social networks. At present, it is available in beta and for iPhones only.
Source: Social Sensor
Tow Center at Columbia University to focus on verification
Craig Silvermann has been named a fellow with the Tow Center for Digital Journalism at Columbia University. He’ll conduct research on how “news organizations deal with rumors and unconfirmed information, and to identify best practices for how journalists can debunk misinformation”. Results and findings will be publish on a dedicated website and in a final paper. We are looking very much forward to the results of Craig’s work and wish him every success in this endeavour. If you want to follow Craig Silverman and stay updated, check out his Twitter account. Of course, we will also report about his work in progress.
Source: Craig Silverman
We interview people who deal with the verification of content in social networks on various levels. You can check out some very insightful findings and positions as part of our interview series here:
More to come!
Source: REVEAL project