Are you looking for opportunities to contribute to the web of meaning that are appropriate to filling some hours in these last lazy days of summer? Something a little less taxing than, say, creating and publishing a Linked Data set on the web?
They’re out there. Here are a few to keep you engaged while you’re soaking up the sun, hopefully on some tropical island with a warm breeze blowing and a cool drink in hand. For those of you at this week’s Semantic Web Technology and Business conference, don’t worry – these should still be waiting for your input when you get back.
- A project of the Semantic Multimedia Research Group at the Hasso Plattner Institut is looking to use the wisdom of the crowd to help with its Fact Ranking Web Application. It needs to evaluate the quality of its heuristic algortihms in determining what data about a subject is the most relevant for the mainstream information consumer. It wants to “create a new ground truth that later will be publicly available and open for all researchers.” Current fact-ranking systems, it says, lack a gold standard corpus to serve as that ground truth for performance evaluation. Those participating in the study will be presented with an entity, and asked to note some facts that occur to them about that entity. They’ll then be displayed additional facts generated automatically and randomly from DBpedia and asked to vote on the importance of each. As an example, it notes that most people would consider the fact that Albert Einstein was a physicist to be more important than the fact that he was about 5’ 6” tall.
- Dartmouth College’s Tiltfactor studio that designs and studies games for social impact this summer extended its Metadata Games platform – which lets users help label image, audio and video collections by providing descriptions or keywords through tagging – via a collaboration with the British Library. Ships Tag, Book Tag, and Portrait Tag are calling on users to engage with over 1 million public domain images found in the British Library’s collection of some 65,000 digitized books from the 17th through the 19th centuries. The aim is to help build the British Library’s metadata for these images, helping to make the content searchable and expanding the collection’s accessibility for public research, reuse, and repurposing, according to a press release about the initiative.
- Phrase Detectives – it’s not hot off the presses, having been launched in 2009. It’s only got 517 completed documents to date (the most recent concluded at the end of July), however, but I still think it’s a fun way to while away a few minutes while helping computers get smarter at understanding text. The game is all about anaphoric references, leading users to search for relationships between words and phrases in a piece of text. The idea behind the game was to use the data gathered to train anaphora resolution systems to improve text summarisation and search engine indexing.
If you’ve got a favorite fun way to advance the semantic web and all its related components, feel free to share it below.