Evaluation of Semantic Context Data… What’s missing?

Reading Time: 1

At the moment I’m working on my thesis “Extraction of Synonyms in User-Generated Content”. During this work I’ve seen that we have a lot of semantic useful data (actually not far enough), but no real tools that are able to classify or evaluate this data in a automatic or semi-automatic way. Take for example the discussion about the context (as context I don’t understand only the location, appointments or other dynamic context, but also the interests of the user or the tags of published papers/post/comments/…), there are a lot of papers around that specifies what a context is and how you could apply it in a theoretically way to different domains. On the other side there are no tools that are able to crawl through your personal data and extract from it the current content.

What do we do if we need better tools? Right, we write them by our own. During my thesis work I’ll write a context crawler that abstracts your FOAF, SIOC, SCOT, … files to a SCOT file with tags that indicates your context. For example in this file you found your interests side by side with the tags of your publications. The work is only part of the thesis and at the moment not really finished. If I’m able to publish the code or the concept you could expect at least a proof-of-concept in the next few months.

Hopefully it’s possible to publish at least this part of the thesis under a public licence.