Related box, freebase-like backed system

[brainstorm]

from @satra:

Much like google generates a related items box based on their knowledge graph, i think the best use of semantic technology would be to create a related box for any question.

This box may contain pointers to code, to websites, to papers, to other relevant discussions.

So this would entail having a box next to a post and building up a database of related areas (like freebase and using freebase where relevant).

[puntonim]

Let's suppose that in a question I mention Django web framework with no actual link. Such "related box" would contain info and links from Django's official website, maybe the Wikipedia page about Django and maybe even the GitHub repo of Django. Something like that? So, how to build such a "database of related areas"? What are the theory and the technologies involved? Could you suggest me something to read to get a better idea?

[puntonim]

From @satra:

This is where the potential for RDF would come in. As an example the related box could contain:

• related questions (if we already have that linked up through RDF, it would be easy to display those)
• related Wikipedia pages (this would require building links from questions to wikipedia)
• related people (again the RDF bit on users and questions would provide some help here)
• related papers
• related data
• related code

In fact we can even provide an option to crowd source these relation. If someone knows a github repo relevant to the question, they could drop the url into the related box. That would allow us to start building up our own RDF database of relations. Finally the related box is really a recommendation engine as roman stated in his message. I think a full-fledged recommendation engine would take up too much effort to be part of the GSOC. However, a simple one could be done based on the RDF extensions.

[satra]

the related box immediately requires the concept of information federation in some way. whether that's done through RDF or through centralized servers remains a practical choice. but the conceptual idea is that we can related information across different sources using a common api instead of having to write a separate one for each information source.

[brainstorm]

This issue was moved to ialbert/biostar-central#297