Problem Database — Idea for a CP-related project

Revision en1, by Um_nik, 2021-10-13 12:55:28

I can sometimes see people doing some CP-related projects for their courses/portfolio/community, and I want to suggest an idea of such a project that would benefit me as a CodeChef Admin and all the problemsetters, I believe.

Issue: There are a lot of problems on different platforms, and it is really hard to check if some problem was already used somewhere or not. Sometimes googling works, but most of the time I have to rely on my own experience and memory, which are far from perfect. It's hard to search for problems, because they usually have some kind of story, and the same problem can be formulated in a lot of different ways.

Proposed solution: Create some kind of smart database, that will collect the problems from all known sources and will have some very smart search engine that would focus on the internal meaning of the problem, instead of how it is formulated.

I have no idea how to do any of that :) I would assume that it would require some kind of very smart ML and a lot of community effort. But I thought that it might be an interesting project for someone and I really hope that it will be done by some amazing person.

Just a database of problems from different sources with any kind of full-text search would also be nice.

I think I'm ready to pay some reasonable money ($1000? I understand that it's small money for a high-level ML person, but I'm not rich) for a well-done job, and maybe we can crowdfund it even more.

Tags side-project, machine learning


  Rev. Lang. By When Δ Comment
en1 English Um_nik 2021-10-13 12:55:28 1509 Initial revision (published)