Abstract
Keyword-based search in text-rich multi-dimensional datasets facilitates many novel applications and tools. In this paper, the objects that are tagged with keywords are embedded in a vector space. For these datasets, queries where asked for the tightest groups of points satisfying a given set of keywords. A novel method called ProMiSH (Projection and Multi Scale Hashing) which uses random projection and hash-based index structures, and achieves high scalability and speedup. Thus, the paper introduces an exact and an approximate version of the algorithms.
Users
Please
log in to take part in the discussion (add own reviews or comments).