You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This project regards an analysis of recent advances in solving the Approximate Jaccard Similarity Search Problem, specifically in regards to how one can achieve sublinear query time using parallel bit counting as presented by Knudsen\cite{fast-similarity-search}. This contains a theoretical analysis of both runtime and correctness of the bit-counting algorithm as well as an empirical comparison to existing methods. The findings include a theoretical run time advantage to using parallel bit counting compared to a simple, linear time algorithm, but empirical results show that this advantage is hard to achieve in practice. Furthermore, reflections are made on how the results might scale on more specialized hardware.
The text was updated successfully, but these errors were encountered:
bachelor-project/abstract/main.tex
Line 9 in 7336b49
The text was updated successfully, but these errors were encountered: