Fig. 5From: Locality-sensitive hashing enables efficient and scalable signal classification in high-throughput mass spectrometry raw dataControlling typical similarity of pairs: probability \(P_{m,n}([x_{i},x_{j}])\) to retrieve a pair for several combined hashes as a function of similarity \(s(x_{i},x_{j})\). By appropriate choice of m and n, found pairs have a high probability of having at least a certain similarity. Note that the x-axis starts at \(s(x_{i},x_{j})=0.55\)Back to article page