In order to estimate the performance of SIPs for the identification of highly similar citations, a set of 10 000 random citation pairs, as well as a 171 MEDLINE duplicates (citations tagged as duplicates in MEDLINE), were used to estimate the algorithm performance, following our previously reported method (Errami et al., 2008).
The following is our algorithm to update cwnd and ssthreshaccording to our estimated bandwidth and loss rate.
Usually, these algorithms treat only the spectral features, while as we discuss in this and following sections, our algorithm converts both spectral features and residual data.
Our results are new and interesting in the following contexts: (i) Our algorithm (3.1) is not regularized by contractions.
The higher GRM(H) is, the higher recovery capability a given matrix H has. Once the weakest and the strongest parts of the matrix H are identified, the following step of our algorithm consists in identifying the global minimum of CRM j) of H (CRM min (j)), that is a minimum of recovery capability in the matrix.
What we will describe in the following turns our watershed algorithm into a hierarchical algorithm in the Beucher (1994) sense.
The following algorithm reflects our experience of post-LC bile leaks at a tertiary care center.
The following algorithm illustrates our approach hereby denoted by CA-FR: 1. Train the cortical network using all features provided.
The following algorithm formalizes our explanations so far; for the stop criterion, we need both, the optimal CCR and BCC efficiency: (g_k^) and (g_k^).
Using scaled-up scoring matrices will eliminate the need for signed number representation in our following algorithm designs.
Finally, we introduce the following proposition to compare our algorithm to LP method [8] theoretically.
