Sentence examples for a selectiveness from inspiring English sources

Overall, except for three degenerated cases where EXCELLENT got 23% faster than FINE, we obtained an average slowdown of 81% with a selectiveness improvement of 130%.

For instance, for L, d, r, q = 100, 14, 5, 6 (p = 11) we obtain a selectiveness of 99.992%, 99.957% and 99.835% for the methods FINE, GOOD and EXCELLENT, respectively.

We further supposed that the extra cost of solving the Parallelogram q-hits Chaining Problem would be smaller for higher values of threshold p. What Table 1 shows us is in agreement with these expectations, since for the cases in which d/ L ≥ 10% and p/ L ≥ 25%, we obtain a selectiveness improvement of 69% in contrast to a time slowdown of 20%.

Here, like in the previous comparison, but now with even more striking numbers, it is clear that EXCELLENT performs better than FINE, bringing a selectiveness improvement of 4.45 against a time slowdown of only 25% for error cases larger than p/ L ≥ 25%.

For instance, if we have a selectiveness improvement of 30% against a slowdown of 75% (like the general average numbers for all 198 cases), it may still be worth it if the algorithm we are going to submit the filtered sequence to is, for instance, cubic, since 1.33 > 1.75.

For instance, we observed that the 15 cases with time slowdown higher than 4 (ranging from 4.28 to 9.36, with average 5.85) are exactly the 15 cases where p/ L ≤ 14%, and we can still verify a selectiveness improvement of 55% in contrast to a time slowdown of 38% for large error cases (d/ L ≥ 10%) with p/ L > 14% (not shown on Table 1).

Using GOOD rather than FINE, we obtained an overall selectiveness improvement (SI) of plus 68% with a slowdown (SD) of only plus 3.2%.

On the other hand, both FINE* and GOOD* have a U-like selectiveness curve with a minimal selectiveness (FINE* reached his minimal for q = 6 and GOOD* reached his minimal for q = 5) – this behaviour is also very typical for these methods.

For instance, the temporal limit of binocular rivalry has several characteristics that are consistent with activity patterns of (binocular) complex cells in the primary visual cortex, e.g. independencies of eye-of-origin information, and contrast-polarity, and a pattern-selectiveness of adaptation [44] [47].

Yield is viewed as a proxy to selectiveness and student preference - it takes into account students with multiple offers to similarly ranked colleges.

Studies have shown that the older the persons are, the higher the number of drop outs, leading to a risk of selectiveness [ 39].