### Table 6. Improvement by fusion based on named entities for several tasks

"... In PAGE 5: ... This upper level can give a hint on how much of the potential for improvement is exploited by an approach. Table6 shows the optimal performance and the improvement by the fusion based on the optimal selection of a system for each category of topics. Because named entity recognition systems are available, this method could easily be implemented within an information retrieval system.... ..."

Cited by 1

### Table 1: kNN fusion based on majority voting.

"... In PAGE 3: ... If there is a tie, the smallest k has the tie breaker. See Table1 for a concrete ex- ample where we assume there are n classes and k ranges from 1 to 7. The choice of k should normally be odd numbers and less than the square root of the number of samples in the data set.... ..."

### Table 1. Verification performance on ALL5 dataset with user-independent fusion based on Quadratic Discriminant. EERs in %

2005

Cited by 2

### Table 2. Verification performance on COMMON5 dataset with user-independent fusion based on Quadratic Discriminant. EERs in %

2005

Cited by 2

### Table 3. Verification performance on ALL5 dataset with user-dependent fusion based on Quadratic Discriminant. EERs in %

2005

Cited by 2

### Table 4. Verification performance on COMMON5 dataset with user-dependent fusion based on Quadratic Discriminant. EERs in %

2005

Cited by 2

### Table 3 Results of rank fusion based on Borda counting for the three dictionaries using (a) sum rule and (b) product rule

2004

"... In PAGE 11: ...2. Results for rank fusion based on Borda counting Table3 (a) and (b) shows the results obtained with the rank fusion method based on Borda counting using the sum rule and the product rule respectively. The pattern of results across the three dictionaries seen with raw score fusion (i.... In PAGE 12: ... Best results overall were found for rank fusion of candi- date scores using the product rule. These best results are shown in italics in Table3 (b). However, the better re- sults for the product rule versus the sum rule did not reach statistical significance.... ..."

### Table 6. Verification performance on COMMON5 dataset with adapted user- dependent fusion based on Quadratic Discriminant (r = 1). EERs in %

2005

Cited by 2

### Table 1 KL-complexity of multi-channel EEG (I=60)

"... In PAGE 11: ....C. 15 C3-P3 11 11 10 9 10 11 11 9 RESULTS The results of our calculations are summarized in Tables 1 and 2. In Table1 one may observe that KL-complexity of the signal differs from about 5-8 for signals in frontal channels up to 40 and more in occipital ones. In general, the smallest KL- complexity is seen on frontal channels, where in healthy subjects there is no dominant wave frequency and frequency spectrum is relatively uniform.... In PAGE 12: ...12 Unfortunately it is seen from the Table1 that no consistency in the influence of Diazepam administration on KL-complexity is observed, neither between different channels in one subject, nor between the same channels in different subjects. In Table 2 one may observe that embedding dimension lies between 7 and 11, so endorsing the presumption about existence of low-dimensional chaotic attractor.... ..."

### Table 1. Multi-channel examples

"... In PAGE 6: ... Since Mu[1] = Mv[1] = 00 and Mu[2] = Mv[2] = 00, we choose y = 2 for both j = 0 and 1 as a key-bit of type 1. The two paths from node 0 to node 64 of class 0 are shown in Table1 . The key-bit is shown with boldface.... In PAGE 6: ...y Case 2.2. We choose y = 2 for both j = 0 and 1. Since c(0) u = c(1) v and c(1) u = c(0) v , we have |Z0| = |Z1| = 1. The two paths from node u = 0000000001 to node v = 1101000000 are shown in Table1 . |P0| = |P1| = d(u,v)+4+|Z1| = 5+ 4+1 = 10, where d(1,64) = H(1,64)+22 = 1+4 = 5.... ..."