Results 1 - 10 of 95,631

TABLE C.11 Z-Scores for Trial-Level Standard Deviations, by Laboratory and Gender Data BL LO HI

in Mind/Machine Interaction Consortium: PortREG Replication Experiments
by R. Jahn, B. Dunne, G. Bradish, Y. Dobyns, A. Lettieri, R. Nelson, J. Mischo, E. Boller, H. Bösch, D. Vaitl, J. Houtkooper, B. Walter 2000
Cited by 4

TABLE II Correct identification rate with different training images of the Equinox dataset.

in Statistical Hypothesis Pruning for Identifying Faces from Infrared Images
by Anuj Srivastava, Xiuwen Liu

Table 1: Identification efficiency, eq. (4), and purity, eq. (5), for the training, validation and test samples.

in A neural network approach to event-by-event cosmic
by S. Riggi, R. Caruso, A. Insolia, M. Scuderi
"... In PAGE 7: ... The identification efficiency and purity, relative to the chosen cut at 0.5, are shown in Table1 for the proton and iron mass classes. We tested our method also in presence of a five components primary flux (proton, helium, oxygen,... ..."

Table 3: Identification efficiency, eq. (4), and purity, eq. (5), for the train and test samples.

in A neural network approach to event-by-event cosmic
by S. Riggi, R. Caruso, A. Insolia, M. Scuderi
"... In PAGE 11: ...Results obtained for the classification efficiency and purity are showed in Table3 : as we expected, the performance drops below of around 20% with respect to the pure simulated events. We point out that this results are to be regarded as a first estimate of the performance of the method, since many factors have to be still taken into account, first of all the atmospheric conditions, in order to report conclusive values for the classification matrix.... ..."

Table 6: Comparison of the best configuration for SX against CX Dataset

in Smart Crossover Operator with Multiple Parents for a
by Pittsburgh Learning Classifier, Jaume Bacardit
"... In PAGE 7: ... The most sensitive parameter of the operator is the number of parents used in SX, followed by PSmartX and being the number of repetitions of the rule selection algorithm the least impor- tant parameter. Table6 contains the comparison of several metrics be- tween CX and the selected configuration of SX. We report three metrics: the test accuracy, the average rule-set size of the generated solutions and the algorithm run-time.... ..."

Table 6: Comparison of the best configuration for SX against CX Dataset

in Smart Crossover operator with multiple parents for a
by Pittsburgh Learning Classifier, Jaume Bacardit, Natalio Krasnogor
"... In PAGE 7: ... The most sensitive parameter of the operator is the number of parents used in SX, followed by PSmartX and being the number of repetitions of the rule selection algorithm the least impor- tant parameter. Table6 contains the comparison of several metrics be- tween CX and the selected configuration of SX. We report three metrics: the test accuracy, the average rule-set size of the generated solutions and the algorithm run-time.... ..."

Table 2. Mean and standard deviation of test set error (in percentage) over 20 divisions of training and test sets, for the five datasets, at the three training set sizes (small, medium and large)

in Which is the best multiclass SVM method? An empirical study
by Kai-bo Duan, S. Sathiya Keerthi 2005
"... In PAGE 5: ... The mean and standard devia- tion of the test set error rate (in percentage) are computed over the 20 runs. The results are reported in Table2 . Full details of all runs can be found at:... In PAGE 6: ...Let us now analyze the results from our numerical study. From Table2 we can see that, PWC PSVM gives the best classification results and has significantly smaller mean values of test error. For WTA SVM, MWV SVM and PWC KLR, it is hard to tell which one is better.... In PAGE 6: ... We have also done a finer comparison of the methods by pairwise t-test.The results further consolidate the conclusions drawn from Table2... ..."
Cited by 11

Table 4: Efficient Configurations

in USING ONLINE AUCTIONS FOR OPTIMAL PRODUCT CONFIGURATION AND ASSORTMENT
by Eli M. Snir, Marion G. Sobol, Eli M. Snir, Marion G. Sobol 2005
"... In PAGE 30: ... From the 800 feasible configurations, only 9 are efficient. Table4 details these efficient configurations, which are on the efficient frontier. Table 4: Efficient Configurations ... ..."

Table 6. Results obtained by choosing increasingly easier datasets for training.

in A multi-engine solver for quantified Boolean formulas
by Luca Pulina, O Tacchella
"... In PAGE 10: ... Notice that the size of the sets increases as the time limit decreases, so this experiment explores the possibility of training AQME with a small number of easy-to-solve formulas, and then deploying it on a large number of hard-to-solve ones. In Table6 we show the results of the above experiment. The table is arranged simi- larly to Table 5, modulo the fact that best- and worst-case performances coincide, since the test sets are obtained deterministically.... In PAGE 10: ... The table is arranged simi- larly to Table 5, modulo the fact that best- and worst-case performances coincide, since the test sets are obtained deterministically. Looking at Table6 , we can see that even when training on relatively easy formulas (the 0.5 group), the four AQME versions per- form substantially better than every other engine.... In PAGE 11: ... The number of instances in each of the eight test sets is (from left to right): 96, 202, 138, 197, 121, 104, 146, and 204. The table is then arranged as Table6 . When the training set is biased in favor of a given solver, a dash indicates that the corresponding test set does not contain any formula that can be evaluated by such solver within the time limit.... ..."

Table 14. Statistical results on the quality of candidate identification by SOM and MPRQ. For No. of Complete Correct and Complete Correct Accuracy , first-rank peptide was used for analysis. For specificity and sensitivity, the results for first-rank peptide / best- match peptide are shown.

in Algorithms for Peptide and PTM Identification using Tandem Mass Spectrometry
by Kang Ning 2007
"... In PAGE 12: ...able 13. Parameters for the generation of databases and theoretical spectra...............................................76 Table14 .... In PAGE 89: ...PRQ) as candidates. We used a search distance radius d = 0.25 as the MPRQ parameter. Notice that similar spectra that correspond to the same 2D point can be losslessly retrieved by our algorithm since our algorithm has built an index for these overlapping spectra. In Table14 , the candidate peptides are scored and ranked by SPC only. The best- ranked result (highest SPC) among all candidates is labeled as first-rank peptide.... In PAGE 89: ...From Table14 , it is clear that both the sensitivity and specificity of our algorithm using SOM and MPRQ is high. The sensitivity and specificity of best-match peptides are much higher that those for first-rank peptides, indicating that (i) SPC alone is not a good ... In PAGE 91: ... Efficiency One of the most important features of our algorithm is speed. For batch processing of multiple spectra queries, we can see from Table14 and Table 16 that our algotithm can complete peptide identification for large spectrum datasets ( gt; 500 spectra) in less than 30 secs (e.... In PAGE 101: ... Another important question is: among the candidate sequences, how many of them are identical to the real peptide sequences. We have given the complete correct accuracy in Table14... ..."
Results 1 - 10 of 95,631