Sentence examples for shuffling labels from inspiring English sources

Nearly all modifications showed a classification accuracy of at least 55% (in H) and 75% (in IMR90), which is above the classification accuracy of 50% expected at chance (we verified that classification accuracy on randomly shuffling labels was found to be ∼50%).

For each shuffle, we associated all terms from GO with the shuffled labels to retain the term dependency.

In addition, we compared the observed classifier performance with the clustering of the compounds with randomly shuffled labels (100 trials).

We confirmed that the deviation of most of the histone modifications from a set of elements with randomly shuffled labels is statistically significant for total normalized read counts within −1 to +1 kb of the element.

To evaluate the robustness of the method we performed a non-parametric statistical test by randomly shuffling 1000 times the labels for each MeSH disease and counting how many times m the AUC computed with randomly shuffled labels is larger than the AUC computed with the true labels.

BACCs were regarded as significant if the number of BACCs of the shuffled label vectors exceeding that of the original label vector was lower than 10 out of 1000 (p < 0.01) [ 58].

Note that randomization might also be performed by permuting the gene labels in the correlation matrix (equivalent to shuffling node labels in the network); in this case, the null distribution will differ from the one obtained here.

For yeast and human dataset, randomization testing was carried out straightforwardly by shuffling gene labels in the expression matrix as described previously [60].

In order to estimate the significance of the texture coding in each neuron, texture information was averaged over time and compared to the average values of information obtained by shuffling texture labels across trials (p<0.05 see Materials and Methods).

The rate of false positives was then estimated by randomly shuffling sample labels 100 times.

By shuffling node labels we sidestep the connection bias of highly annotated (highly connected) gene set members.