Sentence examples for rates for classification from inspiring English sources

Table 4 Estimated Precision (P) and Recall (R) rates for classification of substructure patterns.

In general, balanced training data resulted in the lowest overall error rates for classification experiments (82.3% and 78.3% for the binary and multiclass experiments respectively).

For CA 153 antigen (fixed cutoff: 28 U ml−1) combined with Endo180 (variable cutoff: 0.95 1.65 relative levels), the respective true-positive and false-positive rates for classification of metastasis was 94 97% and 32 48% (Supplementary Figure 2).

In addition, the overall accuracy rate for classification scenarios that include discrimination of human voice is very high.

In Fig. 4, the LF bands signal alone on the rightmost can contribute 79.27 % accuracy rate for classification; similarly, HF can get 88.33 % accuracy rate alone.

This was shown to be the most reliable approach for differentiating accurate and inaccurate predictions and for estimating the prediction accuracy (the root mean square error, RMSE, for classification models or the correct prediction rate for classification models), as described elsewhere [27],[28].

To evaluate the predictive performance, we calculated the prediction mean-squared error for regression and the error rate for classification.

The child's attachment behaviors will be rated for classification into secure-B, or into avoidant-A or ambivalent-C as insecure-organized attachment qualities, according to the standard criteria described by Ainsworth [ 22], and for classification into insecure-disorganized/disoriented-D attachment qualities, according to the Main and Solomon [ 35] coding system.

Table 2 Rates for correct classification of two syllables belonging to the same class accepting 5%% false positives Class comp.

True/false-positive rates for metastasis classification were: 85%/50% for the reference standard, CA 15-3 antigen (28 U ml−1); ⩽97%/⩾36% for Endo180; and ⩽97%/⩾32% for CA 15-3 antigen+Endo180.

The DET curve is a plot of error rates for binary classification systems, in which the lower left curve implies the better performance.