Sentence examples for ranking the data from inspiring English sources

The class probability estimates are rather reliable in this case and can make a difference as compared to randomly ranking the data.

Ranking the data accordingly results in far better AUC ROC values and thus to a more efficient error rate reduction by rejecting the prediction of objects close to the decision boundary (see also below, predictiveness curves).

The startup developed a set of algorithms to crawl and parse the Web's biggest travel information databases (like Wikitravel and Open Street Maps), before ranking the data to determine which pieces have the most relevance to its users and serving them up as rich travel guides.

ROC curve (Fig. 7a) is drawn by first ranking the data based on the prediction score.

This is calculated by ranking the data values from highest to lowest and then splitting the values into five ranges, which do not necessarily contain equal numbers.

We used a relatively simple approach to ranking the data, as currently little is known about the public's knowledge of medicines R&D, and so we felt that ranking the individual areas of medicines R&D within each country would be the most informative approach and in line with the aims of this study.

In a number of cases, there were insufficient data to rank the data and this may exclude sites that may accurately record the paleomagnetic field or produce sample groups too small to provide statistically meaningful comparisons.

We first ranked the data points (x) in each column (i.e. IP experiment).

Zhou et al. proposed a manifold ranking method, which ranks the data with respect to the intrinsic manifold structure collectively revealed by the given data (Zhou et al., 2004).

The rank transformation of the G-EAR data limited the possible merging of results and comparison of effect sizes: rank transformations rank the data smallest to largest and force each data point's rank value to a normal distribution with minimum value 0 and maximum value 1.

We note that although in its original formulation, the one-class SVM is an approach to binary classification, i.e. differentiation between in-class and out-of-class data, it can well be used to rank the data.