This indicates that the confidence score of any link connecting 253- to any source word is at most 1 /N .

From multiple alignments of the same sentence pair, we select high confidence links from different alignments based on their link confidence scores and alignment agreement ratio.

Where C (A) is the confidence score of the alignment A as defined in formula 1.

Select links whose confidence scores are higher than an empirically chosen threshold H as anchor links 1.

2When two equally close alignment links have the same confidence score ), we randomly select one of the tied links as the anchor link.

For each sentence pair, we also calculate the sentence alignment confidence score — log 0 (A|S, T).

measure suggests the possibility of selecting the alignment with the highest confidence score to obtain better alignments.

For each sentence pair in the CE test set, we calculate the confidence scores of the HMM alignment, the Block Model alignment and the MaXEnt alignment, then select the alignment with the highest confidence score .

Next, we built a decision tree classifier that improves the results on average by combining the outputs of the three models together with their confidence scores .

Finally, we will consider other alternatives to the decision tree framework when combining the results of the models with their confidence scores .

As the features of our DT classifier, we use the results of the models that are implemented in Section 4.1, together with the confidence scores calculated for each instance.

To calculate a confidence score for the models, we note that since each model makes its selection based on the language that gives the highest probability, a confidence score should indicate the relative highness of that probability compared to the probabilities of other languages.

where ,u’ and 0’ are the mean and the standard deviation values respectively, for a set of confidence scores calculated for a model on a small development set of 25 annotated queries from each language.

The novelty of our approach is to use “pseudo negative examples” with reliable confidence score estimated by a classifier trained with positive and unlabeled examples.

Using naive Bayes classifier, we can estimate the confidence score c(d, s) that the sense of a data instance “(1”, whose features are f1, f2, ..., fn,

01 foreacth(T—P—N) 02 classify d by WSD system I— (P, T-P) 03 c(d, pos) <— the confidence score that d is

04 predicted as positive defined in equation (2) 05 c(d, neg) <— the confidence score that d is

Distribution of Confidence Scores .

The vast majority of tokens has a confidence score close to l, the median lies at 0.9966.