similarity score

This example is from Wikipedia and may be reused under a CC BY-SA license.
Furthermore, mean similarity scores for all stimuli within each condition, collapsed across participants, were calculated.
In themselves the similarity scores are not very informative, since we have no basis of comparison.
So making parsetrees extend down through the tokens to the separate characters of a text would artificially raise the similarity scores between tree-pairs.
In calculating these mean similarity scores, we corrected for response omissions as was done in the foregoing analysis.
In this case it is necessary to combine evidence from the document rankings rather than document-query similarity scores.
Then, similarity allows us to rank items and even prune those with the lowest similarity scores.
The similarity scores are derived from the ones used in semantic broadening, so that direct synonyms have a similarity of 1.0, and less directly related terms have a lower score.
Word alignment clues can come from various sources such as statistical alignment models, co-occurrence tests, string similarity scores and static dictionaries.
Furthermore, mean similarity scores for the items (nouns and verbs) within each concreteness by cognate status condition, collapsed across the participants within each of the four association conditions, were calculated.
This ensures that groups of identities with high similarity scores contribute more to the local diagonal score than to identities with low similarity scores.
From
Wikipedia