similarity: Compute different similarity coefficients

In Morphonodepredictivemodel/morphonode: The Morphonode Predictive Model for ultrasound signatures detection and malignancy prediction in vulvar cancer

Compute different similarity coefficients

Description

Compute the similarity between two vectors.

Usage

similarity(x, y, f = "cosine", ceiling = 1e+07, ...)

Arguments

x	A numeric vector.
y	A numeric vector.
f	Similarity function. One among "cosine" (default), "jaccard" (for dichotomous vectors only), "pearson", "spearman", "kendall", or "euclidean".
ceiling	If f = "euclidean", the similarity is computed as 1/distance. This argument limits the similarity to a very high value, in case the euclidean distance is equal to 0. The default value is 1E+07.
...	Currently ignored.

Value

A numeric value corresponding to the similarity coefficient.

Author(s)

Fernando Palluzzi [e-mail]

References

Leydesdorff L (2005). Similarity Measures, Author Cocitation Analysis,and Information Theory. In: JASIST 56(7), pp.769-772. <https://doi.org/10.48550/arXiv.0911.4292>

See Also

jaccard, euclidean

Examples

# Sample two random ultrasound profiles from the default dataset
x <- mosaic::sample(mpm.us, 1, replace = FALSE, prob = NULL)
x <- as.numeric(x[, 2:15])
y <- mosaic::sample(mpm.us, 1, replace = FALSE, prob = NULL)
y <- as.numeric(y[, 2:15])

# Compute the cosine similarity
r <- similarity(x, y)
print(r)

Morphonodepredictivemodel/morphonode documentation built on Feb. 15, 2023, 4:51 a.m.