3 ms·

The name may be an example of this: https://en.m.wikipedia.org/wiki/Stigler%27s_law_of_eponymy https://en.m.wikipedia.org/wiki/Stigler%27s_law_of_eponymy It wa

The name may be an example of this: https://en.m.wikipedia.org/wiki/Stigler%27s_law_of_eponymy https://en.m.wikipedia.org/wiki/Stigler%27s_law_of_eponymy It was developed by Grove Karl Gilbert in 1884 as his ratio of verification (v)[1] and now is frequently referred to as the Critical Success Index in meteorology.[2] It was later developed independently by Paul Jaccard…

dalke 4y ago
In my field, cheminformatics, we refer to it as "Tanimoto similarity" because it was (quoting Wikipedia) "independently formulated again by T. Tanimoto." It's an odd set of linkages to get there. First, "Dr. David J. Rogers of the New York Botanical Gardens" proposed a problem to Tanimoto, who published the writeup in an internal IBM report in 1958. (I understand there was a lot of mathematical research in taxonomy at the time.) In 1960 Rogers and Tanimoto published an updated version in Science. In 1973 Adamson and Bush at Sheffield University developed a method for the automatic classification of chemical structures. They tried Dice, phi, and Sneath as their comparison methods but not Tanimoto. In their updated 1975 publication write "Several coefficients have been proposed based on this criterion", with a list of citations, including the Rogers and Tanimoto paper as citation 14. In 1986, Peter Willett at Sheffield revisits this work and finds that Tanimoto gives overall better results when applied to what are now called cheminformatics "fingerprints". He uses "Tanimoto", with no direct citation for the source of that definition. This similarity method is easy to implement, and many organizations already have pre-computed fingerprints (they are used as pre-filters for graph queries), so the concept and nomenclature takes off almost immediately, with "Tanimoto" as the preferred named. It's not until 1991 that can find a paper in my field referring to the earlier work by Jaccard (the paper uses "Tanimoto (Jaccard)"). I have found some papers in related fields (eg, in IR and mass spectra analysis) which reference Tanimoto similarity, but nothing to the extent that my field uses it.
jszymborski 4y ago
Another example of this sort of thing (that is vaguely related in that it's commonly used as a metric) is (what I call) the Matthew's Correlation Coefficient https://en.wikipedia.org/wiki/Phi_coefficient https://en.wikipedia.org/wiki/Phi_coefficient > In machine learning, it is known as the Matthews correlation coefficient (MCC) ... introduced by biochemist Brian W. Matthews in 1975.[1] Introduced by Karl Pearson,[2] and also known as the Yule phi coefficient from its introduction by Udny Yule in 1912

dalke 4y ago

In my field, cheminformatics, we refer to it as "Tanimoto similarity" because it was (quoting Wikipedia) "independently formulated again by T. Tanimoto." It's an odd set of linkages to get there. First, "Dr. David J. Rogers of the New York Botanical Gardens" proposed a problem to Tanimoto, who published the writeup in an internal IBM report in 1958. (I understand there was a lot of mathematical research in taxonomy at the time.) In 1960 Rogers and Tanimoto published an updated version in Science. In 1973 Adamson and Bush at Sheffield University developed a method for the automatic classification of chemical structures. They tried Dice, phi, and Sneath as their comparison methods but not Tanimoto. In their updated 1975 publication write "Several coefficients have been proposed based on this criterion", with a list of citations, including the Rogers and Tanimoto paper as citation 14. In 1986, Peter Willett at Sheffield revisits this work and finds that Tanimoto gives overall better results when applied to what are now called cheminformatics "fingerprints". He uses "Tanimoto", with no direct citation for the source of that definition. This similarity method is easy to implement, and many organizations already have pre-computed fingerprints (they are used as pre-filters for graph queries), so the concept and nomenclature takes off almost immediately, with "Tanimoto" as the preferred named. It's not until 1991 that can find a paper in my field referring to the earlier work by Jaccard (the paper uses "Tanimoto (Jaccard)"). I have found some papers in related fields (eg, in IR and mass spectra analysis) which reference Tanimoto similarity, but nothing to the extent that my field uses it.

jszymborski 4y ago

Another example of this sort of thing (that is vaguely related in that it's commonly used as a metric) is (what I call) the Matthew's Correlation Coefficient https://en.wikipedia.org/wiki/Phi_coefficient https://en.wikipedia.org/wiki/Phi_coefficient > In machine learning, it is known as the Matthews correlation coefficient (MCC) ... introduced by biochemist Brian W. Matthews in 1975.[1] Introduced by Karl Pearson,[2] and also known as the Yule phi coefficient from its introduction by Udny Yule in 1912