I had to re-blog this blog post by Lior Pachter.
It helped me memorize better the FDRs, FPRs, and all of that!
I found Lior’s explanation easy to follow and he also published a ‘cheat’ table (which I already printed and have it on my desk!).
Happy new year to all!
What are confusion matrices?
In the 1904 book Mathematical Contributions to the Theory of Evolution, Karl Pearson introduced the notion of contingency tables. Sometime around the 1950s the term “confusion matrix” started to be used for such tables, specifically for 2×2 tables arising in the evaluation of algorithms for binary classification.
Example: Suppose there are 11 items labeled A,B,C,D,E,F,G,H,I,J,K four of which are of the category blue (also to be called “positive”) and eight of which are of the category red (also to be called “negative”). An algorithm called BEST receives as input the objects without their category labels, i.e. just A,B,C,D,E,F,G,H,I,J,K and must rank them so that the top of the ranking is enriched, as much as possible, for blue items. Suppose BEST produces as output the ranking:
View original post 1,599 more words