This repository contains HER2 training and test sets used for evaluating MSclassifier and other packages in the software article entitled "MSclassifier: median-supplement model-based classification tool for automated knowledge discovery." The training set is comprised of 100 instances and 74 attributes while the test set is comprised of 62 instances and 74 attributes. The training samples were used to obtain results from a 10-fold cross-validation testing of how MSclassifier and other packages accurately predicted HER2-receptor status phenotypes in breast cancer in the article. The data used in the software article was obtained from the supplementary data of "Adabor ES, Acquaah-Mensah GK, Machine learning approaches to decipher hormone and HER2 receptor status phenotypes in breast cancer, Briefings in Bioinformatics 2019; 20 (2): 504–514, https://doi.org/10.1093/bib/bbx138" by permission of Oxford University Press. Here, it is reproduced by permission of Oxford University Press.

{"references": ["Emmanuel S Adabor, George K Acquaah-Mensah, Machine learning approaches to decipher hormone and HER2 receptor status phenotypes in breast cancer, Briefings in Bioinformatics, 2019, Volume 20, Issue 2, Pages 504-514, https://doi.org/10.1093/bib/bbx138"]}