SAVANNAH, Ga.--(BUSINESS WIRE)--Health Discovery Corporation (“HDC”)(OTCBB: HDVY) today announced the issuance of its pioneering U.S. patent covering the well-established and widely-used recursive feature elimination (“RFE”) method for support vector machines (“SVM”), a major addition to its intellectual property portfolio. RFE is a mathematical selection/filtering process that successively eliminates the least important of many data features, then ranks each of the most significant of the relatively few features remaining for simultaneous use in classifying the data using SVMs. The most significant benefit is an increase in the accuracy and usefulness of the classification results, particularly when analyzing voluminous data. A 2002 Machine Learning article by the inventors and others that initially described RFE-SVMs has been repeatedly cited by hundreds of researchers worldwide. Internet search engines provide over a thousand additional references demonstrating the importance and widespread acceptance of this significant enhancement to SVM methodologies.
