A wide variety of classification algorithms exist in the literature. Probably the most popular and among the newest is support vector machines (SVM). Older learning algorithms such as k-nearest-neighbor (KNN), decision trees (DTREE) or Bayes' classifier (BC) are well understood and widely applied. One might ask why we are motivated to use LR for classification instead of the usual candidates.
That LR is suitable for binary classification is made clear in Chapter 4. Our motivation for exploring LR as a fast classifier to be used in data mining applications is its maturity. LR is already well understood and widely known. It has a statistical foundation which, in the right circumstances, could be used to extend classification results into a deeper analysis. We believe that LR is not widely used for data mining because of an assumption that LR is unsuitably slow for high-dimensional problems. In Zhang and Oles [49], the authors observe that many information retrieval experiments with LR lacked regularization or used too few attributes in the model. Though they address these deficiencies, they still report that LR is ``noticeably slower'' than SVM. We believe we have overcome the stability and speed problems reported by other authors.