Distance-based Disagreement Classifiers Combination

Abstract

We present a methodology to analyze Multiple Classifiers Systems (MCS) performance, using the diversity concept. The goal is to define an alternative approach to the conventional recognition rate criterion, which usually requires an exhaustive combination search. This approach defines a Distance-based Disagreement (DbD) measure using an Euclidean distance computed between confusion matrices and a soft-correlation rule to indicate the most likely candidates to the best classifiers ensemble. As case study, we apply this strategy to two different handwritten recognition systems. Experimental results indicate that the method proposed can be used as a low-cost alternative to conventional approaches.

Single-base mismatch profiles for NGS samples

“…In fact, the SBI can be seen as the confusion matrix of a classification problem on the four classes A,G,C,T, where the classifier is the alignment pipeline and the ground truth is the reference genome. Within this framework, several measures are available in literature for translating a confusion matrix into a single performance value: see for instance [2][3][4] for a review of the more classical methods, or [5] for novel measures or [6][7][8] for totally different approaches. Among others, the Matthews Correlation Coefficient (MCC, for short) has recently attracted the attention of many researchers in the field: in particular, it has been designed as the elective measure in initiatives defining methodologic guidelines such as MAQC-II [9] led by the U.S. Food and Drug Administration (FDA).…”