As in sample by sample classification, when a classifier is asked to combine information across multiple samples drawn from the same data source, the results are combined using a strategy such as majority vote. To solve the problem of classification failure (i.e., a hazard function) in multi sample classification, Multi-aggregative factored K-NN Classifier is proposed. This method evaluates the classification of multi sample problems, such as electromyographic (EMG) data, by making aggregate features available to a per-sample classifier. It is found that the accuracy of this approach is superior to that of traditional methods such as majority vote for this problem. The classification improvements of this method, in conjunction with a confidence measure expressing the per-sample probability of classification failure (i.e., a hazard function) is described and measured.