Text Classifications Using Transductive Confidence Machine For K Nearest Neighbors

Transductive confidence machine for k nearest neighbors is a new algorithm based on algorithmic random theory. It not only can give predictions, but also can provide confidence for every prediction, which is significative for machine learning. However, it always needs a great deal of calculation, for every test example must calculate with every class, especially using on text classification with multi-classes and great datum. In our text, we improve the algorithm by using cluster method, and then use them for text classification. According to the results, the improved algorithm has a similar accuracy to the old algorithm, but a much smaller time at the same time.