The large majority of methods proposed in literature for handwriting recognition assume that words are produced drawing large parts of the ink without lifting the pen, other than horizontal bars and dots. This fundamental assumption, however, does not always hold: while some educational systems provide explicit training for producing continuous handwriting, minimizing the number of pen-up during the production of a word, others do not. As a consequence, whenever the handwriting presents pen-up within a word, the recognition performance can drop significantly. In a preliminary study, we presented an algorithm for discriminating among different types of ink appearing in handwriting, namely isolated characters, cursive, dots, horizontal and vertical bars, based on the use of a suitable set of features. In this paper, we have characterized the discriminative power of each considered feature according to different measures and we have proposed a method for combining the different feature rankings. We have also used the Fischer's Linear Discriminant Analysis (LDA) for exhaustively selecting the best feature subsets with increasing number of features. Finally, we have compared the results obtained by using the feature subsets provided by LDA with those obtained with the feature subsets selected according to our feature ranking. The experimental results, on different datasets of handwritten words, showed that our approach successfully achieves its aim allowing to reduce the computational cost without affecting the overall performance of the recognition process.