The traditional weighting schemes used in text categorization for the vector space model (VSM) cannot exploit information intrinsic to texts obtained through on-line handwriting recognition or any OCR process. Especially, top n (n > 1) candidates could not be used without flooding the resulting text with false occurrences of spurious terms. In this paper, an improved weighting scheme for text categorization, that estimates the occurrences of terms from the posterior probabilities of the top n candidates, is proposed. The experimental results show that the categorization performances increase for texts with high error rates.