Skip to main content
. Author manuscript; available in PMC: 2012 Jun 27.
Published in final edited form as: Proc IEEE Workshop Autom Speech Recognit Underst. 2003 Oct 13;1:378–384. doi: 10.1109/ICCV.2003.1238369

TABLE II.

Classification accuracies for basic word features. Accuracies are given for the hold-out verification testing data.

Feature setting Accuracy (%)
Linear unigram 82.93
RBF unigram 87.50
RBF bigram 84.76
RBF stemmed unigram 83.23
RBF LIWC 78.66
RBF POS 77.71
RBF POS+LIWC 78.96
RBF POS+unigram 86.89
RBF POS×unigram 86.58