Furthermore, we choose the words whose CHI statistics feature value is bigger than 3 and the linguistics cues whose feature value is bigger than 0.2 as the feature items, and then conduct the open test using SVM classier. Fig.2 separately shows the recall, precision and Fmeasure values, which obtained by using SVM classifier with three kinds of feature sets on our 793 deceptive texts and 1 191 non-deceptive texts.