Precision determines the fraction of records that actually turns out to be positive in the group the classifier has declared as a positive class.
Precision = TP / ( TP + FP )
Precision Graph
Clearly from the bar graph above , J48 Decision Tree has the highest precision value i.e lowest number of false positive errors committed by this classifier.
Naive Bayes and JRip classifiers almost perform the same with equal number of minimum false positives.
SMO clearly does the worst with maximum number of false positives relative to the other classifiers mentioned above .