Publication bias refers to the phenomenon that statistically significant, "positive" results are more likely to be published than non-significant, "negative" results. Currently, researchers have to manually identify negative results in a large number of publications in order to examine publication biases. This paper proposes an NLP approach for automatically classifying negated sentences in biomedical abstracts as either reporting negative findings or not. Using multinomial naïve Bayes algorithm and bag-ofwords features enriched by parts-ofspeeches and constituents, we built a classifier that reached 84% accuracy based on 5-fold cross validation on a balanced data set.