High-throughput experimental techniques, such as mass spectrometry and yeast two-hybrid assay, provided an extensive list of protein-protein interactions (PPIs) that control the biological processes in for various organisms. Although these experimental approaches produce crucially valuable data, they sufferfrom high false positives results.In this paper, we proposed a new method to detect false positive interactions. Our method usesinteraction generality andK-path ratio as two topological features of the protein-protein interaction networks, which extracted from high quality data, to detect false positive interactions. The results reveal that proposed method can effectively detect false positive interactions in highthroughput PPI data.