Neuraminidase is a significant anti-influenza target that plays crucial role in virus replication cycle. The discovery of 150-cavity in Group-1 neuraminidase provides us a novel mentality of designing inhibitor which can bind with both conserved site and 150-cavity. In order to discover novel dual-site-binding inhibitors, a 3D chemical-feature-based pharmacophore model was established to cover dual-site in neuraminidase. The dual-site-binding model was consistent in predicting the binding conformation of Group-1 neuraminidase inhibitor and applied for virtual screening of Specs database. Compound 4 (ZINC05790048) that aligned well to the model was selected after multiple filtrations for molecular dynamics simulations, indicating improved binding energy with neuraminidase. It can sever as the lead compound for a novel series of inhibitors.