The β-secretase is one of prospective targets against Alzheimer's disease (AD). A three-dimensional quantitative structure-activity relationship (3D-QSAR) model of Hydroethylamines (HEAs) as β-secretase inhibitors was established using Topomer CoMFA. The multiple correlation coefficient of fitting, cross validation and external validation were r2=0.928, qloo2=0.605 and rpred2=0.626, respectively. The 3D-QSAR model was used to search R groups from ZINC database as the source of structural fragments. As a result, a series of R groups with relatively high activity contribution was obtained to design a total of 15 new compounds, with higher activity than that of the template molecule. The molecular docking was employed to study the interaction mode between the new compounds as ligands and β-secretase as receptors, displaying that hydrogen bond and hydrophobicity played important roles in the binding affinity between the new compounds and β-secretase. The results showed that Topomer CoMFA and Topomer Search could be effectively used to screen and design new molecules of HEAs as β-secretase inhibitors, and the designed compounds could provide new candidates for drug design targeting AD.