OBJECTIVE To optimize the extraction technology of modified Tabusen- 2（MT-2），and to investigate inhibitory effects of the extract obtained by the optimal technology on osteoclast differentiation. METHODS The index components of MT- 2 process optimization were selected by using network pharmacology. Based on single factor tests ，the extraction technology of MT- 2 was optimized by Box-Behnken design-response surface methodology according to the comprehensive score of contents of above index components ，and then validated. RAW 264.7 cells were induced by receptor activator of nuclear factor-κB ligand（100 ng/mL） to prepare osteoclast differentiation model. Inhibitory effects of MT- 2 extract（18.6，37.2，74.4 ng/mL）obtained by the optimal technology on osteoclast differentiation were investigated. RESULTS The index components screened by network pharmacology included chlorogenic acid ，terpineol diglucoside ，isochlorogenic acid A ，1，5-dicaffeoylquinic acid ，hydroxysafflower yellow A ， ginsenoside Rg 1 and ginsenoside Rb 1. The optimal extraction technology of MT- 2 was ethanol volume fraction of 60％ ，the solid-liquid ratio of 1 ∶ 14（g/mL），extraction time of 94 min and extraction times of twice. The average comprehensive score obtained by the three validation experiments was 95.50，and the relative error with the predicted value （95.75）was －0.26％. Compared with osteoclastic differentiation model cells ，the cells treated with MT- 2 extract prepared by the optimal technology were mostly mononuclear round cells ，and the number of osteoclasts decreased significantly （P＜0.05），its inhibitory effects tended to strengthen with the increase of drug concentration. CONCLUSIONS The optimal extraction technology of MT- 2 is stable and feasible. Obtained extract can inhibit osteoclast differentiation.