Abstract: This study aims to investigate the network pharmacology of Chinese medicinal formulae for treatment of Alzheimer's disease. Machine learning algorithms were applied to construct classifiers in predicting the active molecules against 25 key targets toward Alzheimer's disease (AD). By extensive data profiling, we compiled 13 classical traditional Chinese medicine (TCM) formulas with clinical efficacy for AD. There were 7 Chinese herbs with a frequency of 5 or higher in our study. Based on the predicted results, we built constituent-target, and further construct target-target interaction network by STRING (Search Tool for the Retrieval of Interacting Genes/Proteins) and target-disease network by DAVID (Database for Annotation, Visualization and Integrated Discovery) and gene disease database to study the synergistic mechanism of the herbal constituents in the Chinese traditional patent medicine. By prediction of blood-brain penetration and validation by TCMsp (traditional Chinese medicine systems pharmacology) and Drugbank, we found 7 typical multi-target constituents which have diverse structure. The mechanism uncovered by this study may offer a deep insight into the action mechanism of TCMs for AD. The predicted inhibitors for the AD-related targets may provide a good source of new lead constituents against AD.