The development of farnesyltransferase inhibitors based on the benzophenone scaffold directed against Plasmodium falciparum is considered a strategy in malaria treatment. In this work, quantitative structure–activity relationship (QSAR) was performed to predict the protein farnesyltransferase (PFT) inhibitory activities for a series of 36 benzophenone derivatives. The data set was divided into two subsets of training and test sets, and the best model using multiple linear regression (MLR), with the values of internal and external validity (R2 = 0.884, R2adj = 0.865, R2pred = 0.821, Q2cv =0.822 and R2p=0.811) was found in agreement with the Tropsha and Golbraikh criteria. The applicability domain (AD) was determined using the Williams plot to describe the chemical space for the model used in this study. The model shows that antimalarial activities of benzophenone depend on logP, bpol, MAXDn, and FMF descriptors. These indications prompted us to design new benzophenones PFT inhibitors and predict the value of their anti-malarial activities based on the MLR equation. Docking results reveal that the newly designed benzophenones bind to the hydrophobic pocket and polar contact with high affinity. The predicted results from this study can help to design novel benzophenone as inhibitors of human PFT with high antimalarial activities.