Drug-Drug Interactions (DDIs) are a leading cause of morbidity and treatment failure. The ability to forecast DDIs in order to avert negative consequences is a critical problem. Because there are so many drug-drug interaction pairings, it is hard to test all of them in vitro or in vivo. The high expenditures of DDI research are a restriction. Many medication interactions are caused by enzyme changes in drug metabolism. Cytochrome P450 enzymes are the most frequent of these enzymes (CYP450). Medicines that change the metabolism of other drugs might be substrates, inhibitors, or inducers of CYP450. To overcome the drawbacks and to improve the analysis, the proposed method is implemented. The proposed approach is mainly based on Drug-to-Drug Interaction based on deep learning approach. It is based on Spotted Hyena Optimization Driven Fuzzy Optimized Recurrent Convolution Neural Network (SHO-FORCNN-D2D). The SHO-FORCNN-D2D is executed using PubMed database into the platform of Python. The obtained accuracy of SHO-FORCNN-D2D is 96% and loss is 0.12 that are comparatively lower than other existing techniques.