The research aims to identify potential antidepressants and elucidate their mechanisms in depression treatment through mining depression-related medical literature. Semantic triplets related to depression were extracted using SemRep to determine potential drugs by limiting semantic relationships and types. Potential drugs and depression-related targets were obtained from databases like PubChem and GeneCards, and their intersection was used to construct a protein-protein interaction (PPI) network. The PPI network was analyzed using Cytoscape to identify core targets. Core targets underwent GO (gene ontology) and KEGG (kyoto encyclopedia of genes and genomes) analysis using R software. Finally, molecular docking analysis of core targets and potential drugs was conducted using AutodockTool software. Results revealed Hydrocortisone, Benzodiazepine, Curcumin, Metformin, Nicotine, and Risperidone as potential antidepressants. These drugs exert their antidepressant effects through processes like inflammation, neurotransmitter regulation, and signaling pathways like MAPK (mitogen activated protein kinase) and TNF (tumor necrosis factor). Additionally, Benzodiazepine and Nicotine pose risks of addiction and abuse in clinical practice, potentially limiting their effectiveness in depression treatment. The study demonstrates that utilizing semantic triplets and network pharmacology can unearth novel insights into depression medications, saving time, costs, and offering new directions for clinical drug use.