BACKGROUND: Curcumin is a plant polyphenol extracted from turmeric, which can be used to treat knee osteoarthritis. However, the specific mechanism is still unclear. OBJECTIVE: Based on bioinformatics and network pharmacology, to explore the molecular mechanism of curcumin in the treatment of knee osteoarthritis, thereby providing a new research direction for the treatment of knee osteoarthritis. METHODS: Curcumin-related targets were analyzed by TCMSP database, STITCH database, Drugbank database, SEA database, SwissTargetPrediction database and Binding DB database. Differentially expressed genes were obtained by GEO2R analysis combined with curcumin-related chips in GEO database. TTD, OMIM, DisGeNET, Drugbank, and GeneCards were used to obtain the disease targets related to knee osteoarthritis and identify the curcumin-knee osteoarthritis common targets and the key targets by setting the degree value. STRING database was used to construct the protein-protein interaction relationship of common targets, and R language was used to analyze gene ontology and Kyoto Encyclopedia of Genes and Genomes pathways of common genes. IGEMDOCK software was used to perform molecular docking analysis on the key targets. RESULTS AND CONCLUSION: A total of 339 curcumin-related targets were obtained by searching the above-mentioned drug databases; a chip with the curcumin-related sequence number GSE10896 was determined by searching the GEO database, and 397 differential genes were screened out. A total of 1 903 disease-related targets were obtained by searching the above-mentioned disease databases, and the typical targets included AKT1, NFKB1, RELA, and IKBKB. The protein-protein interaction network diagram of 129 common targets was constructed by using the STRING database. The gene ontology analysis showed that the biological process of curcumin treatment of knee osteoarthritis mainly covered DNA-binding transcription factor, DNA binding transcription activator activity, RNA polymerase II specificity, endopeptidase activity, protein serine/threonine kinase activity, nuclear receptor activity, ligand activated transcription factor activity, metal peptidase activity, and phosphatase binding. The mainly signal pathways were the PI3K-Akt, MAPK, interleukin-17, tumor necrosis factor and hypoxia-inducible factor-1 signaling pathways. The molecular docking analysis revealed that the key targets such as insulin-like growth factor 1 receptor, estrogen receptor, epidermal growth factor receptor, and CYP19A1 had a good affinity with the target proteins. To conclude, there were eight key targets involved in the treatment of knee osteoarthritis with curcumin, including MAPK3, epidermal growth factor receptor, estrogen receptor 1, MDM2, CYP19A1, MAPK14, estrogen receptor, and insulin-like growth factor 1. These targets inhibit inflammation and chondrocyte apoptosis through PI3K-Akt, MAPK, interleukin-17, tumor necrosis factor, and hypoxia-inducible factor 1 signaling pathways for treating knee osteoarthritis with curcumin. [ABSTRACT FROM AUTHOR]