Soybean is widely used in diets, and numerous reports have highlighted its antioxidant properties. However, constructing a methodology for rapid identifying and predicting a series of antioxidant active ingredients in Soybean presents certain challenges. Therefore, we introduced the spectrum-effect relationship-ingredient knockout identification technique to identify a series of antioxidant active ingredients in soybean. By combining untargeted metabolomics with network pharmacology, we predicted the antioxidant active ingredients and their target sites. We successfully identified 4 antioxidant active compounds (daidzein, genistein, daidzein, and glycitin) and 10 corresponding antioxidant targets (epidermal growth factor receptor (EGFR), estrogen receptor 1 (ESR1), steroid receptor coactivator (SRC), tumor necrosis factor (TNF), kinase insert domain receptor (KDR), AKT serine/threonine kinase 1 (AKT1), growth factor receptor bound protein 2 (GRB2), signal transducer and activator of transcription1 (STAT1), mitogen-activated protein kinase 8 (MAPK8), B-cell lymphoma-2 (BCL2)) by our analysis. The validation results from cell experiments revealed that glycitin exhibited the best antioxidant activity and significantly influenced the expression of EGFR and the proteins associated with nuclear factor erythroid 2-related factor 2/NAD(P)H quinone dehydrogenase 1 (NRF2/NQO1) signaling pathways. These findings were consistent with the predicted outcomes and were further confirmed in a zebrafish model. It suggests that glycitin may exert antioxidant effects by regulating the expression of EGFR, NRF2, and NQO1 proteins. The results demonstrate that a rapid analytical method for determining antioxidant activity was established.