Objective To use data mining method to detect the adverse reaction signal of omalizumab after marketing,and to provide reference for clinical safety and rational drug use．Methods In this study,the report odds ratio method（ROR）and Bayesian confidence propagation neural network（BCPNN）were used to conduct data mining and signal detection for omalizumab-related adverse event（ADE）reports from the FDA Adverse Event Reporting System（FAERS）from the first quarter of 2004 to the second quarter of 2023．Results Through data mining and signal detection,186,353 reports of ADE involving omalizumab were extracted,involving 45,383 patients．Among these reports,the proportion of women（65.31%）was much higher than that of men（24.97%）．The main reporting countries were the United States（64.93%）and Canada（11.96%）．consumers（41.35%）and doctors（36.97%）were the main groups of reporters．The study identified 621 ADE positive signals across 25 system organ classes（SOCs）,including respiratory,chest,and mediastinal diseases（21.29%）and infectious and infectious diseases（10.91%）．Of these,183 signals were assessed as high risk,including 57 new high-risk signals．These findings contribute to a more complete understanding of the safety and potential risks of omalizumab．Conclusions In the clinical application of omalizumab,in addition to the known adverse reactions mentioned in the drug description,special attention should be paid to potential adverse drug events,such as elevated blood pressure,elevated heart rate,intermediate insomnia,and postatic tachycardia syndrome．