The invention discloses a data processing method and system based on clustering anonymization and differential privacy protection. The data processing method comprises the following steps: performingconversion and normalization processing on data in an original data set; performing clustering processing on the data set subjected to conversion and normalization processing to obtain a plurality ofdata clusters with attributes mutually independent; dividing the plurality of data clusters, calculating the centroid value of each equivalent group, and replacing the value of the data record of thecorresponding equivalent group with the centroid value to obtain a plurality of anonymized equivalent groups; adding noise to the equivalent group with the certain data record deleted according to theparallel combination property of differential privacy protection, and obtaining a query set meeting differential privacy; and calculating the similarity, and when the similarity is greater than 0, releasing the data set of the equivalent group meeting the differential privacy protection. According to the method, the calculation expenditure and the privacy budget are effectively saved, the accuracy of a non-interactive differential privacy protection high-dimensional data release query result is improved, and the availability of the release data is ensured.