The application belongs to the technical field of
data processing. An intelligent financial data anomaly auditing method and
system are provided. After a first suspected anomaly
database and a second suspected anomaly
database are combined, a final anomaly
database is obtained. For data that is not determined to be abnormal, financial related features are extracted to construct a
feature vector, and a
support vector machine model is used to calculate the probability of belonging to each type of anomaly. When the probability is greater than a set threshold, the corresponding data is added to the final anomaly database. According to the amount of abnormal data, the
importance weight of the business, and the
impact coefficient on the financial situation, the final anomaly severity
evaluation result is determined. Through the automatic data collection, cleaning,
standardization, and anomaly identification and analysis process, the time and
workload of manual operation are greatly reduced, the financial data audit can be completed in a shorter time, and the efficiency of the audit work is improved.