Order collaborative management method and system based on big data and deep learning

Abstract

The invention discloses an order collaborative management method and system based on big data and deep learning, and the method comprises the steps: extracting business relation features from first associated order data which is in business association with an unaudited order based on a historical big data set of the unaudited order; extracting flow relation characteristics from second associated order data in flow association with the unaudited order; performing feature reconstruction processing on the business relationship features and / or the process relationship features in the historical big data set, and generating fusion reconstruction features according to a feature reconstruction result; determining whether the unaudited order is a compliant order or not through a deep learning model according to the business relation characteristics, the process relation characteristics and the fusion reconstruction characteristics; obtaining a processing efficiency parameter of the first cooperation round; and based on the processing efficiency parameter, dynamically optimizing the first task allocation sequence of the first collaboration round, and generating the second task allocation sequence of the second collaboration round, thereby improving the efficiency and accuracy of order processing.

Images