Data processing method and device

Abstract

The invention discloses a data processing method and device, and relates to the technical field of computers. According to one specific embodiment, the method comprises the steps: obtaining historical time sequence data, wherein the historical time sequence data indicates task order amount information; performing prediction processing on the historical time series data by using multiple time series models, and determining target time series data and a time series model set corresponding to the target time series data according to a prediction value obtained after the prediction processing; according to the target time sequence data and the time sequence features corresponding to the target time sequence data, training each time sequence model in the time sequence model set by using a deep learning algorithm to determine a weight coefficient corresponding to each time sequence model; and determining a prediction result corresponding to the target time sequence data according to the prediction value and the weight coefficient corresponding to each time sequence model. According to the embodiment, the volatility of the prediction result is reduced, the accuracy and interpretability of the prediction result are improved, and the prediction result can meet the layout requirement of the supply chain.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a data processing method and device. Background technique [0002] In response to the rapid development of e-commerce and the sharp increase in online shopping orders, major e-commerce companies have invested a lot in the supply chain of goods, but the realization of the supply chain of goods still needs to rely on sales forecasting technology and sales forecast value. Carry out operations such as replenishment and allocation of online items, and then complete the source enrichment of the supply chain. [0003] In the course of realizing the present invention, the inventor finds that there are at least the following problems in the prior art: [0004] 1. The single time-series model used in the existing data processing methods is difficult to cover the characteristics of all time-series data, resulting in large fluctuations in prediction results and low accuracy; [0...

AI Technical Summary

Problems solved by technology

[0004] 1. The single time-series model used in the existing data processing methods is difficult to cover the characteristics of all time-series data, resulting in large fluctuations in prediction results and low accuracy;

[0005] 2. When the deep learning model is used to process single-volume data of historical tasks, due to insufficient sample data and other reasons, the interpretation of the prediction results is poor, and it is difficult to meet the layout requirements of the supply chain

Images