Affiliations: [a] Dong Fureng Economic & Social Development School, Wuhan University, Wuhan, Hubei, China
| [b] China Electric Power International Forwarding Agency Co., Ltd, Beijing, China
| [c] Beijing Guodiantong Network Technology Co., Ltd., Beijing, China
Correspondence:
[*]
Corresponding author. Linglong Meng, China Electric Power International Forwarding Agency Co., Ltd, Beijing, 100011, China. E-mail: bengmengtangm2y@163.com.
Abstract: A supplier portrait generation method based on Big data analysis and deep learning was proposed to help users make reasonable decisions in core links such as procurement and contract signing. This method establishes a label element analysis model for each level in the vertical label system of power supply enterprises, and divides it into target layer, standard layer, and solution layer based on the logic and attributes of the elements, and establishes a hierarchical structure. Compare the index labels of each level with the labels of the upper and lower levels by considering the logical relationship and correlation between each level. Utilize deep learning algorithms to sort hierarchically, and use a multidimensional structural model to represent and fuse portrait labels of power supply enterprises. Based on the imaging results of supplier vertical rating, combined with objective factors such as material production cycle, supply cycle, market supply and demand, price fluctuations, etc., it helps power enterprises effectively predict the supplier’s performance ability. The simulation results show that the reliability of the power supply enterprise portrait generated by this method is high, and the credibility of the portrait identification system for all levels of power supply enterprises is high. This supplier portrait method can effectively improve the supplier management capabilities of power enterprises.
Keywords: Deep learning BCCM, multi-aspect, electricity supplier, portrait generation, information management
