Enterprise credit score modeling from electricity consumption based on deep ranknet

Qiuying Shen; Wentao Zhang

doi:10.1117/12.2667297

6 March 2023 Enterprise credit score modeling from electricity consumption based on deep ranknet

Qiuying Shen, Wentao Zhang

Proceedings Volume 12553, Fourth International Conference on Optoelectronic Science and Materials (ICOSM 2022); 125531A (2023) https://doi.org/10.1117/12.2667297
Event: 2022 4th International Conference on Optoelectronic Science and Materials (ICOSM2022), 2022, Henfei, China

Abstract

Enterprise credit assessment is important for financial institutes. To enrich the evidence for credit analysis, this paper proposes to use the electricity consumption data to obtain an absolute credit score. Instead of creating the direct mapping between the electricity consumption data and credit score, we train a deep model to predict which enterprise has higher credit score given two enterprises. To learn deep model, we utilize the ranknet model to learn the ranking information from the electricity consumption data. To improve the training efficiency and robustness, we propose a ranking-based representative enterprise sample selection method to optimize the training dataset. During the inference, the learned ranknet model is performed to generate the absolute credit score by a ranking-based score mapping method. The experimental results demonstrate that the method in this paper can achieve accurate enterprise credit evaluation.

Citation Download Citation

Qiuying Shen and Wentao Zhang "Enterprise credit score modeling from electricity consumption based on deep ranknet", Proc. SPIE 12553, Fourth International Conference on Optoelectronic Science and Materials (ICOSM 2022), 125531A (6 March 2023); https://doi.org/10.1117/12.2667297

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available