SCIENCE@home

Towards a Better Microcredit Decision

Mengnan Song,Jiasong Wang,Suisui Su

Published 2022 in arXiv.org

ABSTRACT

—Microcredit refers to personal oriented loans for consumption, which is mainly characterized by small loans, medium-short term and lacking collateral, and serves high-risk populations who are highly likely to be rejected by traditional financial institutions. A crucial challenge that the leading plat- forms are facing is to assess the creditworthiness of applicants and decide whether or not to grant credit. Therefore, credit scorecard emerges and has drawn broad attention, as it takes an important role in the field of risk management. However, the parameters of scorecard are usually estimated using a sample set that excludes rejected applicants, which may prove biased when applied to all applicants. Reject inference comprises techniques to infer the possible repayment behavior of rejected cases. In this paper, we model credit in a brand new view by capturing the sequential pattern of interactions among multiple stages of loan business to make better use of the underlying causal relationship. Specifically, we first define 3 stages with sequential dependence throughout the loan process including credit granting(AR), withdrawal appli- cation(WS) and repayment commitment(GB) and integrate them into a multi-task architecture. Inside stages, an intra-stage multi- task classification is built to meet different business goals. Then we design an Information Corridor to express sequential depen- dence, leveraging the interaction information between customer and platform from former stages via a hierarchical attention module controlling the content and size of the information channel. In addition, semi-supervised loss is introduced to deal with the unobserved instances. The proposed multi-stage interaction sequence(MSIS) method is simple yet effective and experimental results on a real data set from a top loan platform in China show the ability to remedy the population bias and improve model generalization ability.

PUBLICATION RECORD

CITATION MAP

EXTRACTION MAP

CLAIMS

  • No claims are published for this paper.

CONCEPTS

  • No concepts are published for this paper.

REFERENCES

Showing 1-38 of 38 references · Page 1 of 1

CITED BY