Refined analysis and a hierarchical multi-task learning approach for loan fraud detection

Fraud problems in loan application assessment cause significant losses for finance companies worldwide, and much research has focused on machine learning methods to improve the efficacy of fraud detection in some financial domains. However, diverse information falsification in individual fraud remains one of the most challenging problems in loan applications. To this end, we conducted an empirical study to explore the relationships between various fraud types and analyzed the factors influencing information fabrication. Weak relationships exist among different falsification types, and some essential factors play the same roles in different fraud types. In contrast, others have various or opposing effects on these types of frauds. Based on this finding, we propose a novel hierarchical multi-task learning approach to refine fraud-detection systems. Specifically, we first developed a hierarchical fraud category method to break down this problem into several subtasks according to the information types falsified by customers, reducing fraud identification's difficulty. Second, a heterogeneous network with a meta-path-based random walk and heterogeneous skip-gram model can solve the representation learning problem owing to the sophisticated relationships among the applicants' information. Furthermore, the final subtasks can be predicted using a multi-task learning approach with two prediction layers. The first layer provides the probabilities of general fraud categories as auxiliary information for the second layer, which is for specific subtask prediction. Finally, we conducted extensive experiments based on a real-world dataset to demonstrate the effectiveness of the proposed approach.

The effects of person-organization fit on lending behaviors:Empirical evidence from Kiva

Donation-based crowdfunding,as part of impact investment,plays a vital role in promoting economic development and alleviating poverty.In order to enhance the lender's enthusiasm for lending behavior,some platforms,for example Kiva,have introduced groups to facilitate lending.This study examines how the group environment can affect the lenders’behaviors in crowdfunding.It has been found that lenders who join groups contribute 1.2 more loans(about$30-$42)per month than those who do not,but the theoretical mechanism of these differences is unclear.To understand in depth how the group environment affects lending behaviors,we introduce and develop the PersonOrganization fit theory and Free-rider theory in this study.Combining machine-learning techniques with empirical analysis,the results show that the matching degree of motivation between group and lender has a positive effect on the lender behavior,i.e.,lending to loans,and this relationship is weakened by free-riding in large groups.In addition,the group openness can have different effects on lenders of different group sizes.Our research enriches the existing crowdfunding literature and fills the gap in the research on new lending models in crowdfunding,and it will also be useful for crowdfunding platforms in setting the rules for building groups.

Toward Intelligent Financial Advisors for Identifying Potential Clients:A Multitask Perspective

Intelligent Financial Advisors(IFAs)in online financial applications(apps)have brought new life to personal investment by providing appropriate and high-quality portfolios for users.In real-world scenarios,identifying potential clients is a crucial issue for IFAs,i.e.,identifying users who are willing to purchase the portfolios.Thus,extracting useful information from various characteristics of users and further predicting their purchase inclination are urgent.However,two critical problems encountered in real practice make this prediction task challenging,i.e.,sample selection bias and data sparsity.In this study,we formalize a potential conversion relationship,i.e.,user→activated user→client and decompose this relationship into three related tasks.Then,we propose a Multitask Feature Extraction Model(MFEM),which can leverage useful information contained in these related tasks and learn them jointly,thereby solving the two problems simultaneously.In addition,we design a two-stage feature selection algorithm to select highly relevant user features efficiently and accurately from an incredibly huge number of user feature fields.Finally,we conduct extensive experiments on a real-world dataset provided by a famous fintech bank.Experimental results clearly demonstrate the effectiveness of MFEM.