In the field of finance, the underwriting process is an essential step in evaluating every loan application. During this stage, the borrowers' creditworthiness and ability to repay the loan are assessed to ultimately decide whether to approve the loan application. One of the core components of underwriting is credit scoring, in which the probability of default is estimated. As such, there has been significant progress in enhancing the predictive accuracy of credit scoring models through the use of machine learning, but there still exists a need to ultimately construct an approval rule that takes into consideration additional criteria beyond the score itself. This construction process is traditionally done manually to ensure that the approval rule remains interpretable to humans. In this paper, we outline an automated system for optimizing a rule-based system for approving loan applications, which has been deployed at Hyundai Capital Services (HCS). The main challenge lay in creating a high-quality rule base that is simultaneously simple enough to be interpretable by risk analysts as well as customers, since the approval decision should be accountable. We addressed this challenge through principled submodular optimization. The deployment of our system has led to a 14% annual growth in the volume of loan services at HCS, while maintaining the target bad rate, and has resulted in the approval of customers who might have otherwise been rejected.

To better understand how different psychosocial components motivate the use of mobile fitness apps, this study integrates two major theories in behavior prediction, theory of planned behavior (TPB) and theory of normative social behavior (TNSB). An online survey was conducted with a random sample of undergraduate students (N = 558) registered at a large public university in Singapore. Results show that participants’ outcome expectations, descriptive norms, and perceived behavioral control predicted their intention to use mobile fitness apps, but not injunctive norms. Perceived behavioral control exerted stronger influence on use intention among current users than non-users. In the TPB-TNSB integrated model, group identification significantly moderated the relationship between descriptive norms and use intention, regardless of the user status. This study provides an improved understanding of how components of social influence affect the adoption of mobile fitness app by taking a novel approach to integrate two distinctive theories in behavior prediction.

Applying self-determination theory, this study investigates how users’ tracking behaviors relate to self-determination in the use of smart wearables, which enhances their intrinsic motivation (i.e., enjoyment) to use the device across age groups. An online survey with 494 smartwatch users aged 18 to 76 shows that the frequency of fitness data tracking is positively related to competence, autonomy, and relatedness needs fulfillment; only need for autonomy is positively associated with enjoyment. Age moderates the mediating effect of autonomy on the relationship between tracking fitness data and enjoyment. The findings are discussed from theoretical and practical perspectives.

Smart wearables are revolutionising how users communicate and acquire information. Yet, the user benefits of smart wearables largely depend on the devices’ ability to collect and analyze a large amount of user data, shaping smart wearables-privacy paradox. The current study explores user responses to the smart wearables-privacy paradox through a survey with smartwatch users (N = 494). Using a cluster analysis method, we identified three distinct groups of smartwatch users – ambivalent, benefit-oriented, and neutral – based on their responses to the smart wearables-privacy paradox. The ambivalent users, who exhibit high levels of both perceived benefit and privacy concerns, were the largest group, followed by benefit-oriented, and neutral groups. We found that the ambivalent users, compared to the benefit-oriented users, tend to be young, male and highly educated, and to show high levels of technology self-efficacy and smartwatch usage. However, the ambivalent users displayed less positive attitudes and a lower continued intention of using the smartwatches than the benefit-oriented users.

In order to address the increased privacy and security concerns raised by mobile communications, designers of mobile applications and websites have come up with a variety of warnings and appeals. While some interstitials warn about potential risk to personal information due to an untrusted security certificate, others attempt to take users' minds away from privacy concerns by making tempting, time-sensitive offers. How effective are they? We conducted an online experiment (N = 220) to find out. Our data show that both these strategies raise red flags for users - appeals to instant gratification make users more leery of the site and warnings make them perceive greater threat to personal data. Yet, users tend to reveal more information about their social media accounts when warned about an insecure site. This is probably because users process these interstitials based on cognitive heuristics triggered by them. These findings hold important implications for the design of cues in mobile interfaces.