Location

Online

Event Website

https://hicss.hawaii.edu/

Start Date

3-1-2022 12:00 AM

End Date

7-1-2022 12:00 AM

Description

Many companies rely on professional debt-collection agencies to handle their outstanding debts. These agencies conduct a debt collection process consisting of successive, escalating actions with the aim of getting a debtor to settle an overdue claim. The sequence of actions is administered by agents who often have to make decisions on a case-by-case basis. This requires understanding of complex data and making decisions under uncertainty. This decision-making process has hardly been investigated so far. We are proposing Bayesian networks as the analytical basis for a decision support system. Bayesian networks are strong in dealing with uncertainties. They can be used for both predicting the success of a case and making recommendations on actions. The evaluation shows that Bayesian networks have a very good predictive performance which gets even better as the process evolves. With this instrument, the agents can make better-informed decisions in the debt collection process.

Share

COinS
 
Jan 3rd, 12:00 AM Jan 7th, 12:00 AM

Analysis of a Debt Collection Process Using Bayesian Networks

Online

Many companies rely on professional debt-collection agencies to handle their outstanding debts. These agencies conduct a debt collection process consisting of successive, escalating actions with the aim of getting a debtor to settle an overdue claim. The sequence of actions is administered by agents who often have to make decisions on a case-by-case basis. This requires understanding of complex data and making decisions under uncertainty. This decision-making process has hardly been investigated so far. We are proposing Bayesian networks as the analytical basis for a decision support system. Bayesian networks are strong in dealing with uncertainties. They can be used for both predicting the success of a case and making recommendations on actions. The evaluation shows that Bayesian networks have a very good predictive performance which gets even better as the process evolves. With this instrument, the agents can make better-informed decisions in the debt collection process.

https://aisel.aisnet.org/hicss-55/da/machine_learning_in_finance/2