Table 3.
Detailed future research recommendations—Theme: Processes
| References | Future research directions | Sub-themes | Deficit | Authors recommendations | Themes |
|---|---|---|---|---|---|
| Baesens et al. (2005) | Increase data size and examine the impact of time varying inputs | AI and credit | Variables |
• Investigate different variables (e.g., demographic information) and methods (e.g., different feature selection [FS] algorithms) using AI in the credit scoring, analyses, and granting processes • Examine AI-driven credit models using advanced algorithms and practical case studies • Explore new aspects of risks presented with the introduction of AI technologies |
Processes |
| Ince and Aktan (2009) | More research using NN and regression trees for credit scoring models | Method | |||
| Kao et al. (2012) | Investigate the model with different variables | Variables | |||
| Akkoç (2012) | Test the model with different variations | Method | |||
| Koutanaei et al. (2015) | Investigate using other FS algorithms. Perform comprehensive parameters settings | Method | |||
| Ala'raj and Abbod (2016) | Test the model with different variations | Method | |||
| Vahid and Ahmadi (2016) | Test the model developed with different variables | Variables | |||
| Mall (2018) | Test the model developed with different variables | Variables | |||
| Chopra and Bhilare (2018) | Investigate the practical use of the model | Implications | |||
| Jagtiani and Lemieux (2019) | Explore other aspects of risk to borrowers presented by new innovations | New dimension | |||
| Daqar and Arqawi (2020) | Test the model further using demographic information | Variables | |||
| Alborzi and Khanbabaei (2016) | Deploy other ANN, use more relevant variables, and apply association rule technique | AI and services | Method | • Explore different methods (e.g., deploy other artificial neural network approaches) included in the use of AI in financial institutions | |
| Guotai et al. (2017) | Include other financial products | New dimension |