Credit risk assessment is one of the most important aspects of financial decision-making processes. This study presents a systematic review of the literature on the application of Artificial Intelligence (AI) and Machine Learning (ML) techniques in credit risk assessment, offering insights into methodologies, outcomes, and prevalent analysis techniques. Covering studies from diverse regions and countries, the review focuses on AI/ML-based credit risk assessment from consumer and corporate perspectives. Employing the PRISMA framework, Antecedents, Decisions, and Outcomes (ADO) framework and stringent inclusion criteria, the review analyses geographic focus, methodologies, results, and analytical techniques. It examines a wide array of datasets and approaches, from traditional statistical methods to advanced AI/ML and deep learning techniques, emphasizing their impact on improving lending practices and ensuring fairness for borrowers. The discussion section critically evaluates the contributions and limitations of existing research papers, providing novel insights and comprehensive coverage. This review highlights the international scope of research in this field, with contributions from various countries providing diverse perspectives. This systematic review enhances understanding of the evolving landscape of credit risk assessment and offers valuable insights into the application, challenges, and opportunities of AI and ML in this critical financial domain. By comparing findings with existing survey papers, this review identifies novel insights and contributions, making it a valuable resource for researchers, practitioners, and policymakers in the financial industry.

This study explores the determinants of control loss in eating behaviors, employing decision tree regression analysis on a sample of 558 participants. Guided by Self-Determination Theory, the findings highlight amotivation (β = 0.48, p < 0.001) and external regulation (β = 0.36, p < 0.01) as primary predictors of control loss, with introjected regulation also playing a significant role (β = 0.24, p < 0.05). Consistent with Self-Determination Theory, the results emphasize the critical role of autonomous motivation and its deficits in shaping self-regulation. Physical characteristics, such as age and weight, exhibited limited predictive power (β = 0.12, p = 0.08). The decision tree model demonstrated reliability in explaining eating behavior patterns, achieving an R² value of 0.39, with a standard deviation of 0.11. These results underline the importance of addressing motivational deficits in designing interventions aimed at improving self-regulation and promoting healthier eating behaviors.

The goal of this work was to create and assess machine-learning models for estimating the risk of budget overruns in developed projects. Finding the best model for risk forecasting required evaluating the performance of several models. Using a dataset of 177 projects took into account variables like environmental risks employee skill level safety incidents and project complexity. In our experiments, we analyzed the application of different machine learning models to analyze the risk for the management decision policies of developed organizations. The performance of the chosen model Neural Network (MLP) was improved after applying the tuning process which increased the Test R² from −0.37686 before tuning to 0.195637 after tuning. The Support Vector Machine (SVM), Ridge Regression, Lasso Regression, and Random Forest (Tuned) models did not improve, as seen when Test R² is compared to the experiments. No changes in Test R²’s were observed on GBM and XGBoost, which retained same Test R² across different tuning attempts. Stacking Regressor was used only during the hyperparameter tuning phase and brought a Test R² of 0. 022219.Decision Tree was again the worst model among all throughout the experiments, with no signs of improvement in its Test R²; it was −1.4669 for Decision Tree in all experiments arranged on the basis of Gender. These results indicate that although, models such as the Neural Network (MLP) sees improvements due to hyperparameter tuning, there are minimal improvements for most models. This works does highlight some of the weaknesses in specific types of models, as well as identifies areas where additional work can be expected to deliver incremental benefits to the structured applied process of risk assessment in organizational policies.

The Consumer Price Index (CPI) is a vital gauge of economic performance, reflecting fluctuations in the costs of goods, services, and other commodities essential to consumers. It is a cornerstone measure used to evaluate inflationary trends within an economy. In Saudi Arabia, forecasting the Consumer Price Index (CPI) relies on analyzing CPI data from 2013 to 2020, structured as an annual time series. Through rigorous analysis, the SARMA (0,1,0) (12,0,12) model emerges as the most suitable approach for estimating this dataset. Notably, this model stands out for its ability to accurately capture seasonal variations and autocorrelation patterns inherent in the CPI data. An advantageous feature of the chosen SARMA model is its self-sufficiency, eliminating the need for supplementary models to address outliers or disruptions in the data. Moreover, the residuals produced by the model adhere closely to the fundamental assumptions of least squares principles, underscoring the precision of the estimation process. The fitted SARMA model demonstrates stability, exhibiting minimal deviations from expected trends. This stability enhances its utility in estimating the average prices of goods and services, thus providing valuable insights for policymakers and stakeholders. Utilizing the SARMA (0,1,0) (12,0,12) model enables the projection of future values of the Consumer Price Index (CPI) in Saudi Arabia for the period from June 2020 to June 2021. The model forecasts a consistent upward trajectory in monthly CPI values, reflecting ongoing economic inflationary pressures. In summary, the findings underscore the efficacy of the SARMA model in predicting CPI trends in Saudi Arabia. This model is a valuable tool for policymakers, enabling informed decision-making in response to evolving economic dynamics and facilitating effective policies to address inflationary challenges.

High-quality development in China requires higher vocational education, scientific and technological innovation, and sustainable economic development. The spatial distribution patterns of these factors show higher levels in the east and coastal areas compared to the west and inland regions, emphasizing the need for coupling coordination with the social economy. This study examines the impact of sustainable economic development on the coupling coordination degree using the spatial Durbin model. The results show a positive promotion and spillover effect, with regional variations. The main factors affecting the difference in coupling coordination are the amount of technology market contracts, fiscal expenditure on science and technology, patent application authorizations, tertiary industry output value, and the number of R&D institutions. According to the grey prediction model, the coupling coordination degree is expected to increase from 2022 to 2025, but achieving primary coordination may still be challenging in some areas. Therefore, strategies that utilize regional characteristics for coordinated development should be developed to improve the level of coupling coordination and create a mutually beneficial environment.