2026-09-20T21:16:51Z https://www.ideals.illinois.edu/oai-pmh

oai:www.ideals.illinois.edu:2142/125604 2025-02-06 col_2142_5131 col_2142_105429 com_2142_5130 com_2142_105424

application/pdf en text Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-02-04 without embargo terms The student, Leon Hounnou, accepted the attached license on 2024-07-10 at 15:53. The student, Leon Hounnou, submitted this Dissertation for approval on 2024-07-10 at 15:56. This Dissertation was approved for publication on 2024-07-12 at 14:01. DSpace SAF Submission Ingestion Package generated from Vireo submission #21040 on 2025-02-04 at 21:04:52 Harnessing the power of machine learning, Bayesian neural networks, and spatial analysis in modeling a predictive system, credit risk, and organizational performance across continents Hounnou, Leon 2024-07-12 Okumu, Moses McNamara, Paul E He, Jing Rui Ansong, David Predictive System Machine Learning Bayesian Analysis Spatial Econometrics. eng In a world increasingly driven by data and technological advancements, addressing pressing global issues requires innovative research approaches. This thesis aligns with this ideal by exploring the potential of Machine Learning, Bayesian Neural Networks, and Spatial Analysis in Education, International Development, and Finance. The work unfolds over three distinct chapters, each tackling urgent issues and striving for transformative solutions. The first chapter focuses on early graders' literacy outcomes in South Africa, designing a predictive system to identify at-risk students and facilitate timely interventions. The second chapter employs Spatial Analysis techniques to understand the organizational dynamics of Area Stakeholder Panels (ASPs) in Malawi, and evaluates the impact of agricultural interventions on smallholder farmers. The third chapter highlights the use of Bayesian Neural Networks in credit risk prediction in the United States, demonstrating the importance of Bayesian inference in accounting for uncertainties. Collectively, these three chapters underscore the interdisciplinary nature of contemporary research, transcending geographical and sectoral boundaries. They emphasize the transformative power of data-driven insights in various fields, highlighting the potential of advanced analytical techniques to drive positive change. This study reinforces the importance of informed decision-making and proactive strategies for addressing global challenges. It contributes to a more inclusive and globally informed future, ensuring that knowledge is harnessed to address critical issues and drive meaningful change. 2024-08 Thesis https://hdl.handle.net/2142/125604 Copyright 2024 Leon Hounnou Illinois Informatics Institute Informatics University of Illinois at Urbana-Champaign Ph.D. Dissertation