Why choose this project topic?
This topic turns assessing minority-class performance without leaking evaluation information into a measurable analytical question. Comparing precision-recall area, calibration and subgroup error gives a student material for visual results, methodological criticism and a discussion of how the data's limitations change the conclusion.
How does class imbalance in loan models affect assessing minority-class performance without leaking evaluation information, as measured by precision-recall area, calibration and subgroup error?
Choose an accessible licensed or synthetic dataset, agree the unit of analysis and narrow assessing minority-class performance without leaking evaluation information to a reproducible comparison your supervisor can review.
Proposed project objectives
- 01Define the data, assumptions and comparison design for assessing minority-class performance without leaking evaluation information.
- 02Use licensed anonymised or synthetic repayment data, compare weighting and resampling within training folds and report threshold-specific outcomes.
- 03Compare precision-recall area, calibration and subgroup error and report uncertainty and sensitivity to analytical choices.
A suggested research approach
Use licensed anonymised or synthetic repayment data, compare weighting and resampling within training folds and report threshold-specific outcomes. Confirm the data licence and variable definitions before analysis. Preserve an untouched evaluation set where relevant, document exclusions and missingness, and report precision-recall area, calibration and subgroup error with uncertainty or sensitivity checks rather than selecting only favourable results.
What you will need
- Licensed or synthetic repayment table
- Leakage-safe modelling pipeline
- Documented target definition
Keep your project scope clear
A classroom model must not determine real credit eligibility or imply causal risk factors.
Data Science project chapter outline
Use this outline as a starting point. You can edit the chapter titles to match your department’s format during setup.
- Chapter 1Introduction
- Chapter 2Literature Review
- Chapter 3Research Methodology
- Chapter 4Presentation and Analysis of Results
- Chapter 5Summary, Conclusion and Recommendations
Turn this topic into your own final year project.
Your title, department, research question and outline are ready. Add your institution, personalise the details and continue to your project workspace.
Generate the Complete Project Generation uses your word balance. Review the draft and supply your own verified research findings.