Data and Information Quality - University project (Politecnico di Milano)
Data and Information Quality Project The project explores the impact of traditional and emerging Data Quality (DQ) issues on Machine Learning (ML) analysis, with a specific focus on the "Completeness" dimension. It investigates how Missing Not At Random (MNAR) and Missing Completely At Random (MCAR) data affect ML models' performance in a regression task using synthetic datasets. The project includes data preprocessing strategies, imputation techniques, and robustness evaluation to assess and mitigate the effects of missing data on predictive accuracy.
The final code is contained in the DIQ_PROJECT.ipynb file. The DIQ_project.pdf file contains the final report