Causality-Based Scores Alignment in Explainable Data Management

1. 1Identify relevant attribution scores such as Causal Responsibility, Shapley Value, and Causal Effect
2. 2Analyze the compatibility of these scores in different query and database scenarios
3. 3Investigate the impact of score alignment on explainable data management
4. 4Develop strategies to improve score alignment and overall system explainability

Data scientists and researchers on a team benefit from this work as it helps to understand the compatibility of different attribution scores, while product managers can use this knowledge to improve the explainability of their data management systems

💡 Different attribution scores may induce incompatible rankings of tuples, highlighting the need for score alignment in explainable data management

📊 Causality-based scores alignment in explainable data management: a first investigation 📈