Culturally-Aware Missing Data Detection: Integrating Contextual Diversity into Automated Evaluation Tools

While most detection frameworks (see: Wang et al., 2023; Preeti Sharma et al., 2022) are agnostic to context, Bassongui (2024) highlights the need for culturally responsive analysis in fragile or diverse environments. This project would build a novel toolset that blends standard anomaly/missingness detection algorithms with modules for encoding contextual knowledge—such as local norms, conflict dynamics, or stakeholder wisdom—into the evaluation and interpretation of data gaps. For example, a checklist or diagram-based diagnostic (inspired by “thought-diversifying tools”) could be integrated with machine learning models to flag not just statistical anomalies, but also contextually meaningful absences (e.g., missing voices in survey data from marginalized groups). This would fill a critical gap in current automated tools, making them more equitable, interpretable, and relevant for global and culturally complex settings.

References:

1. A Methodology to Detect Traffic Data Anomalies in Automated Traffic Signal Performance Measures. Bangyu Wang, Grant G. Schultz, Gregory S. Macfarlane, Dennis L. Eggett, Matthew C. Davis (2023). Future Transportation.
2. Culturally responsive data analysis for evaluations in contexts of fragility, conflict, and violence. Nassibou Bassongui (2024). Evaluation.
3. Comprehensive analyses of image forgery detection methods from traditional to deep learning approaches: an evaluation. Preeti Sharma, M. Kumar, Hiteshwari Sharma (2022). Multimedia tools and applications.