# Towards a More Reliable Interpretation of Machine Learning Outputs for Safety-Critical Systems using Feature Importance Fusion

arXiv: https://arxiv.org/abs/2009.05501 · HTML: https://mlsanity.com/paper/2009.05501/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 11 Sep 2020.

## Built on by (most cited first, 11 of 11)

- [Estimating the Contamination Factor's Distribution in Unsupervised Anomaly Detection](https://mlsanity.com/paper/2210.10487/) (2022)
- [Causal Feature Selection for Responsible Machine Learning](https://mlsanity.com/paper/2402.02696/) (2024)
- [Mechanistic Interpretation of Machine Learning Inference: A Fuzzy Feature Importance Fusion Approach](https://mlsanity.com/paper/2110.11713/) (2021)
- [McUDI: Model-Centric Unsupervised Degradation Indicator for Failure Prediction AIOps Solutions](https://mlsanity.com/paper/2401.14093/) (2024)
- [EFI: A Toolbox for Feature Importance Fusion and Interpretation in Python](https://mlsanity.com/paper/2208.04343/) (2022)
- [Reconnoitering the class distinguishing abilities of the features, to know them better](https://mlsanity.com/paper/2211.12771/) (2022)
- [What's meant by explainable model: A Scoping Review](https://mlsanity.com/paper/2307.09673/) (2023)
- [Rethinking Unsupervised Outlier Detection via Multiple Thresholding](https://mlsanity.com/paper/2407.05382/) (2024)
- [Pareto Data Framework: Steps Towards Resource-Efficient Decision Making Using Minimum Viable Data (MVD)](https://mlsanity.com/paper/2409.12112/) (2024)
- [Helix 1.0: An Open-Source Framework for Reproducible and Interpretable Machine Learning on Tabular Scientific Data](https://mlsanity.com/paper/2507.17791/) (2025)
- [Anomaly Detection with Adaptive and Aggressive Rejection for Contaminated Training Data](https://mlsanity.com/paper/2511.21378/) (2025)

## Abstract (arXiv)

When machine learning supports decision-making in safety-critical systems, it is important to verify and understand the reasons why a particular output is produced. Although feature importance calculation approaches assist in interpretation, there is a lack of consensus regarding how features' importance is quantified, which makes the explanations offered for the outcomes mostly unreliable. A possible solution to address the lack of agreement is to combine the results from multiple feature importance quantifiers to reduce the variance of estimates. Our hypothesis is that this will lead to more robust and trustworthy interpretations of the contribution of each feature to machine learning predictions. To assist test this hypothesis, we propose an extensible Framework divided in four main parts: (i) traditional data pre-processing and preparation for predictive machine learning models; (ii) predictive machine learning; (iii) feature importance quantification and (iv) feature importance decision fusion using an ensemble strategy. We also introduce a novel fusion metric and compare it to the state-of-the-art. Our approach is tested on synthetic data, where the ground truth is known. We compare different fusion approaches and their results for both training and test sets. We also investigate how different characteristics within the datasets affect the feature importance ensembles studied. Results show that our feature importance ensemble Framework overall produces 15% less feature importance error compared to existing methods. Additionally, results reveal that different levels of noise in the datasets do not affect the feature importance ensembles' ability to accurately quantify feature importance, whereas the feature importance quantification error increases with the number of features and number of orthogonal informative features.