English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

In a world where Machine Learning (ML) is increasingly deployed to supportdecision-making in critical domains, providing decision-makers withexplainable, stable, and relevant inputs becomes fundamental. Understanding howmachine learning works under missing data and how this affects featurevariability is paramount. This is even more relevant as machine learningapproaches focus on standardising decision-making approaches that rely on anidealised set of features. However, decision-making in human activities oftenrelies on incomplete data, even in critical domains. This paper addresses thisgap by conducting a set of experiments using traditional machine learningmethods that look for optimal decisions in comparison to a recently deployedmachine learning method focused on a classification that is more descriptiveand mimics human decision making, allowing for the natural integration ofexplainability. We found that the ML descriptive approach maintains higherclassification accuracy while ensuring the stability of feature selection asdata incompleteness increases. This suggests that descriptive classificationmethods can be helpful in uncertain decision-making scenarios.

Preservation of Feature Stability in Machine Learning Under Data Uncertainty for Decision Support in Critical Domains