Contextual Outlier Interpretation (COIN)

Catalogue of Tools & Metrics for Trustworthy AI

These tools and metrics are designed to help AI actors develop and use trustworthy AI systems and applications that respect human rights and are fair, transparent, explainable, robust, secure and safe.

Overview Tools Metrics About the catalogue

Website

Contextual Outlier INterpretation (COIN) is a method designed to explain the abnormality of existing outliers spotted by detectors. The interpretability for an outlier is achieved from three aspects: outlierness score, att that contribute to the abnormality, and contextual description of its neighborhoods.

Trustworthy AI Relevance

This metric addresses Explainability and Robustness by quantifying relevant system properties. Explainability: COIN produces contextual, feature-level explanations that describe why an instance is anomalous relative to its neighbors or local subpopulation, making model behaviour and atypical outputs more understandable to users and operators. This improves comprehensibility of decisions and supports transparent communication of unusual model outputs.

About the metric

You can click on the links to see the associated metrics

Objective(s):

Robustness
Explainability

Purpose(s):

Event/anomaly detection

Lifecycle stage(s):

Build & interpret model

Target users:

Developer

Risk management stage(s):

Define
Assess
Govern

Modify this metric

Disclaimer: The tools and metrics featured herein are solely those of the originating authors and are not vetted or endorsed by the OECD or its member countries. The Organisation cannot be held responsible for possible issues resulting from the posting of links to third parties' tools and metrics on this catalogue. More on the methodology can be found at https://oecd.ai/catalogue/faq.