Beyond Importance Scores: Interpreting Tabular ML by Visualizing Feature Semantics

11/10/2021
by   Amirata Ghorbani, et al.
1

Interpretability is becoming an active research topic as machine learning (ML) models are more widely used to make critical decisions. Tabular data is one of the most commonly used modes of data in diverse applications such as healthcare and finance. Much of the existing interpretability methods used for tabular data only report feature-importance scores – either locally (per example) or globally (per model) – but they do not provide interpretation or visualization of how the features interact. We address this limitation by introducing Feature Vectors, a new global interpretability method designed for tabular datasets. In addition to providing feature-importance, Feature Vectors discovers the inherent semantic relationship among features via an intuitive feature visualization technique. Our systematic experiments demonstrate the empirical utility of this new method by applying it to several real-world datasets. We further provide an easy-to-use Python package for Feature Vectors.

READ FULL TEXT

page 4

page 7

research
04/26/2021

LCS-DIVE: An Automated Rule-based Machine Learning Visualization Pipeline for Characterizing Complex Associations in Classification

Machine learning (ML) research has yielded powerful tools for training a...
research
12/23/2021

AcME – Accelerated Model-agnostic Explanations: Fast Whitening of the Machine-Learning Black Box

In the context of human-in-the-loop Machine Learning applications, like ...
research
01/20/2019

Quantifying Interpretability and Trust in Machine Learning Systems

Decisions by Machine Learning (ML) models have become ubiquitous. Trusti...
research
08/08/2022

EFI: A Toolbox for Feature Importance Fusion and Interpretation in Python

This paper presents an open-source Python toolbox called Ensemble Featur...
research
02/11/2019

LS-Tree: Model Interpretation When the Data Are Linguistic

We study the problem of interpreting trained classification models in th...
research
02/04/2022

The impact of feature importance methods on the interpretation of defect classifiers

Classifier specific (CS) and classifier agnostic (CA) feature importance...
research
07/16/2020

Relative Feature Importance

Interpretable Machine Learning (IML) methods are used to gain insight in...

Please sign up or login with your details

Forgot password? Click here to reset