Explaining prediction models and individual predictions with feature contributions

We present a sensitivity analysis-based method for explaining prediction models that can be applied to any type of classification or regression model. Its advantage over existing general methods is that all subsets of input features are perturbed, so interactions and redundancies between features ar...

Full description

Saved in:
Bibliographic Details
Published in:Knowledge and information systems 2014-12, Vol.41 (3), p.647-665
Main Authors: Štrumbelj, Erik, Kononenko, Igor
Format: Article
Language:English
Subjects:
Additives
Algorithms
Analysis
Applied sciences
Classification
Computer Science
Computer science
control theory
systems
Data mining
Data Mining and Knowledge Discovery
Data processing. List processing. Character string processing
Database Management
Decision support systems
Equivalence
Exact sciences and technology
Information Storage and Retrieval
Information systems
Information Systems and Communication Service
Information Systems Applications (incl.Internet)
IT in Business
Knowledge
Mathematical models
Memory organisation. Data processing
Methods
Regression
Regression analysis
Regular Paper
Running
Software
Studies
Support vector machines
Visualization
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We present a sensitivity analysis-based method for explaining prediction models that can be applied to any type of classification or regression model. Its advantage over existing general methods is that all subsets of input features are perturbed, so interactions and redundancies between features are taken into account. Furthermore, when explaining an additive model, the method is equivalent to commonly used additive model-specific methods. We illustrate the method’s usefulness with examples from artificial and real-world data sets and an empirical analysis of running times. Results from a controlled experiment with 122 participants suggest that the method’s explanations improved the participants’ understanding of the model.
ISSN:0219-1377
0219-3116
DOI:10.1007/s10115-013-0679-x