A Workflow for Visual Diagnostics of Binary Classifiers using Instance-Level Explanations

Josua Krause, Aritra Dasgupta, Jordan Swartz, Yindalon Aphinyanaphongs, Enrico Bertini

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Abstract

    Human-in-the-loop data analysis applications necessitate greater transparency in machine learning models for experts to understand and trust their decisions. To this end, we propose a visual analytics workflow to help data scientists and domain experts explore, diagnose, and understand the decisions made by a binary classifier. The approach leverages 'instance-level explanations', measures of local feature relevance that explain single instances, and uses them to build a set of visual representations that guide the users in their investigation. The workflow is based on three main visual representations and steps: one based on aggregate statistics to see how data distributes across correct / incorrect decisions; one based on explanations to understand which features are used to make these decisions; and one based on raw data, to derive insights on potential root causes for the observed patterns. The workflow is derived from a long-term collaboration with a group of machine learning and healthcare professionals who used our method to make sense of machine learning models they developed. The case study from this collaboration demonstrates that the proposed workflow helps experts derive useful knowledge about the model and the phenomena it describes, thus experts can generate useful hypotheses on how a model can be improved.

    Original languageEnglish (US)
    Title of host publication2017 IEEE Conference on Visual Analytics Science and Technology, VAST 2017 - Proceedings
    EditorsTobias Schreck, Brian Fisher, Shixia Liu
    PublisherInstitute of Electrical and Electronics Engineers Inc.
    Pages162-172
    Number of pages11
    ISBN (Electronic)9781538631638
    DOIs
    StatePublished - Jul 2 2017
    Event2017 IEEE Conference on Visual Analytics Science and Technology, VAST 2017 - Phoenix, United States
    Duration: Oct 1 2017Oct 6 2017

    Publication series

    Name2017 IEEE Conference on Visual Analytics Science and Technology, VAST 2017 - Proceedings

    Other

    Other2017 IEEE Conference on Visual Analytics Science and Technology, VAST 2017
    Country/TerritoryUnited States
    CityPhoenix
    Period10/1/1710/6/17

    Keywords

    • Interpretation
    • Machine Learning
    • Visual Analytics

    ASJC Scopus subject areas

    • Computer Graphics and Computer-Aided Design
    • Human-Computer Interaction
    • Information Systems
    • Media Technology

    Fingerprint

    Dive into the research topics of 'A Workflow for Visual Diagnostics of Binary Classifiers using Instance-Level Explanations'. Together they form a unique fingerprint.

    Cite this