Visual Exploration of Feature-Class Matrices for Classification Problems

Wolfgang Kienreich, Christin Seifert

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

    23 Downloads (Pure)

    Abstract

    When a classification algorithm does not work on a data set, it is a non-trivial problem to figure out what went wrong on a technical level. It is even more challenging to communicate findings to domain experts who can interpret the data set but do not understand the algorithms. We propose a method for the interactive visual exploration of the feature-class matrix used to represent data sets for classification purposes. This method combines a novel matrix reordering algorithm revealing patterns of interest with an interactive visualization application. It facilitates the investigation of feature-class matrices and the identification of reasons for failure or success of a classifier on the feature level. We discuss results obtained by applying the method to the Reuters text collection.
    Original languageEnglish
    Title of host publicationProceedings of the 3rd International EuroVis Workshop on Visual Analytics (EuroVA)
    EditorsK. Matkovic, G. Santucci
    PublisherEurographics Association
    Number of pages5
    ISBN (Print)978-3-905673-89-0
    DOIs
    Publication statusPublished - 1 Jun 2012
    Event9th International EuroVis Workshop on Visual Analytics, EuroVA 2012 - Brno, Czech Republic
    Duration: 4 Jun 20124 Jun 2012
    Conference number: 9
    http://www.eurova.org/previous-events/eurova-2012

    Conference

    Conference9th International EuroVis Workshop on Visual Analytics, EuroVA 2012
    Abbreviated titleEuroVA
    CountryCzech Republic
    CityBrno
    Period4/06/124/06/12
    Internet address

    Fingerprint Dive into the research topics of 'Visual Exploration of Feature-Class Matrices for Classification Problems'. Together they form a unique fingerprint.

    Cite this