ClaVis: An Interactive Visual Comparison System for Classifiers

Authors. Frank Heyen, Tanja Munz, Michael Neumann, Daniel Ortega, Ngoc Thang Vu, Daniel Weiskopf, Michael Sedlmair

Venue. AVI (2020) Full Paper

Type. Full Paper

Materials. DOI PDF supplemental

Abstract. We propose ClaVis, a visual analytics system for comparative analysis of classification models. ClaVis allows users to visually compare the performance and behavior of tens to hundreds of classifiers trained with different hyperparameter configurations. Our approach is plugin-based and classifier-agnostic and allows users to add their own datasets and classifier implementations. It provides multiple visualizations, including a multivariate ranking, a similarity map, a scatterplot that reveals correlations between parameters and scores, and a training history chart. We demonstrate the effectivity of our approach in multiple case studies for training classification models in the domain of natural language processing.

@inproceedings{10.1145/3399715.3399814,
    title        = {ClaVis: An Interactive Visual Comparison System for Classifiers},
    author       = {Heyen, Frank and Munz, Tanja and Neumann, Michael and Ortega, Daniel and Vu, Ngoc Thang and Weiskopf, Daniel and Sedlmair, Michael},
    year         = {2020},
    booktitle    = {Proceedings of the International Conference on Advanced Visual Interfaces},
    publisher    = {Association for Computing Machinery},
    series       = {AVI '20},
    doi          = {10.1145/3399715.3399814},
    url          = {https://doi.org/10.1145/3399715.3399814},
    abstract     = {We propose ClaVis, a visual analytics system for comparative analysis of classification models. ClaVis allows users to visually compare the performance and behavior of tens to hundreds of classifiers trained with different hyperparameter configurations. Our approach is plugin-based and classifier-agnostic and allows users to add their own datasets and classifier implementations. It provides multiple visualizations, including a multivariate ranking, a similarity map, a scatterplot that reveals correlations between parameters and scores, and a training history chart. We demonstrate the effectivity of our approach in multiple case studies for training classification models in the domain of natural language processing.},
    articleno    = {9},
    numpages     = {9},
    keywords     = {visual analytics, Visualization, machine learning, classifier comparison}
}

Acknowledgements. Funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) – Project-ID 251654672 – TRR 161 (A08) and under Germany’s Excellence Strategy – EXC-2075 – 39074001