Looking deeper into deep learning model: attribution-based explanations of TextCNN

Wenting Xiong, Iftitahu Ni'mah, Juan M. G. Huesca, Werner van Ipenburg, Jan Veldsink, Mykola Pechenizkiy

    Onderzoeksoutput: Bijdrage aan congresPaperAcademic

    24 Downloads (Pure)

    Samenvatting

    Layer-wise Relevance Propagation (LRP) and saliency maps have been recently used to explain the predictions of Deep Learning models, specifically in the domain of text classification. Given different attribution-based explanations to highlight relevant words for a predicted class label, experiments based on word deleting perturbation is a common evaluation method. This word removal approach, however, disregards any linguistic dependencies that may exist between words or phrases in a sentence, which could semantically guide a classifier to a particular prediction. In this paper, we present a feature-based evaluation framework for comparing the two attribution methods on customer reviews (public data sets) and Customer Due Diligence (CDD) extracted reports (corporate data set). Instead of removing words based on the relevance score, we investigate perturbations based on embedded features removal from intermediate layers of Convolutional Neural Networks. Our experimental study is carried out on embedded-word, embedded-document, and embedded-ngrams explanations. Using the proposed framework, we provide a visualization tool to assist analysts in reasoning toward the model's final prediction.
    Originele taal-2Engels
    Aantal pagina's9
    StatusGepubliceerd - 8 nov 2018
    EvenementNIPS 2018 Workshop on Challenges and Opportunities for AI in Financial Services: the Impact of Fairness, Explainability, Accuracy, and Privacy - Montreal, Canada
    Duur: 7 dec 20187 dec 2018

    Congres

    CongresNIPS 2018 Workshop on Challenges and Opportunities for AI in Financial Services
    LandCanada
    StadMontreal
    Periode7/12/187/12/18

    Vingerafdruk

    Duik in de onderzoeksthema's van 'Looking deeper into deep learning model: attribution-based explanations of TextCNN'. Samen vormen ze een unieke vingerafdruk.

    Citeer dit