Deep Text Prior: Weakly Supervised Learning for Assertion Classification

Vadim Liventsev, Irina Fedulova, Dmitry Dylov

    Результат исследований: Глава в книге, отчете, сборнике статейМатериалы для конференциирецензирование

    3 Цитирования (Scopus)

    Аннотация

    The success of neural networks is typically attributed to their ability to closely mimic relationships between features and labels observed in the training dataset. This, however, is only part of the answer: in addition to being fit to data, neural networks have been shown to be useful priors on the conditional distribution of labels given features and can be used as such even in the absence of trustworthy training labels. This feature of neural networks can be harnessed to train high quality models on low quality training data in tasks for which large high-quality ground truth datasets don’t exist. One of these problems is assertion classification in biomedical texts: discriminating between positive, negative and speculative statements about certain pathologies a patient may have. We present an assertion classification methodology based on recurrent neural networks, attention mechanism and two flavours of transfer learning (language modelling and heuristic annotation) that achieves state of the art results on MIMIC-CXR radiology reports.

    Язык оригиналаАнглийский
    Название основной публикацииArtificial Neural Networks and Machine Learning – ICANN 2019
    Подзаголовок основной публикацииWorkshop and Special Sessions - 28th International Conference on Artificial Neural Networks, Proceedings
    РедакторыVera Kurková, Igor V. Tetko, Pavel Karpov, Fabian Theis
    ИздательSpringer Verlag
    Страницы243-257
    Число страниц15
    ISBN (печатное издание)9783030304928
    DOI
    СостояниеОпубликовано - 2019
    Событие28th International Conference on Artificial Neural Networks, ICANN 2019 - Munich, Германия
    Продолжительность: 17 сент. 201919 сент. 2019

    Серия публикаций

    НазваниеLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Том11731 LNCS
    ISSN (печатное издание)0302-9743
    ISSN (электронное издание)1611-3349

    Конференция

    Конференция28th International Conference on Artificial Neural Networks, ICANN 2019
    Страна/TерриторияГермания
    ГородMunich
    Период17/09/1919/09/19

    Fingerprint

    Подробные сведения о темах исследования «Deep Text Prior: Weakly Supervised Learning for Assertion Classification». Вместе они формируют уникальный семантический отпечаток (fingerprint).

    Цитировать