F
Franck Dernoncourt
Researcher at Adobe Systems
Publications - 170
Citations - 3270
Franck Dernoncourt is an academic researcher from Adobe Systems. The author has contributed to research in topics: Computer science & Sentence. The author has an hindex of 22, co-authored 109 publications receiving 2161 citations. Previous affiliations of Franck Dernoncourt include Massachusetts Institute of Technology & University of Central Florida.
Papers
More filters
Posted Content
A Discourse-Aware Attention Model for Abstractive Summarization of Long Documents
Arman Cohan,Franck Dernoncourt,Doo Soon Kim,Trung Bui,Seokhwan Kim,Walter Chang,Nazli Goharian +6 more
TL;DR: This work proposes the first model for abstractive summarization of single, longer-form documents (e.g., research papers), consisting of a new hierarchical encoder that models the discourse structure of a document, and an attentive discourse-aware decoder to generate the summary.
Proceedings ArticleDOI
Sequential Short-Text Classification with Recurrent and Convolutional Neural Networks
Ji Young Lee,Franck Dernoncourt +1 more
TL;DR: This article proposed a model based on recurrent neural networks and convolutional neural networks that incorporates the preceding short texts for dialog act prediction. But this model is not suitable for short texts and it cannot handle many short texts occur in sequences (e.g., sentences in a document or utterances in a dialog), and most existing ANN-based systems do not leverage the preceding sentences when classifying a subsequent one.
Proceedings ArticleDOI
A Discourse-Aware Attention Model for Abstractive Summarization of Long Documents
Arman Cohan,Franck Dernoncourt,Doo Soon Kim,Trung Bui,Seokhwan Kim,Walter Chang,Nazli Goharian +6 more
TL;DR: The authors propose a hierarchical encoder that models the discourse structure of a document and an attentive discourse-aware decoder to generate the summary, which significantly outperforms state-of-the-art models.
Journal ArticleDOI
De-identification of patient notes with recurrent neural networks.
TL;DR: The first de-identification system based on artificial neural networks (ANNs), which requires no handcrafted features or rules, unlike existing systems, is introduced, which outperforms the state-of-the-art systems.
Journal ArticleDOI
Comparing deep learning and concept extraction based methods for patient phenotyping from clinical narratives.
Sebastian Gehrmann,Sebastian Gehrmann,Franck Dernoncourt,Franck Dernoncourt,Yeran Li,Yeran Li,Eric Carlson,Eric Carlson,Joy T. Wu,Joy T. Wu,Jonathan Welt,Jonathan Welt,John Foote,John Foote,Edward T. Moseley,Edward T. Moseley,David W. Grant,David W. Grant,Patrick D. Tyler,Patrick D. Tyler,Leo Anthony Celi +20 more
TL;DR: The results indicate that CNNs are a valid alternative to existing approaches in patient phenotyping and cohort identification, and should be further investigated, and the deep learning approach presented in this paper can be used to assist clinicians during chart review or support the extraction of billing codes from text by identifying and highlighting relevant phrases for various medical conditions.