C
Cheryl Clark
Researcher at Mitre Corporation
Publications - 14
Citations - 521
Cheryl Clark is an academic researcher from Mitre Corporation. The author has contributed to research in topics: Set (abstract data type) & Statistical classification. The author has an hindex of 8, co-authored 14 publications receiving 461 citations.
Papers
More filters
Journal ArticleDOI
The MITRE Identification Scrubber Toolkit: Design, training, and assessment
John S. Aberdeen,Samuel Bayer,Reyyan Yeniterzi,Benjamin Wellner,Cheryl Clark,David A. Hanauer,Bradley A. Malin,Lynette Hirschman +7 more
TL;DR: The open source MITRE Identification Scrubber Toolkit (MIST) provides an environment to support rapid tailoring of automated de-identification to different document types, using automatically learned classifiers to de-identified and protect sensitive information.
Journal ArticleDOI
Negation's not solved: generalizability versus optimizability in clinical natural language processing.
Stephen Wu,Stephen Wu,Timothy A. Miller,James J. Masanz,Matt Coarr,Scott Halgrim,David Carrell,Cheryl Clark +7 more
TL;DR: This work proposes that an optimizable solution does not equal a generalizable solution, and introduces a new machine learning-based Polarity Module for detecting negation in clinical text, and extensively compare its performance across domains.
Journal ArticleDOI
Identifying Smokers with a Medical Extraction System
Cheryl Clark,Kathleen Good,Lesley Jezierny,Melissa Macpherson,Brian Wilson,Urszula Chajewska +5 more
TL;DR: A medical information extraction system that combines a rule-based extraction engine with machine learning algorithms to identify and categorize references to patient smoking in clinical reports and shows overall accuracy in the 90s on all data sets used.
Journal ArticleDOI
Hiding in plain sight: use of realistic surrogates to reduce exposure of protected health information in clinical text
David Carrell,Bradley A. Malin,John S. Aberdeen,Samuel Bayer,Cheryl Clark,Ben Wellner,Lynette Hirschman +6 more
TL;DR: Experimental results suggest approximately 90% of residual identifiers can be effectively concealed by the HIPS approach in text containing average and high densities of personal identifying information, which suggests HIPS is feasible, but requires further evaluation.
Journal ArticleDOI
Habitat-Lite: A GSC Case Study Based on Free Text Terms for Environmental Metadata
Lynette Hirschman,Cheryl Clark,K. Bretonnel Cohen,Scott A. Mardis,Joanne S. Luciano,Renzo Kottmann,James R. Cole,Victor Markowitz,Nikos C. Kyrpides,Norman Morrison,Lynn M. Schriml,Dawn Field +11 more
TL;DR: The goal of the work described here is to provide a light-weight, easy-to-use (small) set of terms ("Habitat-Lite") that captures high-level information about habitat while preserving a mapping to the recently launched Environment Ontology (EnvO).