A Labeling Methodology for Identifying Business Product Innovation in Pharmaceutical Articles

Report
Authors:Ratcliff, Nathaniel, PV-BII SDADUniversity of Virginia Kattampallil, Neil, PV-BII SDADUniversity of Virginia ORCID icon orcid.org/0000-0002-4092-1897
Abstract:

In collaboration with the National Center for Science and Engineering Statistics (NCSES), this effort stems from a larger effort aimed at identifying business product innovations across various economic sectors. Specifically, the purpose of this effort was to create a human-labeled data set for identifying business and product innovation using articles covering the pharmaceutical sector. The human-labeled data is intended to help train supervised machine learning algorithms like BERT (Bidirectional Encoder Representations from Transformers) to enable greater accuracy for the automation of detecting business product innovations.

Keywords:
product innovations , human-labeled data , supervised machine learning algorithms , BERT (Bidirectional Encoder Representations from Transformers)
Rights:
Attribution 4.0 International (CC BY)
Contributor:Lyman, Kimberly, PV-BII SDADUniversity of Virginia
Language:
English
Source Citation:

Ratcliff, N. J., & Kattampallil, N. A. (2022). A labeling methodology for identifying business product innovation in pharmaceutical articles. Proceedings of the Biocomplexity Institute, Technical Report. TR# 2022-013 University of Virginia. https://doi.org/10.18130/eps9-t189

Publisher:
University of Virginia
Published Date:
02/16/2022
Sponsoring Agency:
U.S. National Science Foundation (NSF)
Notes:

This material is based on work supported by U.S. Department of Agriculture (58-3AEU-7-0074) and the National Science Foundation (Contract #49100420C0015). Any opinions, finding, and conclusions or recommendations expressed in this material are those of the authors.

Persistent Link:
https://doi.org/10.18130/eps9-t189