Generation and evaluation of linked data derived from information extraction methodologies

Authors
Gross, Ian
ORCID
Loading...
Thumbnail Image
Other Contributors
McGuinness, Deborah L.
Ji, Heng
Hendler, James A.
Issue Date
2018-05
Keywords
Computer science
Degree
MS
Terms of Use
Attribution-NonCommercial-NoDerivs 3.0 United States
This electronic version is a licensed copy owned by Rensselaer Polytechnic Institute, Troy, NY. Copyright of original work retained by author.
Full Citation
Abstract
By employing the extensible knowledge graph curation and analysis platform called Whyis, we can validate the constructed interactions discovered by our workflow in comparison to interactions stated in a domain-specific assertion database. Assessment of this data flow showcase the associated benefits, difficulties, precision, and potential usages alongside established ontologies. The data workflow implemented in this study provides an innovative approach for knowledge discovery, connectivity, and curation of information using a linked data methodology and assertion-based evaluation criteria.
Many approaches currently exist in the pursuit of knowledge representation from unstructured documents. With the extensive amount of information available, the need to derive a common meaning behind data has become more prevalent than ever before. In this thesis, we develop a workflow to support knowledge extraction and representation of data from the biomedical domain. To determine the hidden meaning behind text, we utilize popular Information Extraction methods to extract events from biomedical papers and store the output from these methods into a combined textual object and annotation representation format. This data is converted into an RDF Graph format based on groundings to scientific ontologies and provenance semantics.
Description
May 2018
School of Science
Department
Dept. of Computer Science
Publisher
Rensselaer Polytechnic Institute, Troy, NY
Relationships
Rensselaer Theses and Dissertations Online Collection
Access
CC BY-NC-ND. Users may download and share copies with attribution in accordance with a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License. No commercial use or derivatives are permitted without the explicit approval of the author.