Repository logo
 

ChemDataExtractor: A Toolkit for Automated Extraction of Chemical Information from the Scientific Literature.

Accepted version
Peer-reviewed

Loading...
Thumbnail Image

Type

Article

Change log

Authors

Swain, Matthew C 
Cole, Jacqueline M 

Abstract

The emergence of "big data" initiatives has led to the need for tools that can automatically extract valuable chemical information from large volumes of unstructured data, such as the scientific literature. Since chemical information can be present in figures, tables, and textual paragraphs, successful information extraction often depends on the ability to interpret all of these domains simultaneously. We present a complete toolkit for the automated extraction of chemical entities and their associated properties, measurements, and relationships from scientific documents that can be used to populate structured chemical databases. Our system provides an extensible, chemistry-aware, natural language processing pipeline for tokenization, part-of-speech tagging, named entity recognition, and phrase parsing. Within this scope, we report improved performance for chemical named entity recognition through the use of unsupervised word clustering based on a massive corpus of chemistry articles. For phrase parsing and information extraction, we present the novel use of multiple rule-based grammars that are tailored for interpreting specific document domains such as textual paragraphs, captions, and tables. We also describe document-level processing to resolve data interdependencies and show that this is particularly necessary for the autogeneration of chemical databases since captions and tables commonly contain chemical identifiers and references that are defined elsewhere in the text. The performance of the toolkit to correctly extract various types of data was evaluated, affording an F-score of 93.4%, 86.8%, and 91.5% for extracting chemical identifiers, spectroscopic attributes, and chemical property attributes, respectively; set against the CHEMDNER chemical name extraction challenge, ChemDataExtractor yields a competitive F-score of 87.8%. All tools have been released under the MIT license and are available to download from http://www.chemdataextractor.org .

Description

Keywords

Cluster Analysis, Data Mining, Databases, Chemical, Humans, Natural Language Processing, Pattern Recognition, Automated, Periodicals as Topic

Journal Title

J Chem Inf Model

Conference Name

Journal ISSN

1549-9596
1549-960X

Volume Title

56

Publisher

American Chemical Society (ACS)
Sponsorship
Isaac Newton Trust (1507(l))