Creating Neuroscientific Knowledge Organization System Based on Word Representation and Agglomerative Clustering Algorithm

How to navigate the ever-growing scientific literature database is a pressing problem. Knowledge Organization System (KOS) with hierarchical categorization, which can help us navigate through the literature. We have presented a systematical method to create such KOS which can effectively help us fin...

Full description

Saved in:
Bibliographic Details
Published inFrontiers in neuroinformatics Vol. 14; p. 38
Main Authors Huangfu, Cunqing, Zeng, Yi, Wang, Yuwei
Format Journal Article
LanguageEnglish
Published Lausanne Frontiers Research Foundation 18.08.2020
Frontiers Media S.A
Subjects
Algorithms
Clustering
Cognition & reasoning
Keywords
knowledge oganization system
Knowledge organization
literature analysis
Neuroscience
Neurosciences
Ontology
Research methodology
Researchers
Semantics
Subject heading schemes
word embedding
Online AccessGet full text

Cover

More Information
Summary:How to navigate the ever-growing scientific literature database is a pressing problem. Knowledge Organization System (KOS) with hierarchical categorization, which can help us navigate through the literature. We have presented a systematical method to create such KOS which can effectively help us find sub-topics, provide search keywords, analyze articles related to specific research domains, and characterize the features of article clusters. The method is based on word representation and agglomerative clustering algorithm, which can significantly decrease the inevitable expert manual work required for the KOS organization. We have collected 35832 research keywords and 11497 research methods from PubMed Central XMLs.We have organized them into a hierarchal structure; all the hierarchical nodes have explainable specific meanings. Through two case studies, we demonstrated that the term system is versatile in bibliometric research and scientific literature retrieval.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
This article was submitted to Neuroinformatics, a section of the journal Frontiers in Neuroinformatics
Reviewed by: Mihail Bota, University of Southern California, United States; Guang-Zhong Wang, Shanghai Institute of Nutrition and Health (CAS), China
Edited by: Jan G. Bjaalie, University of Oslo, Norway
ISSN:1662-5196
1662-5196
DOI:10.3389/fninf.2020.00038