共 50 条
Literature-related discovery (LRD): Methodology
被引:42
|作者:
Kostoff, Ronald N.
[1
]
Briggs, Michael B.
Solka, Jeffrey L.
[2
]
Rushenberg, Robert L.
[3
]
机构:
[1] Off Naval Res, Arlington, VA 22217 USA
[2] USN, Surface Weapons Ctr, Dahlgren Div, Dahlgren, VA 22448 USA
[3] DDL OMNI Engn LLC, Mclean, VA 22201 USA
关键词:
literature-based discovery;
text mining;
information retrieval;
clustering;
semantic filters;
D O I:
10.1016/j.techfore.2007.11.010
中图分类号:
F [经济];
学科分类号:
02 ;
摘要:
Literature-related discovery (LRD) is linking two or more literature concepts that have heretofore not been linked (i.e., disjoint), in order to produce novel, interesting, plausible, and intelligible knowledge. LRD has two components: Literature-based discovery (LBD) generates potential discovery through literature analysis alone, whereas literature-assisted discovery (LAD) generates potential discovery through a combination of literature analysis and interactions among selected literature authors. In turn, there are two types of LBD and LAD: open discovery systems (ODS), where one starts with a problem and arrives at a solution, and closed discovery systems (CDS), where one starts with a problem and a solution, then determines the mechanism(s) that links them. The generic methodology for identifying potential discovery candidates through ODS LRD, focusing mainly on its ODS LBD component, is described in this paper. A comprehensive flow chart showing the details of our systematic potential discovery generation process, including the evolution of the flow chart steps through each of the studies performed, is presented. Also shown is a vetting procedure that insures potential discoveries claimed are potential discoveries realized. The semantic filters that replace the numerical filters of other ODS LBD approaches are overviewed. The rationale for addressing the five topics studied (Raynaud's Phenomenon (RP), Cataracts, Parkinson's Disease (PD), Multiple Sclerosis (MS), and Water Purification (WP)) is summarized. (C) 2007 Elsevier Inc. All rights reserved.
引用
收藏
页码:186 / 202
页数:17
相关论文