TY - GEN
T1 - Content mining to support design
AU - Chonde, Seifu
AU - Kumara, Soundar
AU - Storer, Joey
PY - 2015
Y1 - 2015
N2 - In this paper, we introduce a methodology that makes use of a large number of journal articles to support the design problem. In our methodology we discuss tools to determine proper content in individual citations, link the content in individual and across multiple citations to develop a network, and subsequently use network science to describe the topology and evolution of the network. We label these steps Reading, Assembly, and Interpretation, respectively. We develop a tripartite design network structure and discuss network partitioning and projection into unipartite and bipartite networks for visualization. The chemical design problem is used as a design paradigm with an open access subset of over 700,000 journal articles. In a chemotherapy case study, the most frequent drugs, uses of these drugs, and the distributions of these drugs are discussed. Network science metrics and algorithms are introduced to investigate network topology and evolution. Finally, we conclude by discussing other applications of this work.
AB - In this paper, we introduce a methodology that makes use of a large number of journal articles to support the design problem. In our methodology we discuss tools to determine proper content in individual citations, link the content in individual and across multiple citations to develop a network, and subsequently use network science to describe the topology and evolution of the network. We label these steps Reading, Assembly, and Interpretation, respectively. We develop a tripartite design network structure and discuss network partitioning and projection into unipartite and bipartite networks for visualization. The chemical design problem is used as a design paradigm with an open access subset of over 700,000 journal articles. In a chemotherapy case study, the most frequent drugs, uses of these drugs, and the distributions of these drugs are discussed. Network science metrics and algorithms are introduced to investigate network topology and evolution. Finally, we conclude by discussing other applications of this work.
UR - http://www.scopus.com/inward/record.url?scp=84970940029&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84970940029&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:84970940029
T3 - IIE Annual Conference and Expo 2015
SP - 1672
EP - 1681
BT - IIE Annual Conference and Expo 2015
PB - Institute of Industrial Engineers
T2 - IIE Annual Conference and Expo 2015
Y2 - 30 May 2015 through 2 June 2015
ER -