TY - JOUR
T1 - Academic Influence Aware and Multidimensional Network Analysis for Research Collaboration Navigation Based on Scholarly Big Data
AU - Zhou, Xiaokang
AU - Liang, Wei
AU - Wang, Kevin I.Kai
AU - Huang, Runhe
AU - Jin, Qun
N1 - Publisher Copyright:
IEEE
Copyright:
Copyright 2018 Elsevier B.V., All rights reserved.
PY - 2018/7/25
Y1 - 2018/7/25
N2 - Scholarly big data, which is a large-scale collection of academic information, technical data, and collaboration relationships, has attracted increasing attentions, ranging from industries to academic communities. The widespread adoption of social computing paradigm has made it easier for researchers to join collaborative research activities and share academic data more extensively than ever before across the highly interlaced academic networks. In this study, we focus on the academic influence aware and multidimensional network analysis based on the integration of multi-source scholarly big data. Following three basic relations: Researcher-Researcher, Researcher-Article, and Article-Article, a set of measures is introduced and defined to quantify correlations in terms of activity-based collaboration relationship, specialty-aware connection, and topic-aware citation fitness among a series of academic entities (e.g., researchers and articles) within a constructed multidimensional network model. An improved Random Walk with Restart (RWR) based algorithm is developed, in which the time-varying academic influence is newly defined and measured in a certain social context, to provide researchers with research collaboration navigation for their future works. Experiments and evaluations are conducted to demonstrate the practicability and usefulness of our proposed method in scholarly big data analysis using DBLP and ResearchGate data.
AB - Scholarly big data, which is a large-scale collection of academic information, technical data, and collaboration relationships, has attracted increasing attentions, ranging from industries to academic communities. The widespread adoption of social computing paradigm has made it easier for researchers to join collaborative research activities and share academic data more extensively than ever before across the highly interlaced academic networks. In this study, we focus on the academic influence aware and multidimensional network analysis based on the integration of multi-source scholarly big data. Following three basic relations: Researcher-Researcher, Researcher-Article, and Article-Article, a set of measures is introduced and defined to quantify correlations in terms of activity-based collaboration relationship, specialty-aware connection, and topic-aware citation fitness among a series of academic entities (e.g., researchers and articles) within a constructed multidimensional network model. An improved Random Walk with Restart (RWR) based algorithm is developed, in which the time-varying academic influence is newly defined and measured in a certain social context, to provide researchers with research collaboration navigation for their future works. Experiments and evaluations are conducted to demonstrate the practicability and usefulness of our proposed method in scholarly big data analysis using DBLP and ResearchGate data.
KW - Academic Influence
KW - Analytical models
KW - Big Data
KW - Collaboration
KW - Correlation
KW - Data models
KW - Multidimensional Network Analysis
KW - Navigation
KW - Research Collaboration
KW - Scholarly Big Data
KW - Scholarly Recommendation
KW - Social network services
UR - http://www.scopus.com/inward/record.url?scp=85050630020&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85050630020&partnerID=8YFLogxK
U2 - 10.1109/TETC.2018.2860051
DO - 10.1109/TETC.2018.2860051
M3 - Article
AN - SCOPUS:85050630020
JO - IEEE Transactions on Emerging Topics in Computing
JF - IEEE Transactions on Emerging Topics in Computing
SN - 2168-6750
ER -