Description:
Complex network analysis has gained research interests in a wide range of fields. Network motif, which is one of the most popular network properties, is a statistically significant network subgraph. In this paper, we propose a fast methodology, called Four-node Motif Detection Algorithm (FMDA), to extract four-node motifs in complex networks. Specifically, we employ a two-way spectral clustering method to cut big networks into small sub-graphs, and then identify motifs by recognition algorithm to reduce the computational complexity. After that, we use three isomorphic four-node motifs to analyze network structure by American Physical Society (APS) data set.
Description:
An academic team is a highly cohesive collaboration group of scholars, which has been recognized as an effective way to improve scientific output in terms of both quality and quantity. However, the high staff turnover brings about a series of problems that may have negative influences on team performance. To address this challenge, we first detect the tendency of the member who may potentially leave. Here, the outlierness is defined with respect to familiarity, which is quantified by using collaboration intensity. It is assumed that if a team member has a higher familiarity with scholars outside the team, then this member might probably leave the team. To minimize the influence caused by the leaving of such an outlier member, we propose an optimization solution to find a proper candidate who can replace the outlier member. Based on random walk with graph kernel, our solution involves familiarity matching, skill matching, as well as structure matching. The proposed approach proves to be effective and outperforms existing methods when applied to computer science academic teams.
Description:
An academic team is a highly cohesive collaboration group of scholars, which has been recognized as an effective way to improve scientific output in terms of both quality and quantity. However, the high staff turnover brings about a series of problems that may have negative influences on team performance. To address this challenge, we first detect the tendency of the member who may potentially leave. Here, the outlierness is defined with respect to familiarity, which is quantified by using collaboration intensity. It is assumed that if a team member has a higher familiarity with scholars outside the team, then this member might probably leave the team. To minimize the influence caused by the leaving of such an outlier member, we propose an optimization solution to find a proper candidate who can replace the outlier member. Based on random walk with graph kernel, our solution involves familiarity matching, skill matching, as well as structure matching. The proposed approach proves to be effective and outperforms existing methods when applied to computer science academic teams.
Description:
This work aims to explore whether the sustainability of scientific collaboration including collaboration duration and collaboration times can be predicted. For this purpose, we propose a series of features including structural similarity indices, authorship properties, and research interests. Experimental results on a real-world dataset show that our proposed model outperforms baseline model by 10% in MAE. Our study may shed light on investigating scientific collaboration from the perspective of sustainability.
Description:
Link prediction is an important task for analyzing social networks which also has other applications such as bioinformatics and e-commerce. Network representation learning (NRL), which can significantly enhance the performance for link prediction, has attracted much attention in recent years. However, the existing NRL methods mainly focus on observed network structures without considering hidden prediction knowledge in the representation space. Meanwhile, some random walk based NRL methods are dissatisfactory to learn link knowledge in dense networks with large scales. In this paper, we propose a predictive representation learning (PRL) model, which unifies node representations and motif-based structures, to improve prediction ability of NRL. We firstly enhance node representations based on motif-biased random walks and then employ L2-SVM to learn motif-connected node-pairs. By jointly optimizing two objectives of existent and nonexistent edges representations, we preserve more information of nodes in representation space based on supervised learning. To evaluate the performance of our proposed model, we implement experiments on 5 real data sets. Simulation results illustrate that our proposed model achieves better link prediction performance compared with other state-of-the-arts methods.
Description:
The scale of scholarly data has been expanded due to the fact that scientific productions are increasing rapidly and new scholars affiliate academia incessantly. Scholars are shifting their research patterns from individual research to academic teamwork due to the complexity of scientific issues. In order to achieve higher reputations and better performance, academic teams with leaders are constructed to speed up knowledge sharing and problem solving. It is significant to explore team-based issues with the increasing interests of information exploration in big scholarly data. However, existing academic team definitions are commonly not quantitative, which makes it difficult to identify real academic teams. In this work, we propose a collaboration relationship evaluation index called Collaboration Intensity Index (CII), which is a two-way and quantitative index to evaluate collaboration intensity between two scholars in the network. Then, we construct a new type of co-author network with edges weighted by CII, which differs from the original co-author networks. This network reflects the newly scientific research patterns inside or outside academic teams. Furthermore, we propose TRAC (Team Recognition Algorithm based on CII) to identify academic teams from large co-author networks. Finally, we use DBLP data set, which contains 1,250,440 scholars and 1,575,949 published papers, to identify teams by TRAC. Comparing with fast unfolding algorithm and real team data, the effectiveness of our method can be demonstrated.