Role In Evolution Of Networks Research Articles

Link prediction based on the mutual information with high-order clustering structure of nodes in complex networks

In complex networks, link prediction can forecast missing links and identify spurious interactions has wide applications in the real world. Although the high-order structure plays a vital role in network evolution, its effect on link prediction is not always taken into consideration in traditional prediction algorithms. In this paper, we come up with a novel link prediction approach based on Mutual information of the High-Order Clustering structure (MHOC). The MHOC approach integrates the effects of multiple higher-order structures of nodes and quantifies the different contributions of common neighbors with the aid of the diverse high-order clustering coefficients of nodes based on information entropy for predicting missing links. Experimental results demonstrate that in the real world network, the high-order clustering patterns of nodes can improve link prediction accuracy significantly.

Link prediction in complex networks based on Significance of Higher-Order Path Index (SHOPI)

Finding missing links in an observed network (static) or predicting those links that may appear in the future (dynamic) is the aim of the link prediction (LP) task. LP plays a significant role in network evolution, as shown by several works (Barabasi and Albert,1999; Kleinberg, 2000; Leskovec et al., 2005). However, this problem is still challenging for the authors. Most approaches are based on the topological properties of networks like degree, clustering coefficient, path index, etc. The common neighbor approaches are based on the idea “Friends of a friend are also friends,” i.e., a large number of common friends between two persons (nodes) signifies more similarity between them and more likely to be connected. In the resource allocation process, a large number of connections of common neighbors of two nodes are vulnerable for leaking information (resources) through them. Based on this idea, we proposed a new similarity index SHOPI (Link prediction based on S̄ignificance of H̄igher Ōrder P̄ath Īndex) that tries to constrain the information leak through the common neighbors by penalizing them. Moreover, higher-order paths (as defined by six degrees of separation) are used as discriminating features with penalizing the longer paths available between the seed node pair. The experimental results on twelve real-world network datasets (collected from diverse areas) show that SHOPI outperforms the baseline methods. Moreover, SHOPI is more robust than the existing Katz index and comparable to the local path index (LP). The statistical test shows the significant difference of the proposed method (i.e., SHOPI) with the baseline approaches.

Emergence of Innovation Networks from R&D Cooperation with Endogenous Absorptive Capacity

This paper extends the existing literature on strategic R&D alliances by presenting a model of innovation networks with endogenous absorptive capacity. The networks emerge as a result of bilateral cooperation over time between firms occupying different locations in the knowledge space. Social capital is ignored, and firms ally purely on the basis of knowledge considerations. Partner selection is driven largely by absorptive capacity which is itself influenced by cognitive distance and investment allocation between inventive and absorptive R&D. Cognitive distance between firms changes as a function of the intensity of cooperation and innovation. Within different knowledge regimes, we examine the structure of networks that emerge and how firms perform within such networks. Our model replicates some stylised empirical results on network structure and the contingent effects of network position on innovative performance. We find networks that exhibit small world properties which are generally robust to changes in the knowledge regime. Second, subject to the extent of knowledge spillovers, certain network strategies such as occupying brokerage positions or maximising accessibility to potential partners pay off. Third and most importantly, absorptive capacity plays an important role in network evolution: firms with different network strategies indeed differ in the build-up of absorptive capacity.

The gendered transition to college: The role of culture in ego-network evolution

This article argues that gender norms and students’ entering network structure affect their transition to college. More specifically, just as Bourdieu (1977) posited that the education system privileges the culture of the dominant classes, I argue that elite universities favor those students who enter with sparse, diverse networks – the network capital of the adult upper class – and that their presence creates incentives that lead all students to adopt this structure. I predict that cultural mandates that encourage women to cultivate and manage ties actively will foster a more satisfying social transition than for men, who rely more on the very dense networks that the elite university environment undermines. I find that after one year at an elite university, students’ networks are sparser and more diverse. Interviews reveal that men from dense networks experience a particularly difficult social transition to the university. Because gendered cultural norms contribute so greatly to tie formation in this sample, I conclude that culture plays a key role in network evolution.

Learning and Predicting the Evolution of Social Networks

Graph Evolution Rules help in analyzing the evolution of large networks and can be used to predict the future creation of links among nodes.