Machine Learning in Network Centrality Measures

Abstract

Complex networks are ubiquitous to several Computer Science domains. Centrality measures are an important analysis mechanism to uncover vital elements of complex networks. However, these metrics have high computational costs and requirements that hinder their applications in large real-world networks. In this tutorial, we explain how the use of neural network learning algorithms can render the application of the metrics in complex networks of arbitrary size. Moreover, the tutorial describes how to identify the best configuration for neural network training and learning such for tasks, besides presenting an easy way to generate and acquire training data. We do so by means of a general methodology, using complex network models adaptable to any application. We show that a regression model generated by the neural network successfully approximates the metric values and therefore are a robust, effective alternative in real-world applications. The methodology and proposed machine learning model use only a fraction of time with respect to other approximation algorithms, which is crucial in complex network applications.

Enhancing Internal Learning in Teams: The Role of Network Centrality and Psychological Capital of Undergraduate Students

“…The main purpose of this study was to analyze the mediating role of PsyCap in the relationship between network centrality and internal learning in teams. Knowing more about the relationship between centrality in advice networks and internal learning of undergraduate students is relevant, given the different positive learning outcomes, such as cognitive and innovative adaptation of individuals, the development of performance in different contexts and engagement (Gray and Meister, 2006;Wang et al, 2014;You, 2016;Grando et al, 2018).…”

“…Our third contribution comes from the notion that centrality in advice networks may be useful in fostering PsyCap. These findings fill an important gap in the PsyCap field, which led certain authors such as Grando et al (2018) to highlight the importance of explaining the role of an individual's position on PsyCap fostering (Grando et al, 2018). For Heled et al (2015), the creation of the dissemination mechanisms that enable individuals to share their knowledge and insights has been theorized as a factor that promotes PsyCap.…”

“…The learning development can be considered complex because it can flow within a community since its members can have similar experiences, share the same language, and have a mutual understanding of specific problems (Huber, 1991). In particular, learning has been positively related to the cognitive and innovative adaptation of individuals, measures of network centrality, and performance improvement in different contexts (Gray and Meister, 2006;Wang et al, 2014;Grando et al, 2018). In educational settings, You (2016) has found that the development of students' learning skills allows them to overcome uncertainties regarding the labor market, due to high unemployment rates and low job security, and facilitates the achievement of future goals.…”

“…Although no internal learning studies have included the social network theory directly, more recent research has shown interest in conducting studies that analyze the relationship between individual's psychological capabilities and his position on different types of social networks (Grando et al, 2018). However, there have been just a few attempts to run studies that allow us to examine the effects of the relationship between psychological capital (PsyCap) and network centrality in the learning process.…”

Indexability-Based Dataset Partitioning

“…Centrality is classically defined in relation to a spanning graph. Various definitions of centrality exist [5,9], from the simplest based on node degree, to those exploiting a spectral decomposition of the graph (such as PageRank and others [21]).…”