Loved by learners at thousands of companies
If you've ever wanted to understand more about social networks, information networks, or even the neural networks of our brains, then you need to know network science! It will demonstrate network analysis using several R packages, including dplyr, ggplot2, igraph, ggraph as well as visNetwork. You will take on the role of Interpol Analyst and investigate the terrorist network behind the Madrid train bombing in 2004. Following the course, you will be able to analyse any network with basic centrality and similarity measures and create beautiful and interactive network visualizations.
The hubs of the networkFree
The challenge in this chapter is to spot the most highly connected terrorists in the network. We will first import the dataset and build the network. Then we will learn how to visualize it in different layouts using ggraph package. Later on, we will compute two basic yet important centrality measures in network science - degree and strength. We will use them to spot highly connected terrorists. We will finally touch two alternative centrality measures, betweenness and closeness.Network Science50 xpExplore the dataset100 xpBuild and explore the network (part 1)100 xpBuild and explore the network (part 2)100 xpVisualizing networks50 xpVisualize the network (part 1)100 xpVisualize the network (part 2)100 xpCentrality measures50 xpFind the most connected terrorists100 xpFind the most strongly connected terrorists100 xpMore on centrality50 xp
In its weakness lies its strength
In this chapter we will spot the most influential ties among terrorists in the network. We will use a centrality measure on ties, called betweenness, and will learn how to visualize the network highlighting connections with high betweenness centrality. Moreover, we will provide some alternative evidence regarding Mark Granovetter's theory of strength of weak ties, confirming that looser connections are crucial as demonstrated in the Madrid terrorism network.Tie betweenness50 xpBetweenness of ties100 xpFind ties with high betweenness100 xpVisualizing centrality measures50 xpVisualize node centrality100 xpVisualize tie centrality100 xpFilter important ties100 xpThe strength of weak ties50 xpHow many weak ties are there?100 xpVisualize the network highlighting weak ties100 xpVisualize the sub-network of weak ties100 xpMore on betweenness50 xp
The challenge in this chapter is to discover pairs of similar (and dissimilar) terrorists. We will introduce the adjacency matrix as a mathematical representation of a network and use it to find terrorists with similar connection patterns. We will also learn how to visualize similar and dissimilar pairs of individuals using ggraph.Connection patterns50 xpVisualizing connection patterns100 xpThe adjacency matrix (part 1)100 xpThe adjacency matrix (part 2)100 xpPearson correlation coefficient50 xpComputing Pearson similarity100 xpNegative and positive similarity50 xpExplore correlation between degree and strength100 xpMost similar and most dissimilar terrorists50 xpTransforming the similarity matrix100 xpJoin similarity and nodes data frames100 xpFind most similar and dissimilar pairs100 xpVisualize similarity100 xp
In this chapter we will discover cells of similar terrorists. We will explore hierarchical clustering to find groups of similar terrorists building on the notion of similarity of connection patterns developed in the previous chapter. Furthermore, we will explore the visNetwork package to produce fulfilling interactive network visualizations.Hierarchical clustering50 xpCluster the similarity network100 xpCut the dendrogram100 xpAnalyze clusters100 xpVisualize the clusters100 xpInteractive visualizations50 xpBasic visualization100 xpChange the layout100 xpHighlight nearest nodes and ties100 xpSelect nodes and groups of nodes100 xpCongratulations!50 xp
PrerequisitesIntroduction to the Tidyverse
Professor of Data Science at the University of Udine (Italy)
I am currently interested in data science, complex networks, bibliometrics and generative art. A long time ago I was fond of logic and even before of artificial intelligence. Outside academia, I approached theater and contemporary dance. I love to demystify complex ideas and to cross-fertilize science and art.