IS428 Assignment 2 - Viz Makeover

Chan Huang Suan

16th November 2019

1. Administrative: Install Package

packages = c('igraph', 'tidygraph', 'ggraph', 'ggrepel', 'visNetwork', 'lubridate', 'tidyverse')

for(p in packages){library
  if(!require(p, character.only = T)){
    install.packages(p)
  }
  library(p, character.only = T)
}

2. Data Management

2.1 Data Wrangling

GAStech_nodes <- read_csv("data/GAStech_email_node.csv")
GAStech_edges <- read_csv("data/GAStech_email_edge-v2.csv")

2.2 Wrangling Data

GAStech_edges$SentDate  = dmy(GAStech_edges$SentDate)
GAStech_edges$Weekday = wday(GAStech_edges$SentDate, label = TRUE, abbr = FALSE)

GAStech_edges_aggregated <- GAStech_edges %>%
  filter(MainSubject == "Work related") %>%
  group_by(source, target, Weekday) %>%
    summarise(Weight = n()) %>%
  filter(source!=target) %>% 
  #remove email that is sent to self
  filter(Weight > 1) %>%
  ungroup()

2.3 Creating network objects using tidygraph

GAStech_graph <- tbl_graph(nodes = GAStech_nodes, edges = GAStech_edges_aggregated, directed = TRUE)
GAStech_graph

## # A tbl_graph: 54 nodes and 1456 edges
## #
## # A directed multigraph with 1 component
## #
## # Node Data: 54 x 4 (active)
##      id label              Department    Title                             
##   <dbl> <chr>              <chr>         <chr>                             
## 1     1 Mat.Bramar         Administrati~ Assistant to CEO                  
## 2     2 Anda.Ribera        Administrati~ Assistant to CFO                  
## 3     3 Rachel.Pantanal    Administrati~ Assistant to CIO                  
## 4     4 Linda.Lagos        Administrati~ Assistant to COO                  
## 5     5 Ruscella.Mies.Hab~ Administrati~ Assistant to Engineering Group Ma~
## 6     6 Carla.Forluniau    Administrati~ Assistant to IT Group Manager     
## # ... with 48 more rows
## #
## # Edge Data: 1,456 x 4
##    from    to Weekday   Weight
##   <int> <int> <ord>      <int>
## 1     1     2 Monday         4
## 2     1     2 Tuesday        3
## 3     1     2 Wednesday      5
## # ... with 1,453 more rows

Task 1 - Static Organization Graph

2. Network Metrics Analysis

GAStech_graph %>%
  activate(edges) %>%
  arrange(desc(Weight))

## # A tbl_graph: 54 nodes and 1456 edges
## #
## # A directed multigraph with 1 component
## #
## # Edge Data: 1,456 x 4 (active)
##    from    to Weekday Weight
##   <int> <int> <ord>    <int>
## 1    40    41 Tuesday     23
## 2    40    43 Tuesday     19
## 3    41    43 Tuesday     15
## 4    41    40 Tuesday     14
## 5    42    41 Tuesday     13
## 6    42    40 Tuesday     12
## # ... with 1,450 more rows
## #
## # Node Data: 54 x 4
##      id label           Department     Title           
##   <dbl> <chr>           <chr>          <chr>           
## 1     1 Mat.Bramar      Administration Assistant to CEO
## 2     2 Anda.Ribera     Administration Assistant to CFO
## 3     3 Rachel.Pantanal Administration Assistant to CIO
## # ... with 51 more rows

2.1 Computing centrality indices (Original)

g <- GAStech_graph %>%
  mutate(betweenness_centrality = centrality_betweenness()) %>%
  mutate(closeness_centrality = centrality_closeness()) %>%
  ggraph(layout = "nicely") + 
  geom_edge_link(aes()) +
  geom_node_point(aes(colour = closeness_centrality, size=betweenness_centrality))

g + theme_graph()

2.2 Computing Centrality Indices (Improved)

g <- GAStech_graph %>%
  mutate(betweenness_centrality = centrality_betweenness()) %>%
  mutate(closeness_centrality = centrality_closeness()) %>%
  ggraph(layout = "nicely") + 
    geom_edge_density(aes(fill = Weight)) +
    geom_edge_link(aes(width = Weight), alpha = 0.2) + 
    geom_node_point(aes(color = closeness_centrality, size = betweenness_centrality)) +
    scale_color_viridis()

g + theme_graph() + facet_edges(~Weekday)

Questions: 3 Aspects of the Graph that can be improved.

Analysing relationship between pairs of vertexes is made difficult due to the opacity of the graph’s edge.
The dimension ‘Weekday’ could be use to help facet the graph to show interaction between the subjects for each weekday.
All edges are present on the graph. It would help the user to visualize important edges by their weight that is assigned. This helps user to identify important relationships that would be crucial to identify.

3. Building Interactive Network Graph with visNetwork

3.1 Data preparation

GAStech_edges_aggregated <- GAStech_edges %>%
  left_join(GAStech_nodes, by = c("sourceLabel" = "label")) %>%
  rename(from = id) %>%
  left_join(GAStech_nodes, by = c("targetLabel" = "label")) %>%
  rename(to = id) %>%
  filter(MainSubject == "Work related") %>%
  group_by(from, to) %>%
    summarise(weight = n()) %>%
  filter(from!=to) %>%
  filter(weight > 1) %>%
  ungroup()

3.2 Assinging groups to data

GAStech_nodes <- GAStech_nodes %>%
  rename(group = Department)

3.3 Interactivity Graph (Original)

visNetwork(GAStech_nodes, GAStech_edges_aggregated) %>%
  visIgraphLayout(layout = "layout_with_fr") %>%
  visOptions(highlightNearest = TRUE, nodesIdSelection = TRUE)

Questions: 3 Aspects of the Graph that can be improved.

Vertexes do not show the in-degree or out-degree relationship with other vertexes.
Colored edges do not add value to the analysis of the graph. It is a distraction to view the network when one vertex is highlighted.
Names shown under the shape of the vertex, makes it difficult to read and identify if there is any edges overlapping over the text.

3.4 Interactive Graph (Amended)

#Calculate Out-Degree values for each node
graph <- graph.data.frame(GAStech_edges_aggregated, directed = T)
degree_value <- degree(graph, mode = "out")
GAStech_nodes$value <- degree_value[match(GAStech_nodes$id, names(degree_value))]

visNetwork(GAStech_nodes, GAStech_edges_aggregated, main = "Interactive Graph - Outward interaction") %>%
  # Optimization
  visIgraphLayout(layout = "layout_with_fr") %>%
  visEdges(smooth = FALSE) %>%
  visPhysics(stabilization = FALSE) %>%
  # Customization
  visInteraction(dragNodes = TRUE, dragView = TRUE,zoomView = TRUE) %>%
  visNodes(shape = "box" ,shadow = list(enabled = TRUE, size = 10)) %>%
  visEdges(arrows = "to" ,shadow = FALSE,color = list(highlight = "#C62F4B")) %>%
  visOptions(highlightNearest = list(enabled = TRUE, hover = FALSE, algorithm = "hierarchical")
             , nodesIdSelection = list(enabled = TRUE, selected = "1", values = unique(GAStech_nodes$id)))