3.1 Installing and Loading R Packages

Code

pacman::p_load(tidyverse, jsonlite, SmartEDA, tidygraph, ggraph)

3.2 Importing Knowledge Graph Data

In the code below, ‘fromJSON()’ of jsonlite package is used to import MC1_graph.json file into R and save the output object.

Code

kg <- fromJSON("data/MC1_graph.json")

3.3 Inspect Structure

Code

str(kg, max.level = 1)

List of 5
 $ directed  : logi TRUE
 $ multigraph: logi TRUE
 $ graph     :List of 2
 $ nodes     :'data.frame': 17412 obs. of  10 variables:
 $ links     :'data.frame': 37857 obs. of  4 variables:

3.4 Extract and Inspect

Code

nodes_tbl <- as_tibble(kg$nodes)
edges_tbl <- as_tibble(kg$links)

3.5 Initial EDA

Code

ggplot(data = edges_tbl,
       aes(y = `Edge Type`)) + 
  geom_bar()

3.6 Creating Knowledge Graph

3.6.1 Mapping from node id to row index

Code

id_map <- tibble(id = nodes_tbl$id,
                 index = seq_len(
                   nrow(nodes_tbl)))

This ensures each id from your node list is mapped to the correct row number.

3.6.2 Map source and target IDs to row indices

Code

edges_tbl <- edges_tbl %>%
  left_join(id_map, by = c("source" = "id")) %>%
  rename(from = index) %>%
  left_join(id_map, by = c("target" = "id")) %>%
  rename(to = index)

3.6.3 Invalid Edges

Code

edges_tbl <- edges_tbl %>%
  filter(!is.na(from),!is.na(to))

3.6.4 Creating the Graph

Code

graph <- tbl_graph(nodes = nodes_tbl,
                   edges = edges_tbl,
                   directed = kg$directed)

3.6.5 Visualising the Knowledge Graph

Code

set.seed(1234)

Code

#| eval: false
ggraph(graph,layout = "fr") +
  geom_edge_link(alpha = 0.3,
                 colour = "gray") +
  geom_node_point(aes(color = `Node Type`),
                  size = 4) +
  geom_node_text(aes(label = name),
                 repel = TRUE,
                 size = 2.5) + 
  theme_void()

Warning: ggrepel: 17411 unlabeled data points (too many overlaps). Consider
increasing max.overlaps

3.7 Cleaning Up

3.7.1 Filter Edges to only “MemberOf”

Code

graph_memberof <- graph %>%
  activate(edges) %>%
  filter(`Edge Type` == "MemberOf")

3.7.2 Extract only connected nodes (i.e., used in these edges)

Code

used_node_indices <- graph_memberof %>%
  activate(edges) %>%
  as_tibble() %>%
  select(from, to) %>%
  unlist() %>%
  unique

3.7.3 Keep only those nodes

Code

graph_memberof <- graph_memberof %>%
  activate(nodes) %>%
  mutate(row_id = row_number()) %>%
  filter(row_id %in% used_node_indices) %>%
  select(-row_id) #optional cleanup

3.7.4 Plot the subgraph

Code

ggraph(graph_memberof,
       layout = "fr") + 
  geom_edge_link(alpha = 0.5,
                 colour = "gray") +
  geom_node_point(aes(color = `Node Type`),
                  size = 1) + 
  geom_node_text(aes(label = name),
                 repel = TRUE,
                 size = 2.5) +
  theme_void()

Warning: ggrepel: 790 unlabeled data points (too many overlaps). Consider
increasing max.overlaps

--- title: "In Class Exercise 3" subtitle: "Mini Challenge 1" format: html date: 05/17/2025 date-format: long date-modified: last-modified editor: visual --- # 3.1 Installing and Loading R Packages ```{r} pacman::p_load(tidyverse, jsonlite, SmartEDA, tidygraph, ggraph) ``` # 3.2 Importing Knowledge Graph Data In the code below, 'fromJSON()' of jsonlite package is used to import MC1_graph.json file into R and save the output object. ```{r} kg <- fromJSON("data/MC1_graph.json") ``` # 3.3 Inspect Structure ```{r} str(kg, max.level = 1) ``` # 3.4 Extract and Inspect ```{r} nodes_tbl <- as_tibble(kg$nodes) edges_tbl <- as_tibble(kg$links) ``` # 3.5 Initial EDA ```{r} ggplot(data = edges_tbl, aes(y = `Edge Type`)) + geom_bar() ``` # 3.6 Creating Knowledge Graph ## 3.6.1 Mapping from node id to row index ```{r} id_map <- tibble(id = nodes_tbl$id, index = seq_len( nrow(nodes_tbl))) ``` This ensures each id from your node list is mapped to the correct row number. ## 3.6.2 Map source and target IDs to row indices ```{r} edges_tbl <- edges_tbl %>% left_join(id_map, by = c("source" = "id")) %>% rename(from = index) %>% left_join(id_map, by = c("target" = "id")) %>% rename(to = index) ``` ## 3.6.3 Invalid Edges ```{r} edges_tbl <- edges_tbl %>% filter(!is.na(from),!is.na(to)) ``` ## 3.6.4 Creating the Graph ```{r} graph <- tbl_graph(nodes = nodes_tbl, edges = edges_tbl, directed = kg$directed) ``` ## 3.6.5 Visualising the Knowledge Graph ```{r} set.seed(1234) ``` ```{r} #| eval: false ggraph(graph,layout = "fr") + geom_edge_link(alpha = 0.3, colour = "gray") + geom_node_point(aes(color = `Node Type`), size = 4) + geom_node_text(aes(label = name), repel = TRUE, size = 2.5) + theme_void() ``` # 3.7 Cleaning Up ## 3.7.1 Filter Edges to only "MemberOf" ```{r} graph_memberof <- graph %>% activate(edges) %>% filter(`Edge Type` == "MemberOf") ``` ## 3.7.2 Extract only connected nodes (i.e., used in these edges) ```{r} used_node_indices <- graph_memberof %>% activate(edges) %>% as_tibble() %>% select(from, to) %>% unlist() %>% unique ``` ## 3.7.3 Keep only those nodes ```{r} graph_memberof <- graph_memberof %>% activate(nodes) %>% mutate(row_id = row_number()) %>% filter(row_id %in% used_node_indices) %>% select(-row_id) #optional cleanup ``` ## 3.7.4 Plot the subgraph ```{r} ggraph(graph_memberof, layout = "fr") + geom_edge_link(alpha = 0.5, colour = "gray") + geom_node_point(aes(color = `Node Type`), size = 1) + geom_node_text(aes(label = name), repel = TRUE, size = 2.5) + theme_void() ```