changed bbox of network, added variable in json data to set label threshold, changed to British spelling convention

README.md

@@ -1,22 +1,22 @@
-# GESARA Named Entity Network Visualization
+# GESARA Named Entity Network Visualisation
 
-This project generates an [interactive visualization](https://bellingcat.github.io/gesara-entity-viz/) of [named entities](https://spacy.io/usage/linguistic-features#named-entities) in English-language posts archived in a database of Telegram channels that have posted about the GESARA conspiracy theory.
+This project generates an [interactive visualisation](https://bellingcat.github.io/gesara-entity-viz/) of [named entities](https://spacy.io/usage/linguistic-features#named-entities) in English-language posts archived in a database of Telegram channels that have posted about the GESARA conspiracy theory.
 
-This visualization was developed by Bellingcat based on an excellent [Sigma.js demo](https://github.com/jacomyal/sigma.js/tree/main/demo), and uses [react-sigma-v2](https://github.com/sim51/react-sigma-v2) to interface sigma.js with React.
+This visualisation was developed by Bellingcat based on an excellent [Sigma.js demo](https://github.com/jacomyal/sigma.js/tree/main/demo), and uses [react-sigma-v2](https://github.com/sim51/react-sigma-v2) to interface sigma.js with React.
 
-You can view the live visualization [here](https://bellingcat.github.io/gesara-entity-viz/). With GitHub pages configured, after making changes to the `main` branch, you need th run the command `npm run deploy` for the latest changes to be reflected in the live visualization.
+You can view the live visualisation [here](https://bellingcat.github.io/gesara-entity-viz/). With GitHub pages configured, after making changes to the `main` branch, you need to run the command `npm run deploy` for the latest changes to be reflected in the live visualisation.
 
 ## Python Scripts
 
-In the `scripts/` subdirectory, you can run Python scripts that were used to generate the network and visualization:
+In the `scripts/` subdirectory, you can run Python scripts that were used to generate the network and visualisation:
 
 ### `generate_network.py`
 
 Extracts the data from a PostgreSQL database, cleans the entity data, generates a NetworkX graph, prunes the edges using the [Marginal Likelihood Filter](https://github.com/naviddianati/GraphPruning), and exports the pruned graph.
 
-### `generate_visualization.py`
+### `generate_visualisation.py`
 
-After visualizing the network using [Gephi](https://gephi.org/) (using the Force Atlas 2 algorithm, with the "LinLog mode" and "Prevent Overlap" options enabled, and exporting as the file `entity_network_layout.graphml`), this script converts the node, edge, and cluster data into a format readable by this sigma.js project.
+After visualising the network using [Gephi](https://gephi.org/) (using the Force Atlas 2 algorithm, with the "LinLog mode" and "Prevent Overlap" options enabled, and exporting as the file `entity_network_layout.graphml`), this script converts the node, edge, and cluster data into a format readable by this sigma.js project.
 
 ## NPM Scripts
 

scripts/generate_visualisation.py

@@ -12,7 +12,7 @@ COLORS = colorcet.glasbey_dark
 
 OUTPUT_JSON = "../public/dataset_entities.json"
 
-NODE_SCALING = 0.5
+NODE_SCALING = 0.35
 # GraphML file generated by Gephi
 INPUT_GRAPHML = "data/entity_network_layout.graphml"
 CLUSTERS = [
@@ -46,7 +46,9 @@ CLUSTERS = [
     {"key": "37", "clusterLabel": "Payment platforms"},
     {"key": "42", "clusterLabel": "Vote audit"},
 ]
-BOUNDING_BOX = {"x": [-300, 400], "y": [-600, 150]}
+BOUNDING_BOX = {"x": [100, 200], "y": [-370,-50]}
+
+LABEL_THRESHOLD = 15
 
 
 if __name__ == "__main__":
@@ -84,6 +86,7 @@ if __name__ == "__main__":
         ]
         + [{"key": "100", "clusterLabel": "Other", "color": "#999999"}],
         "bbox": BOUNDING_BOX,
+        'labelThreshold': LABEL_THRESHOLD
     }
 
     with open(OUTPUT_JSON, "w") as f:

src/types.ts

@@ -19,7 +19,8 @@ export interface Dataset {
   nodes: NodeData[];
   edges: [string, string][];
   clusters: Cluster[];
-  bbox: {'x': Extent, 'y': Extent}
+  bbox: {'x': Extent, 'y': Extent},
+  labelThreshold: number
 }
 
 export interface FiltersState {

src/views/DescriptionPanel.tsx

@@ -14,18 +14,17 @@ const DescriptionPanel: FC = () => {
       }
     >
       <p>
-        This visualisation represents a <i>network</i> of{" "}
+        This interactive visualisation represents a <i>network</i> of{" "}
         <a target="_blank" rel="noreferrer" href="https://spacy.io/usage/linguistic-features#named-entities">
           named entities
         </a> in English-language posts archived in a database of Telegram channels that have posted about the GESARA conspiracy theory. Each{" "}
-        <i>node</i> represents an entity, <i>edges</i> between nodes indicate that one or more posts contain both entities
+        <i>node</i> represents an entity, <i>edges</i> between nodes indicate that one or more posts contain both entities.
-        .
       </p>
       <p>
-        This kind of visualization shows the ecosystem of the people, organizations, and ideas these conspiracy Telegram channels talk about, as well as the connections between them.
+        This kind of visualisation shows the ecosystem of the people, organisations, and ideas these conspiracy Telegram channels talk about, as well as the connections between them.
       </p>
       <p>
-        Some social media channels were identified by researchers from{" "}
+        Some Telegram channels were identified by researchers from{" "}
         <a target="_blank" rel="noreferrer" href="https://www.bellingcat.com/">
           Bellingcat
         </a>{" "}and{" "}
@@ -59,14 +58,13 @@ const DescriptionPanel: FC = () => {
         .
       </p>
       <p>
-        The network was visualized using{" "}
+        The network was visualised using{" "}
         <a target="_blank" rel="noreferrer" href="https://gephi.org/">
         Gephi
-        </a>. Node sizes are related to the number of channels the entity was posted about in the database.
+        </a>. The radius of each node is proportional to the number of channels in the database whose posts mention the entity.         Nodes are coloured based on a{" "}
-        Nodes are colored based a{" "}
         <a target="_blank" rel="noreferrer" href="https://arxiv.org/abs/0803.0476">
         community detection algorithm
-        </a>.
+        </a>. Edges are weighted by the number of posts that mention both entities.
         For visualisation purposes, edges were pruned using the{" "}
         <a target="_blank" rel="noreferrer" href="https://github.com/naviddianati/GraphPruning">
           Marginal Likelihood Filter

src/views/Root.tsx

@@ -53,7 +53,7 @@ const Root: FC = () => {
           defaultNodeType: "image",
           labelDensity: 0.07,
           labelGridCellSize: 60,
-          labelRenderedSizeThreshold: 10,
+          labelRenderedSizeThreshold: dataset.labelThreshold,
           labelFont: "Lato, sans-serif",
           zIndex: true,
         }}