Data Intelligence & Network Visualisation Platform

Raw social data → embeddings → communities, explored as an interactive influence graph.

The problem

Relationship analysis across large unstructured, scraped datasets was manual.

The system, drawn

Hover a node to trace its links. Size = influence, colour = community.

Fig. 03Unstructured social data → vector embeddings → communities and influenceSimplified redrawing · illustrative data · real mechanism

How it works · 3 parts

01
ML pipeline
Vector embeddings and semantic clustering producing structured cluster and influence outputs.
02
Graph visualisation
Entities as nodes, relationships as edges; node size encodes influence, colour encodes community; click-through exploration.
03
Integration
Reconciles heterogeneous sources with mismatched identifiers via fallback mapping and synthetic datasets; modular repos so new sources drop in.

Raw social data → embeddings → communities, explored as an interactive influence graph.

The problem

Relationship analysis across large unstructured, scraped datasets was manual.

The system, drawn

Hover a node to trace its links. Size = influence, colour = community.

Fig. 03Unstructured social data → vector embeddings → communities and influenceSimplified redrawing · illustrative data · real mechanism

01
ML pipeline
Vector embeddings and semantic clustering producing structured cluster and influence outputs.
02
Graph visualisation
Entities as nodes, relationships as edges; node size encodes influence, colour encodes community; click-through exploration.
03
Integration
Reconciles heterogeneous sources with mismatched identifiers via fallback mapping and synthetic datasets; modular repos so new sources drop in.