Overview#
Graph Link Analysis transforms investigation data into an interactive visual network, making the hidden relationships between people, organisations, addresses, and digital indicators immediately apparent.
When dozens of entities are spread across separate investigations, the connections between them only become visible when placed together on a shared canvas. A well-laid-out graph makes clusters of associated entities obvious, reveals previously unnoticed bridges between communities, and allows an analyst to trace a path from a known subject to an unknown one in seconds of visual inspection rather than hours of manual cross-referencing. The Graph Link Analysis module renders this as a live, interactive node-edge diagram backed by server-computed and client-side layout algorithms drawn from published academic standards. Analysts can export the full graph in GEXF 1.3 format for further analysis in tools such as Gephi.
Key Features#
-
Force-Directed Graph Layout: Node positions are computed with the Fruchterman-Reingold force-directed algorithm (1991). Repulsive forces push all node pairs apart while attractive forces pull connected pairs together. A geometrically cooled temperature limits displacement each iteration, allowing the layout to converge to a low-energy configuration that places related entities close together and reduces edge crossings.
-
Energy-Minimisation Layout: The Kamada-Kawai algorithm (1989) computes positions by minimising the difference between Euclidean node distances and graph-theoretic shortest-path distances. Spring constants and desired edge lengths are derived from the all-pairs shortest-path matrix. This algorithm produces clean, readable layouts for sparse graphs where geodesic distance between entities is meaningful.
-
Community Detection: The platform identifies clusters of strongly connected entities using the Louvain and Leiden community-detection algorithms, surfacing natural groupings such as fraud rings, shared infrastructure networks, or associated organisations within a larger investigation graph.
-
GEXF 1.3 Export: The complete investigation graph can be exported as a GEXF 1.3 XML file. Each node carries entity type, display name, and colour attributes; each edge carries relationship type and weight. The GEXF format is readable by Gephi and other open graph analysis tools without any proprietary plugins.
-
Scalable Client-Side Rendering: Client-side rendering uses simplified force simulation for moderate-sized graphs. Larger graphs delegate to D3-force (ISC Licence) for performant browser-side simulation that remains interactive at scale.
-
Constraint-Based Layout: The layout engine supports constraint-based arrangements via cola.js (MIT Licence, Monash University), enabling hierarchical grouping, fixed-position anchors, and non-overlap constraints for dense investigation graphs.
-
Graph Statistics: The module provides node count, relationship count, average degree, maximum degree, connected components, isolated nodes, clustering coefficient, and entity-type distributions. Statistics are available as structured data for embedding in briefing packages and for driving analytical overlays on the canvas.
Use Cases#
Financial Crime Investigation#
Lay out persons, companies, accounts, and addresses from a financial crime investigation on a shared canvas to expose shell-company chains, nominee directors, and common bank accounts that are invisible in tabular records.
Counter-Fraud and Organised Crime#
Identify high-degree hub nodes that act as brokers or coordinators within a fraud ring, and trace multi-hop paths between a known offender and previously unconnected subjects.
Cross-Investigation Pattern Discovery#
Place entities from multiple separate investigations on a single canvas to reveal shared infrastructure, common associates, or overlapping networks across otherwise unconnected cases.
Intelligence Briefing Production#
Produce a positioned, publication-ready graph layout at a fixed canvas size suitable for inclusion in intelligence briefings, court-disclosure packages, and senior management summaries.
External Graph Analysis#
Export the investigation graph in GEXF 1.3 format for ingestion into Gephi, NetworkX, or other graph analysis tools for community detection, centrality analysis, or custom visualisation workflows.
Integration#
Graph Link Analysis is accessible through the platform's GraphQL API. Queries retrieve graph data for a given investigation, returning typed node and edge objects ready for rendering or export. Mutations allow nodes and edges to be created, updated, or merged within an investigation context. Real-time subscriptions notify connected clients of graph changes as new evidence or entity links are added, enabling collaborative analyst sessions without manual page refreshes.
The GEXF export endpoint accepts an investigation identifier and returns a standards-compliant XML document that can be saved directly or piped into downstream tooling. All access is scoped to the requesting organisation's data; no cross-tenant graph data is ever returned.
Client applications can perform layout computation locally using the platform's published layout library, updating node positions without a server round-trip for a responsive canvas experience on frequently edited graphs.
Authentication follows the platform's standard OAuth 2.0 and JWT bearer-token flow. Developers integrating a custom front-end or external pipeline use the same credentials and scopes as any other platform API consumer.
Open Standards#
- GEXF 1.3 (Graph Exchange XML Format): The export format used for all graph downloads. Maintained by the GEXF Working Group at https://gexf.net/; open format with no licensing restrictions. Compatible with Gephi, NetworkX, and any GEXF-aware tool.
- Fruchterman-Reingold force-directed placement (Fruchterman & Reingold, 1991, Software, Practice & Experience 21(11)): Published algorithm used for the primary force-directed layout mode.
- Kamada-Kawai energy-minimisation (Kamada & Kawai, 1989, Information Processing Letters 31(1)): Published algorithm used for the energy-minimisation layout mode, particularly suited to sparse graphs.
- Louvain community detection (Blondel et al., 2008, Journal of Statistical Mechanics): Modularity-optimisation algorithm used for identifying entity clusters within the graph.
- Leiden community detection (Traag, Waltman & van Eck, 2019, Scientific Reports): Refined community-detection algorithm used as an alternative to Louvain for graphs requiring guaranteed well-connected communities.
- GraphQL (June 2018 specification, The GraphQL Foundation): Query, mutation, and subscription interface through which all graph data is accessed and modified.
- OAuth 2.0 / JWT (RFC 6749 / RFC 7519): Authorisation and bearer-token standards governing all API access, ensuring tenant-scoped data isolation on every request.
Security and Compliance#
All graph queries are tenant-scoped. An analyst can only retrieve, render, or export entities and relationships that belong to their own organisation. Cross-organisation data is never exposed, regardless of whether two investigations reference the same underlying subject.
Graph size limits are enforced server-side to prevent unbounded computation from affecting platform availability for other tenants. Community detection and layout operations on very large graphs are capped and return a representative subgraph with a clear indication of truncation.
Exported GEXF files contain only the data the requesting user is authorised to view; the export service applies the same authorisation checks as the interactive graph queries.
Last Reviewed: 2026-04-14 / Last Updated: 2026-04-14