Terrorist Organizations - Demo Script¶

This demo showcases Aletheia's GraphRAG capabilities using a multi-jurisdiction terrorist organization database containing designations from the US State Department, UK Home Office, and Australia Home Affairs.

Prerequisites¶

1. FalkorDB running: redis-cli ping should return PONG
2. Claude Desktop with MCP servers configured:
3. Graph - Terrorist Orgs: Direct Cypher queries
4. Aletheia - Terrorist Orgs: Semantic search via Graphiti

Part 1: Graph Overview (FalkorDB UI)¶

Open FalkorDB Browser UI at http://localhost:3000

1.1 Graph Statistics¶

// Count nodes by type
MATCH (n)
RETURN labels(n) as type, count(*) as count
ORDER BY count DESC

Expected results:

Talking point: The graph contains ~300 terrorist organizations from three jurisdictions, with rich metadata about aliases, sanctions, and designating authorities.

1.2 Relationship Types¶

// Count relationships by type
MATCH ()-[r]->()
RETURN type(r) as relationship, count(*) as count
ORDER BY count DESC

Expected results:

Talking point: Key relationships include SANCTION (linking orgs to designating authorities) and HAS_ALIAS (tracking alternative names and spellings).

1.3 Visual: Multi-Jurisdiction Organization¶

// Show an organization with multiple designations
MATCH (o:Organization)-[r]->(e)
WHERE o.name CONTAINS 'Hamas'
RETURN o, r, e
LIMIT 20

Talking point: Hamas is designated by US, UK, AND Australia - show how the graph captures cross-jurisdiction designations.

1.4 Visual: Alias Network¶

// Show alias relationships for a complex organization
MATCH (o:Organization)-[:HAS_ALIAS]->(a)
WHERE o.name CONTAINS 'Lashkar'
RETURN o, a

Talking point: Lashkar-e-Taiba has many aliases (LeT, Milli Muslim League, Al-Anfal Trust) - essential for intelligence analysis.

Part 2: Semantic Search with Aletheia MCP (Claude Desktop)¶

Switch to Claude Desktop with "Aletheia - Terrorist Orgs" MCP enabled

Demo Question 1: Cross-Jurisdiction Analysis¶

Question:

"What authorities have designated Hamas as a terrorist organization?"

Expected answer:

US State Department, UK Home Office, and Australia

Capability demonstrated: - Multi-hop traversal: Organization -> Sanction -> Authority - Cross-jurisdiction aggregation: Combines data from three different sources - Entity resolution: Handles "Hamas" variations across jurisdictions

Why this is impressive: Each jurisdiction has different naming conventions and program IDs. Aletheia unifies this into a coherent answer.

Demo Question 2: Organizational Evolution (Multi-hop)¶

Question:

"What is the relationship between AQIM and the Salafist Group for Call and Combat?"

Expected answer:

AQIM is an alias for the Salafist Group for Call and Combat - the Salafist Group evolved into AQIM (al-Qaeda in the Islamic Maghreb)

Capability demonstrated: - Historical name tracking: Understanding organizational evolution - 2-hop reasoning: Organization -> Alias -> Related Organization - Semantic understanding: "relationship between" triggers relationship discovery

Demo Question 3: Geographic Filtering¶

Question:

"What Irish organizations are designated as Foreign Terrorist Organizations?"

Expected answer:

New Irish Republican Army and Continuity Irish Republican Army

Capability demonstrated: - Geographic entity filtering: Finds organizations by regional association - Semantic matching: "Irish" matches organizations with Ireland context - Authority-specific filtering: Focuses on US FTO list

Demo Question 4: Network Tracing (3-hop)¶

Question:

"Trace the evolution: What was AQIM called before, and what network is it part of?"

Expected answer:

AQIM was previously the Salafist Group for Call and Combat (GSPC), and is part of the al-Qaeda network

Capability demonstrated: - 3-hop reasoning: Current name -> Previous name -> Parent network - Temporal evolution: Historical organizational changes - Network affiliation: Understanding al-Qaeda franchise structure

Demo Question 5: Alias-based Discovery¶

Question:

"What organization uses the alias GIA?"

Expected answer:

Armed Islamic Group (Groupe Islamique Arme)

Capability demonstrated: - Reverse alias lookup: From abbreviation to full name - Multi-language support: French alias mapped to English name - Intelligence tradecraft: How analysts actually search for organizations

Part 3: Advanced Queries (FalkorDB)¶

Switch to "Graph - Terrorist Orgs" MCP or FalkorDB UI

3.1 Cross-Jurisdiction Organizations¶

// Find organizations mentioned by multiple authorities
MATCH (o:Organization)
WHERE o.summary CONTAINS 'US State Department'
  AND o.summary CONTAINS 'UK Home Office'
RETURN o.name, o.summary
LIMIT 10

Talking point: Show how to find organizations under multiple jurisdictions using structured queries.

3.2 Authority Analysis¶

// List designating authorities
MATCH (pb:PublicBody)
RETURN pb.name, pb.summary

Expected results: - US State Department - UK Home Office - UK Home Secretary - Attorney-General of Australia - Australia Home Affairs

3.3 Alias Mapping¶

// Find all aliases for a specific organization
MATCH (o:Organization)-[:HAS_ALIAS]->(a)
WHERE o.name CONTAINS 'ISIS'
RETURN o.name, collect(a.name) as aliases

3.4 Network Affiliates¶

// Find al-Qaeda network organizations
MATCH (o:Organization)
WHERE o.name CONTAINS 'Qaeda' OR o.name CONTAINS 'AQIM' OR o.name CONTAINS 'AQAP'
RETURN o.name, o.summary

Part 4: Comparative Analysis¶

Aletheia vs Direct Cypher¶

Natural language (Aletheia):

"What Irish republican organizations are proscribed in the UK?"

Equivalent Cypher (Graph MCP):

MATCH (o:Organization)
WHERE (o.summary CONTAINS 'UK' OR o.summary CONTAINS 'proscribed')
  AND (o.name CONTAINS 'Irish' OR o.name CONTAINS 'IRA'
       OR o.summary CONTAINS 'Irish')
RETURN o.name, o.summary

Talking point: Cypher requires knowing: 1. Which field contains "UK" designation info 2. All possible name variations (Irish, IRA, Republican) 3. How "proscribed" is stored vs "designated"

Aletheia handles semantic variations automatically.

Key Takeaways¶

1. Multi-Jurisdiction Intelligence: Unified view across US, UK, and Australian designations

2. Alias Resolution: Handles abbreviations (PKK, AQIM, LeT), alternative spellings (Hizballah/Hezbollah), and historical names

3. Organizational Evolution: Tracks how groups change names, merge, or affiliate with networks

4. Geographic Analysis: Filter by region without knowing exact country associations

5. Network Mapping: Understand franchise structures (al-Qaeda network, ISIS affiliates)

6. Dual Access:

7. Aletheia MCP: Natural language for analysts
8. Graph MCP: Precise Cypher for developers and structured exports

Appendix: Entity Reference¶

Designating Authorities¶

Sample Organizations by Category¶

Common Aliases¶

Demo Flow Summary¶

Sample Demo Narrative¶

"Let me show you how Aletheia enables natural language queries over complex intelligence data. We have a knowledge graph containing terrorist organization designations from three jurisdictions: US, UK, and Australia.

First, let's see the structure in FalkorDB... [show graph stats]

Now, using Aletheia, I can ask: 'What authorities have designated Hamas?' Without knowing the exact schema or query language, I get a unified answer across all three jurisdictions.

This is powerful for analysts who need to quickly understand cross-border designations without writing complex queries."