Search
Sam partitioned data by case and used for speed. No more JOIN explosions. Epilogue: The Conviction Using Neo4j, the agency linked a money trail, phone calls, and meeting locations across 12 suspects. The prosecutor presented a graph visualization—not as evidence, but as an investigation tool. The jury understood instantly.
“Three hops,” Alex whispered. “We can now predict risk chains.” Using collaborative filtering , Sam wrote a query to find people similar to a suspect based on shared locations and contacts:
“We need a faster way to follow relationships,” Alex said. neo4j in action pdf
MATCH (p:Person name: 'Charlie')-[:VISITED|KNOWS]->(common)<-[:VISITED|KNOWS]-(other:Person) WHERE p <> other RETURN other.name, count(common) AS similarity ORDER BY similarity DESC This returned unknown associates—perfect for expanding investigations. The agency integrated Neo4j with Kafka. Every new tip became a new relationship. A trigger query ran every minute:
MATCH (bob:Person name: 'Bob')-[:CALLED]->(phone:Phone) MATCH (phone)<-[:USED]-(suspect:Person)-[:VISITED]->(loc:Location address: 'Main St 42') RETURN suspect.name, phone.number Result: "Charlie" , "555-1234" . Sam partitioned data by case and used for speed
His tech lead, Sam, introduced Neo4j—a where data is stored as nodes (entities) and relationships (connections). Chapter 2: Building the Knowledge Graph Sam modeled their first case:
“It took 2 milliseconds,” Sam said. “And we didn’t even index anything yet.” Alex needed to know: how is Alice connected to a known criminal, Mr. X? “We can now predict risk chains
CREATE (alice:Person name: 'Alice', age: 34) CREATE (bob:Person name: 'Bob', age: 29) CREATE (alice)-[:KNOWS]->(bob) A witness said: “Bob called a phone number, and that phone was used near the crime scene.”