Dynamic segregation of nodes based on text matching

[sailibakalkar]

In my graph database I have multiple address nodes I want them to be segregated on the basis on text matching function.
If the address has the same street or building just the room no. is different those addresses should form a cluster or there own community. Similarly all the address nodes in my graph databases should get segregated based on such similar text matching. If i have 1000 address they should get bifurcated automatically on the basis of same building name or area. If that possible using cypher query? I tried the text similarity functions of Neo4j but they require some input to match phrases from db...can this happen dynamically?

[david_allen]

You might want to consider using an APOC trigger for this, consult the APOC documentation. The idea would be, whenever a node gets written to the database, you apply your text matching function, and then you link that node via a relationship to whichever "cluster" node you want to assign for it.