Can we do analytics like recommendations alone with neo4j or do we need python /spark or some other tool to do analytics

neo4j
(Vijaykotikalapudi3) #1

I would like to know , Is there analytical algorithms or recommendations in neo4j or just to store and retrieve graph data.

0 Likes

(Bjoern Oest Hansen) #2

Depending on what you want to do you can do a lot just by querying for patterns. But if I were you I would have a look at the APOC algorithms https://neo4j-contrib.github.io/neo4j-apoc-procedures/ there you can find quite a selection of different algorithms for analyzing your data

0 Likes

(Vijaykotikalapudi3) #3

Hi bjoernoesth,

Thank you very much for you reply.

Why I've asked this question is , we have confusion in using either Neo4j or Azure Cosmos DB.
and we are planning to do some recommendations using graph data.

If Neo4j is only for storing and retrieve graph data, then I'll go with Azure Cosmos DB.
Please suggest, if there is any best feature or which will be more useful for recommendation systems and why

0 Likes

(Bjoern Oest Hansen) #4

Hi , here are a few examples of recommendations systems in Neo4J.

https://neo4j.com/graphgists/?category=real-time-recommendations

https://neo4j.com/graphgist/restaurant-recommendations

I guess it all depends on how you structure your data and how you want to recommend things. Is it a "other customers also bought" kind of recommendation you are after?

0 Likes

(Ameyasoft) #5

Here is another link on recommendation systems:

https://neo4j.com/graphgist/multidimensional-approach-to-recommender-systems

0 Likes