Also the data the recommendation is based on is not static and not only relying on the recommendations.
So there are people buying, liking etc. new products come in some are promoted, there are hypes etc. Lots of variables that affect the data that is fed into the recommendation, and then people don't necessarily choose what they are recommended. So it is more like an emergent behavior of a more or less chaotic system. It might end up us all watching Game of Thrones but then that's what you should do :) But as it filters out things you've already did the next recommendation for you will be something new. HTH Michael Am 03.01.2014 um 15:12 schrieb M. David Allen <[email protected]>: > Javad, > > The graph database would probably just store the list of things you could > choose amongst for your recommendations; a graph database doesn't do > recommendations, and hence whether or not those recommendations would > reenforce one another would depend on the kind of algorithm you're using for > doing the recommendation. If it was done incorrectly, yes the problem you're > asking about could occur, but my guess is that it wouldn't have anything to > do with neo4j. > > Can you describe more what you're trying to do? > > > > On Thursday, January 2, 2014 2:50:47 PM UTC-5, Javad Karabi wrote: > When using a graph database for a recommendations engine, would not the > suggested recommendations continue to reenforce themselves, thus kind of > creating a feedback loop? > How does one prevent this from happening? > > -- > You received this message because you are subscribed to the Google Groups > "Neo4j" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > For more options, visit https://groups.google.com/groups/opt_out. -- You received this message because you are subscribed to the Google Groups "Neo4j" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/groups/opt_out.
