I should first point out I have no background in mathematics nor in 
programming, but am teaching myself the basics to be able to oversee a
project, hire a team of developers and begin to plot out the data model, so
apologies if my terminology is not correct. After a lot of research Neo4J
seems the logical choice me and the start-up, mainly because of the
simplicity in which queries can be compiled in cypher, but I would like to
ask the community a question and hope that someone might be able to point
in the right direction for further reading or possibly a foundation from
which to build a query from.
I am creating an application that will take a set of data values about the
user, a set of personal preferences, and a set of data about an item,
ratings on different aspects of the item. What I would like to achieve is
(have also attached a graphic to help with visualising this)
1. A nearest neighbour weighting for their personal preferences matching a
minimum of any 7 out of 10 responses (groups in the graphics).
2. A nearest neighbour based on the scores for the product, 15 different
values (in the graphics I've shown 8), with results being returned biased
by the above personal preferences nearest neighbour weighting
So in the graphical example, I could return all the green relationships
from Group 1 and run a count on them
I have read the following and it seems to be a start but I would like some
advice on where to look next to develop this further
Thanks in advance for any advice and assistance
Dave Clissold 's gravatar image asked Mar 14 2017 at 05:24 in Neo4j by Dave Clissold

0 Answers