Hi all,
Thanks for keeping up the active community. I have already found numerous helpful replies scatted through this Discourse.
I am Brendan, I work for a digital marketing company in South Africa and live in small town Canada. I have years of experience building applications in .Net and leveraging MS SQL server, so Graph DBs are a nice change.
I am hoping to use a graph database to solve a "Single Customer View" type problem that we have in our business. I think I have a solution, but I am battling with JAVA and Neo4j using up 100% of CPU and RAM on my dev machine.
Hoping to get assistance on what the best route would be to get a Proof of concept built before scaling out to the handle the live data.
@tard_gabriel sorry, I didn't realize it was so domain specific.
We have a large number of "leads". These are people who have submitted their information to express interest in product. These leads have numerous data points like name, contact info and personal demographics.
A person (e.g. Gabriel) can submit their details more than once for different products on different source platforms. Each of their submissions may be slightly different because of missing info or typeos, etc.
I am looking at using a graph to "plot" all leads and all properties and then sum the weighted connections. Cell number has a higher weight than firstname. If I see that Gabriel connects to Brendan with a total weight of more than X, I will assume that they are the same person (customer).
