I'm trying to analyze and find some patterns in the browsing behavior of our site's userbase. To go about this a graph database seemed like a good choice compared to a relational database since there are hundreds of millions of records. I extracted our our User Information, their session id's, activity_uuid and page name into nodes with 4 labels (users, session, activity and page). The relationship that I have formed is as follows: (User)-[:Browsers]->(Session)-[:Click]->(Activity)-[:View]->(Page)
and the activity node is related to itself with the [:Next] relationship so that we can have a sequence of events in a browsing session.The Page information is also present as one of the properties in the Activity node along with timestamp.
Since I'm new to Cypher I'm wondering how do I go about writing query for finding out browsing pattern of users who bought something (page=Order Conformation) vs those who didn't or new users to the site (property in the session node) vs old and difference in their browsing behavior.
