iDating Makes Connections with Graph Databases
repost from Online Personals Watch
What do Snap Interactive, Bang With Friends, Hinge and Meetic all have in common? They all use Neo4j graph database.
Neo4j is an open-source graph database, implemented in Java. The developers describe Neo4j as “embedded, disk-based, fully transactional Java persistence engine that stores data structured in graphs rather than in tables”. Neo4j is the most popular graph database around, and was developed by Neo Technology. – Mark Brooks
What’s a graph database, and why should the iDating industry care?
It’s a new type of database that is optimized for querying connections between people, things, interests, or really anything that can be connected.
The iDating industry cares about interactions and connections. Those two concepts are closely linked. If someone has a connection to another person, through a shared friend or a shared interest, they are much more likely to interact.
How do most iDating databases work in comparison?
Conventional databases, such as MySQL and Oracle, are not designed to help figure out if two people share a connection. They were designed to quickly retrieve specific information about an individual, such as for computing a payroll. Using these type of databases, there is no efficient way to figure out even slightly complex connections.
For certain types of queries, such as friends of friends, a graph database is 1000 times faster than a relational database. This is important for mobile devices, and for building web pages quickly. People will not wait 60 seconds for your query to complete.
What kind of benefits have your clients seen?
More connections, and more interactions.
Interactions are much more likely when there is a connection between you and the person you are looking at. Neo4j enables our customers to model and make sense of the friend data for millions of users. If you find a friend of a friend online, you are three times as likely to connect to that person than to a stranger. For a friend of a friend of a friend, you are twice as likely to connect than if there is no relationship.
Many sites are bound to the second degree, but expanding to the third degree creates an explosion of value. The 3rd degree provides huge volume. People have hundreds of friends, but thousands of FOF and tens of thousands of FOFOFs.
With Neo4j, our customers easily expand to the third degree. It works natively, and with no tricks. Using MySQL, it is difficult to come up with a reasonable solution. With any relational database, there is no efficient way to find 3rd degree connections.
You power BangWithFriends. Do you work with any other dating sites or apps?
Yes, in addition to Down (previously Bang With Friends), we power Snap Interactive, Hinge, Avalable (Singles Solution), Meetic. Additionally, there are some other dating sites that are building new features in Neo4j as we speak. (David Fox, Big Data Engineer at Snap Interactive, recently spoke at GraphConnect conference in New York).