Included this week are several great videos to share with you, so find your headphones and pop some popcorn before settling into another edition of This Week In Neo4j!
This week we have two new data science focused online trainings that are now available, a blog series about finding fraud with Neo4j, a new release of NSMNTX (a Neo4j plugin for working with RDF data), a graph algorithms talk from OSCON, a post about using GraphQL with Neo4j, a look at exploring League of Legends in Neo4j, Cypher index hints, powering data discovery systems with Neo4j, and a look at migrating dimensional models into property graphs.
Before digging in to Twin4j this week, if you’ve used the Neo4j GraphQL integrations or GRANDstack I’d appreciate if you could take a few minutes to fill out this short survey to help guide the future of Neo4j GraphQL.
Thanks and enjoy!
Cheers, William Lyon and the Developer Relations team
Featured Community Member: Andreas Berger
Our featured community members this week is Andreas Berger
Andreas Berger is a freelance software developer from Dresden, Germany. Usually he works for clients in energy, retail and telecommunication. He fell in love with Neo4j about 2 years ago and started to use it successfully in some of his client projects.
Andreas Berger – This Week’s Featured Community Member
Andreas has been actively contributing to a number of open-source projects, for Neo4j those were Spring-Data-Neo4j and Neo4j-OGM. Most recently he contributed a massive amount of work to the neo4j-graphql-java project, improving it in many areas and adding substantial new features. Thanks a lot for your contributions Andreas!
If you want to connect to Andreas, please follow him on Twitter.
New Data Science And Applied Graph Algorithms Online Trainings
The GraphAcademy team has released two new free online courses to help you take advantage of graph based data science and machine learning techniques.
The first course, Data Science with Neo4j shows how to use Neo4j with Python data science tools like Pandas, matplotlib, and scikit-learn, including how to use Neo4j graph algorithms like PageRank and graph based feature extraction for machine learning models.
The second course, Applied Graph Algorithms is designed for the application developer who wants to take advantage of graph algorithms to enhance applications. This course starts with a simple business search web application that uses React, Neo4j, and the Yelp public dataset then shows how to use graph algorithms like PageRank, community detection, and similarity metrics to add personalization features.
Read more about the new courses or check out this brief overview video:
Finding Fraud With Neo4j
Graph champ Max De Marzi wrote a two-part series on finding fraud with Neo4j.
In the first post he walks us through modeling credit card transactions as a graph and how to create a linked list of the transactions. Then he shows us how to use the Cypher variable length path operator and date range filters to zero in on merchants where credit card fraud occurred and all the users impacted. He then shows how to optimize the model for scale and performance. In the second post Max digs into finding fraud rings with Neo4j and the graph algorithms plugin for Neo4j. He uses the union find algorithm to find connected components and group suspicious actors into fraud rings.
Like all of Max’s posts these are filled with lots of Cypher examples so you can see exactly how to apply these techniques.
Connecting the Dots: Graph Databases and Laravel
Keith Damiani presented Connecting the Dots: Graph Databases and Laravel at Laracon US 2019. Keith starts off the talk with a fun graph-based introduction, compares relational databases and graph databases, then hits us with some graph theory. Then after walking through some common graph database use cases and discussing the advantages of using graphs for each he shows how to use Cypher and Neo4j Browser. Finally he introduces the various options for using Neo4j with Laravel including NeoEloquent, the GraphAware PHP client, and a new driver that he’s working on which he used to build Larapals a social network with the goal of helping you meet people that are outside your current network of friends. We’re looking forward to seeing the new driver when Keith releases it!
New Release of NSMNTX For Working With RDF In Neo4j
Jesús Barrasa has released a new version of NSMNTX, a Neo4j plugin that enables the use of RDF (a W3C standard model for data interchange) in Neo4j. NSMNTX makes it possible to store RDF data in Neo4j, export property graph data from Neo4j as RDF, as well as model mapping and inferencing on Neo4j graphs. As part of the release Jesús has also written a very comprehensive user manual for NSMNTX with lots of usage examples. You can also find examples of using NSMNTX on Jesús’ blog
Graph Algorithms: Predict Real-World Behavior
The talk covers graph algorithms, when and why you should use them, how to use them in Neo4j and in the context of a web application powered by Neo4j. Then we also introduce the Graph Algorithms Playground (also known as NEuler) a graph app for Neo4j Desktop that makes it super easy to use graph algorithms without writing any code. If you haven’t tried NEuler you can install it from the Graph Apps Gallery.
For our friends in the Pacific Northwest, see how many Portland memes you can find hidden in the presentation 😉
GraphQL With Neo4j, League of Legends, Data Discovery at Airbnb and Lyft, Cypher Index Hints, and Dimensional Models In Neo4j
- Michael Porter gives us some insight into his experience working with GRANDstack in his post Neo4j & GraphQL A Match Made in Heaven. And no, Michael, the match wasn’t made in heaven but rather in Neo4j Labs 😉
- Jimmy Crequer shows us how to explore League of Legends data in Neo4j, including data import, verifying the data model and querying with Cypher in this post, Playing Around With League of Legends with Neo4j -Prologue. Be sure to follow Jimmy on Medium to catch the next posts where he plans to show how to build a more complex graph and dig into match statistics.
- In the post How LinkedIn, Uber, Lyft, Airbnb and Netflix are Solving Data Management and Discovery for Machine Learning Solutions, Jesus Rodriguez digs into how companies like Airbnb and Lyft are using Neo4j to power data discovery systems as part of their machine learning stacks.
- Luanne Misquitta shows us how to get more performance out of Cypher by hinting which index to use in Cypher: Using Index Hints
- Ever wondered how to deal with dimensional models in Neo4j? Lju Lazarevic shows us how to migrate dimensionally modelled data into Neo4j in this post.
Tweet of the Week
Our tweet of the week is an early review of the new data science online trainings.
First impressions of the new free data science with Neo4j course are very positive. I really appreciate @neo4j investing in making these materials available to the community. https://t.co/0J3KlpnSWA #datascience #python— Nathan Smith (@nsmith_piano) August 23, 2019
Feel free to share your feedback with us on twitter as well, we love hearing from you!