Have we mentioned how much we love you lately? We love our customers, because you’re the best. Because you’re so great, we’ve got an awesome offer for you!
We have the best prices in the graph game, but we love you SO MUCH that we’re gonna do you a real solid. 10% off any new instance, any price, any provider, for 12 months. That’s the kind of awesome stuff you get when you subscribe to the GS Newsletter. Only the link in this newsletter will work – accept no imitations.
Offer expires in one week!
John David Martin from Unicon has a cool post on the Neo4j blog about creating a CMS with Neo4j and Elasticsearch. In this case, he demonstrates how he created a tool to provide personally relevant content via text search, utilizing the Page Rank algorithm to score results.
Natural Language Processing Made Easy
Another William Lyon article on pulling data from an API and doing analysis on it caught my eye, this time about grabbing data from Best Buy product reviews and analyzing the text for opinions. William’s article Building a text adjacency graph from product reviews with the Best Buy API shows how you’d grab the data, but then it turns over the subject of doing natural language analysis to a couple other blog posts.
The one I liked the most was Natural Language Analytics made simple and visual with Neo4j by Michael Hunger, which does a great job of demonstrating how you split up natural language into a graph structure, and then do stuff like find the most important phrase of the text.
William also pointed to Max de Marzi’s post Summarize Opinions With A Graph. It’s from a few years ago, but it’s still very relevant, focusing primarily on the concepts used in breaking down natural language into analyze-able graphs.
Eric Spiegelberg has a guest post on the GraphAware blog about internationalization powered by Neo4j. This focuses on using the Spring framework and an implementation of the MessageSource interface that retrieves message definitions from a Cypher-powered graph like Neo4j. I’d love to see examples like this for other stacks.
• • •
Are you doing something cool with graphs at Graph Story? Let us know! We want to talk about what our customers are doing here in the newsletter.
Until next week,
CTO, Graph Story