TrustRadius
Neo4j : Good option for small dataset with API support
https://www.trustradius.com/non-relational-databasesNeo4jUnspecified8.524101
No photo available
June 03, 2019

Neo4j : Good option for small dataset with API support

Score 6 out of 101
Vetted Review
Verified User
Review Source

Overall Satisfaction with Neo4j

As [we are] working in the US healthcare domain, we have to deal with large databases like UMLS. It also contains more than 12 relationships among the nodes. To analyze and build a relational module, Neo4j is good choice for us. Mostly Neo4j is used in the R&D department for analysis. But it is also in use at application level to support queries on a large dataset.
  • It's very efficient on large datasets to support the multiple relations between the nodes.
  • Inserting or updating any node or relation is also very easy through the UI or a script
  • Provides very good graphical representation to analyse or present a dataset
  • Very good interactive UI for analysis of any dataset
  • Though the performance is good on a small dataset it requires a well configured server for a large dataset
  • Also graphical representation for less complex dataset is good but for complex dataset in which more than 10 relation possible graphs are not good
  • Also the interactive UI for a complex dataset is little bit complex
  • It adds the ability to quickly load any data set with less time
  • Also we are able to do a POC on different graphs very easily through a good query language; Good for less heavy applications
  • For easy query language and better graphical representation on small dataset
  • Also easy to set up and handle on the server.
  • On top of that Neo4j also provides an API support to interact through any system.
[Based on] Query Language, Performance on small and large data sets, integration and deployment, analysis, API support, Interactive UI.
Neo4j is well suited for POC and analysis purposes on some subsets of large data sets; also very efficient query language to query the knowledge. But if you have to deal with a large and complex data set, it's not a good option.