Let me preface by saying that I'm a librarian with some basic web and database skills - not a real programmer. I'm a quick study though, but I don't always know where to begin or how deep I'm going to have to go with a project.
What we have: Our library transactions (check-outs, renewals, as well as research questions and also the book catalog and customer database) are tracked in a MySQL database (via a library-specific product). We have some functions on our website that use PHP to pull stuff from the database, and I'm reasonably competent at manipulating that. We also use Crystal Reports to do some basic stats and reporting.
What I want to do: customer profiling, discovery of trends, and other basic data analytics and data mining.
"Big data" is of course the buzzword going around, but from my reading it seems clear that what we really need is just basic analytics. I've tried doing some research to find a product or service or platform that could be useful. I don’t have much of a budget, though if I prove that I can get some useful stuff going then I might get some budget next year.
So what I’m looking for here is advice about which product or service (if any) I should be looking into. Any advice is appreciated!!
The options I’m looking at:
- http://ubiq.co/ - surprisingly cheap web based reporting for MySQL
- http://infinidb.co/ - would use the open source version, but might have to learn Hadoop? Also might be too “big data” for my needs
- https://spark.apache.org– supposed to be an easier alternative to Hadoop (I’m 90% sure our server runs on Apache – not sure if that’s relevant)
- http://www.cloudera.com/ - seems like a bigger deal than I need at this time
- Learn to use http://en.wikipedia.org/wiki/Apache_Hive on top of Hadoop
- Just get way better at Crystal Reports
- Just use PowerPivot in Excel (after taking a course in it because I suck at basic pivot tables)
- http://watsonanalytics.com/ - sounds like it might be exactly my speed – any experience with it?
TLDR– Idiot noob wants data analytics on MySQL. What do? Halp.
[link][8 comments]