Epinomy is a search engine, concept manager and autotagger built on top of MarkLogic.

What can you do with Epinomy?

Epinomy is a shopping cart for enterprise information assets.

Everybody knows what a shopping cart is. If you've ever bought something online, you went through a process of searching, refining and selecting a specific asset from among millions of items.  Now, if you think of your enterprise information assets as "products" that people want to "buy", then you understand the purpose behind Epinomy.

What is an "Enterprise Information Asset"?

An enterprise information asset is a piece of information that is owned or produced by an organization.   It might be a document like a contract or purchase order, or it might be a video or PowerPoint presentation or it may be a time series or transaction log.   The goal of Epinomy is to provide a universal index that can store and retrieve all of your enterprise information assets from a central, easy to use interface.

What are Semantics?

Not only can Epinomy behave as a database and search engine, but it also leverages the power of semantics to enhance search.  Broadly speaking, semantics is the study of meaning.  Computer science hasn't really figured out how to tease true meaning from text, because that would imply that we've developed a sentient computer.

Computers are not yet sentient.  

Any organization that has a combination of structured and unstructured data can benefit from Epinomy. Epinomy is based on an exciting platform called MarkLogic. MarkLogic is a search engine, NoSQL database and application server all rolled into one very fast, very scalable architecture.   It provides unique features that allow Epinomy to do some amazing things.

Epinomy blends together three kinds of data:  Tables (structured data), Text (unstructured data) and Triples (semantic data) in a single universal index.

This synergy between three different kinds of data is difficult to achieve because the kind of software the excels at handling unstructured data, like search engines, is very different from the kind of software that handles structured data, like relations database management systems.  The third kind of data, semantic data, is relatively new in the IT landscape and is best handled by a specialized database called a "triple store" or "graph database".

When you think of the kinds of operational data that an organization handles, you will recognize the they all have to deal with both structured and unstructured data.

Examples of Structured Data

  • Balance sheet and accounting data
  • Telematics from machinery
  • Sales figures
  • Performance Indicators
  • Flow Rates
  • Shipping Logs
  • Transactions
  • Quotes
  • Time Series

Examples of Unstructured Data

  • Office Documents
  • PDFs
  • EMail
  • Social Media
  • Comments
  • Reference Guides
  • Business Reports
  • News Feeds
Google+