With the Mindbreeze InSpire Graph Index, recording, updating, and maintaining metadata attached to content items is quick and easy. The graph index stores “items” and “metadata” while creating easily updatable references between these objects.

The use of references and the graph index has two main significant implications for Mindbreeze InSpire and the indexed data. Firstly, when a change to metadata must be made (for example, a change to the Access Control List (ACL) information), the ACL metadata item is updated in the graph database to generate changes to all necessary documents via their references in just milliseconds.

Secondly, relationships can be inferred, weighted, and understood through these references, allowing for information insight to be gathered and tons of semantic operations (higher-order analytics and logic).

For example, suppose two indexed employees have an author reference to the same indexed document. In that case, Mindbreeze can infer and determine that these two people must have a co-author relationship. This knowledge could enhance the user's search experience. With the graph database, the longer Mindbreeze is active, and the more references and connections created, the more intelligent Mindbreeze becomes.

How is Metadata Indexed?                      

Metadata indexed as structure from the connector or any metadata generated during the index processing pipeline is stored as an information object. The information object is linked to the relevant information objects (e.g. document content or parts of it).

While processing more and more content, every information object is enriched with additional information from multiple sources, again via referencing information. This simply means the more information you feed into Mindbreeze InSpire, the more semantics it learns and reflects and enriches the holistic view of an object. This is a perfect basis to express semantics and derive natural NLP features directly from the graph index itself.

Indexing with other Insight Engines and Products

A federator is a service that wraps another index engine and allows you to use the rest of the Mindbreeze InSpire features. Nevertheless, some features depend on core parts of the index, like the graph representation, which means that some functionality cannot be represented when we use a federator to wrap an indexing service.

If needed, the most relevant documents and information can be pushed forward from any other index source to the Mindbreeze InSpire index – a very popular functionality for sources like Microsoft Exchange.