Taxonomies and Metadata Mapping: An example.

The integration between structured and unstructured content relies in the ability to map context from one to the other. In order to get into some detail, let us use a very simple example of a taxonomy and use this to demonstrate how the mapping would be done. Let’s assume that we have sales and marketing documents and related unstructured content.

What is a Taxonomy, really?

Most companies organize their information into folders and sub folders. In our example this might be:

However, a Taxonomy is more than simply a set of folders - it is really a classification scheme for unstructured content. A classification scheme consists both of the buckets (or folders, sub folders) to put documents in, as well as the means to classify the documents.

This brings up the possibility of alternate taxonomies or different hierarchies to structure your content. The exact same underlying documents might be classified in another way, which is what makes a taxonomy interesting.

Yet another way to classify the documents may be by industry or product line or by sales rep.

The key to building a flexible taxonomy is actually not in the folders, but rather, the means for classifying documents into the taxonomy itself. This capability is what allows us to have different taxonomies for the same underlying data. ShareDox provides you with pre-built taxonomies, which you can customize to your own needs. It enables you to create a taxonomy and provides different views (hierarchies) to let you organize your information around your business.

How does ShareDox allow you to classify your assets?

It is the association of metadata (properties) to digital assets as well as the association between entities in the taxonomy (product line, industry, products etc.) that enable ShareDox to decide how to classify digital assets into the appropriate folders.

The way that ShareDox works is to create a “virtual card” for each of the assets. This virtual card contains embedded information about the asset (format, size, color depth, resolution etc.) and also contains its business metadata properties (taxonomy relevance, title, description, language etc.). In our previous example, a virtual card might have four document properties: Document Type, Customer and Product. The Product field might have a value of “Widget XYZ”; the Document Type field might have a value of “Proposal”, etc. The taxonomy would describe products and that products relate to different industries. So in our example, the document would not be associated with a specific industry, but because it’s associated to a product and the product is associated to an industry, it allows us to organize it into our structure.

Notice that the document does NOT have a direct metadata property for Industry.

The Taxonomy describes the hierarchies and in our example, describes that:

Therefore if we know that a document is related to Product A, we also know that this document is related to Industry X, because Industry X contains Product A.

The taxonomy applies the structure to this information and enables us to then structure the information is ways which make sense for different users. So if we have a folder called "Case Studies" - this simply has a criteria which says: "All documents whose Document Type=Case Study" are included.

By focusing in on the inherent properties (metadata) of the assets, the user can browse ANY of the sample taxonomies that we presented and still get to the underlying documents in a logical and orderly way.