Unstructured and Structured Data
- Elaine Foley
The Picturepark Content Platform supports different kinds of metadata concepts, one is unstructured data input like free text and the other is structured data input similar to excel tables.
Unstructured Data
You can define for each and every content all required information as free text, also putting all information in one single field if you like. An Image itself is also unstructured information as you have to look at it to see what it is about.
👍Pro: simple setup
👎Con: a data mess (wrong spellings, typos, a nightmare when you change a product name and many many more
Structured Data
You can use structured data input in the form of tags which is common in popular systems your users work with. For images, this means adding AI Tagging in Picturepark to get keywords for images and video, which you give context and additional information.
👍Pro: prepared for AI and automation (computers love structured data), data consistency, easy quality control
👎Con: you must define your content model
Picturepark also offers Business Rules that help you in structuring your unstructured data by parsing filenames, parsing embedded metadata information or autotagging capabilities.
Schema.org
A schema is a way to determine the structure, content, and to some extent, the semantics of any content - with regard to metadata: a form where data is entered. See metadata schemas: DC, IPTC, XMP and many more. Other schemas are also well known in database design, on the web (open graph) and others. Different schemas as per schema.org
Generic Item: https://schema.org/Thing
Vehicle: https://schema.org/Vehicle
Individual Product: https://schema.org/IndividualProduct
Product: https://schema.org/Product
Person: https://schema.org/Person
Accommodation: https://schema.org/Accommodation
Event: https://schema.org/Event
Website: https://schema.org/WebSite
Checkout page: https://schema.org/CheckoutPage
News Article: https://schema.org/NewsArticle