Immediacy Taxonomy and Categorisation Manager
User generated tagging or ‘folksonomies’ have grown in
popularity through the growth of the blogosphere and online
services such as Flickr and YouTube.
The following is a useful definition…
“Folksonomy is a democratic approach to information
classification. As with democracy it has it problems but is perhaps
the least messy of all alternatives. The power of the crowd is
almost always better than the power of a few intelligent,
well-meaning people in a hierarchy as the state-controlled systems
of the 20th century have learned.”
It’s easy to see how this can raise a dilemma in a lot of
organisations. Compliance has come to the forefront in recent years
and for certain industry sectors means that ‘controlled systems’
are vital in order to avoid the legal consequences of
non-compliance. In these scenarios, the democratic nature of
tagging would undermine established and recognised International
Standards for records management.
Auto-categorisation capabilities offer the opportunity to bridge
the gap between metadata categorisation and classification
requirements and the ability for the user to add knowledge in the
web content and document creation process.
Most organisations are finding that volumes of electronic
information are multiplying rapidly and valuable knowledge can
become highly fragmented, duplicated and lost. Managing it
effectively becomes a real challenge.
It is widely recognised that the more that information can be
categorised at the point of creation the more relevant and easier
the search and retrieve will be. It is also acknowledged that
Search technologies do not provide all the answers when it comes to
enterprise information management.
We think that the best approach is a combination of Search
Technologies, such as text mining, and user involvement.
The Immediacy Taxonomy and Categorisation Manager (TCM)
provides just this sort of approach. We have developed some
sophisticated algorithms that will extract the keywords, keyphrases
and a summarised description from an item of content, be that a web
page or document. This process also enables the system to
categorise this content automatically against the pre-defined
taxonomy.
Via the TCM interface, users can select content for
categorisation and once the process is complete, will be presented
with the Engine’s interpretation of the content/document’s
keywords, logical category and a summary description. A whole range
of basic metadata, as determined by schemes such as eGMS and Dublin
Core and including items such as title, creation date and author
are also saved automatically as part of this process.
Presented with this automatically created information, the user
then has the option to accept what has been generated or filter and
enhance it further. Slider controls are available in the interface
to widen or narrow the keyword and description creation to make it
more precise if necessary.
The basic essence and power of this capability is that it does
not put all the onus on the user to create metadata, keywords and
descriptions from scratch with each item of content added to the
system. It is much easier to edit and extract than originate this
information and therefore improves the likelihood and willingness
of users to participate in categorising and describing the
information they create and manage.
This, in turn, helps make major Information Management
challenges such as search and retrieval, knowledge capture and
records/retention management considerably
easier and cost-effective.
Content Categorisation
UK local authorities, for example, have been encouraged to use a
mandated taxonomy called Integrated Public Sector Vocabulary (IPSV)
as the basis for categorising and displaying content on their
websites. As part of its Transformational Government initiatives,
the UK Government is keen to ensure that its various agencies
manage their information in consistent ways. Imagine the potential
mess, for example, of each of the 400 UK local authorities
determining their own terms for the services they provide, from
rubbish collection to social services and beyond.
However, one look at IPSV, and the challenge is apparent. It has
in the region of 9,000 terms to describe local authority services.
Authorities are encouraged to adopt the top level terms for each
category but have more flexibility to choose descriptions beneath
this. Flexibility is important when you consider that not all
authorities are the same. A land-locked local authority is not
going to require terms relating to sea defences, coastal erosion
and beaches.
Because Immediacy has a large local authority customer base, the
Immediacy Categorisation Engine has been designed to address these
very real needs in managing and version controlling taxonomies and
helping to categorise information automatically against them.
There are other published taxonomies beyond IPSV, for other
Government agencies such as the Police Service or the Sweet &
Maxwell legal taxonomy, so we have designed ICE as a generic
service. You can also create and manage your own taxonomy or
classification scheme from scratch.