Houldsworth Mill,Houldsworth Street, Reddish,SK5 6DAtel: 0161 975 6110 email: recipes@cakesolutions.net JavaEE application development & consultancy Ruby on Rails website development
Content management system (CMS) is a buzzword that many companies use,
but few come to realise its full potential. CMS can be as simple as
a template-based website implemented in PHP or
a complex enterprise knowledge management system. If all you need is a
simple website that shows the latest news and articles, we can help
you by making use of the various open-source content management systems.
This approach is most cost and time efficient.
However, if you need more from your CMS, you will need to consider
a bespoke implementation.
On this page, we will explain some aspects of CMS implementation
you should consider.
One is inclined to think that a system that enforces completeness is a
better solution: it will ensure that you can never reach a dead end (a
broken link, for example).
Let's examine this concept more closely. When you start writing a
document, you cannot know exactly what the final version will contain. Yet,
you can make references very early on, even before you start working the
referenced section. Or, how many times have you said, "Let's skip this,
we'll return to it later"? It is apparent that humans can naturally
handle incomplete information. It is embedded in the very nature of
the way we think.
Creating a system that goes against our way of working will only
reduce productivity and bring frustration to the users.
We believe that systems that can deal with incomplete information allow the users to be more productive and ultimately offer better results than systems that enforce completeness, even though we can only offer anecdotal evidence to support our claim.
Semantic web is a way of representing information that allows machine understanding and learning. Semantic web is an incomplete and conflicting system. We have already explored the advantages of incomplete systems, let's explore conflicting systems. In a conflicting system, we can find pieces of information that contradict each other. We deal with conflicting information every day, therefore, our CMS must be able to deal with the contradictions, too.
Take the morning's weather forecast. How many times
have you listened to the weather forecast on the radio, where the
presenter promised 23 °C and blue skies. Yet, you looked out of the
window and saw dark gray sky and streams of raindrops.
You had two contradicting pieces of information, but you used your
experience and reason to conclude that it's actually rather miserable
outside, despite the reassurances from the BBC.
Therefore, a truly flexible CMS would be able to deal with the consequences of the semantic web and allow directly conflicting information to be uploaded, but will be able to use its context to resolve the truthfulness of each piece of information.
In the final section, we will take a look at the more theoretical concepts of an intelligent CMS.
We will begin with taxonomies, which are part of the semantic web theory.
In short, taxonomy represents the system's common-sense; it classifies
the relationship of the most important terms the system will be dealing with.
Taxonomy allows us to implement machine learning component.
A comprehensive semantic-web CMS should be able to learn from the content the users upload. The key to this machine learning process is the taxonomy and textual analysis. Interested readers can download our paper on taxonomy and machine learning. The underlying concept of machine learning is that the system should be able to analyse the meta data in the content the users upload to it. It should be able to construct a knowledge base containing rules that it can use to automatically classify content with no meta data. This approach calls for a forward-chaining rule engine with complex conflict resolution strategy.
The taxonomy and machine learning components allow us to implement truly intelligent content management system. When the users enter a search term, for example, the system attempts to analyse it in order to "understand" it. This understanding means that it will attempt to match taxonomical terms in the search term.
Intelligent CMS would understand term "university of manchester on tax and market entry in the North West"
not as just a sequence of some characters. Instead, the system would
be able to determine that the users are looking content that mentions a University of Manchester, which is a
type of an institution; tax, which is the taxation subject; business advice on market entry
in the North West region of the UK, which is a country in Western Europe.
In addition to the search accuracy, we can also offer search suggestions. From the
taxonomy, we can find what terms are related to the taxonomical terms
that appeared in the search term.
There is much more to intelligent content management systems. We can help you realise the value of the content you produce on daily basis to help you engage your employees and customers. For more information, see the consultancy services we offer.
What we Do
Team Blogs
Sub-millisecond Java By Jan Machacek on 19th November 2008
Spring User Group November 5 Minutes By Jan Machacek on 16th November 2008
Manchester Wheelers' 125th Anniversary By Jan Machacek on 10th November 2008
Spring User Group Meeting By Jan Machacek on 7th November 2008
Vertically Aligning Images with jQuery By Andrew Chalkley on 30th October 2008
Brr! By Jan Machacek on 29th October 2008
Features