Diving deeper than metadata, down to 'contextual' metadata

bridgwatera | No Comments
| More

Content management gets a bad press.

Allow me to explain -- what I mean is that content management and document management often suffers from a rather unsexy, not as exciting as it could be press.

After all, this is the realm of banks and lawyers and industry verticals that operate around an administratively heavy (electronic) paper-based existence, right?

But there is a fascinating element to this subject, albeit slightly hidden though it may be. This is because content management has changed.

For a start, we now exist in the era of social content management. This is where social web 2.0 collaboration tools (or social business tools if you prefer) are being used to exchange and edit content in increasingly dynamic ways.

This reality, if you buy this 'social business'-themed argument, means that so-called 'systems of record' inside organisations are now working with (and being complemented by) systems of engagement.

In other words, accounts systems and other corporate applications are being impacted by electronic user collaboration in the form of email and more.

Not that this in itself is new, but there has been a progression here -- and systems of record themselves now face the additional challenge of not only tracking a firm's own processes, but also accommodating for what Forrester Research defines as "out of process" applications from third parties or those that only happen infrequently.

What all this means is that documents need to be more concisely defined with additional metadata information. Knowing who created a document, when and where it was created as well as how long it is in terms of characters and what other embedded elements it might contain is no longer enough.

Used by business analysts and typically tuned and tweaked by software developers to suit a particular business case, tools such as IBM Content Analytics and Cognos Consumer Insight are being used to gain the extra depth needed here.

This is not just analytics; this is 'contextual' analytics.

IBM cloud.jpg

Contextual analytics is carried out via the use of "annotators" which help to define the content management metatags that are needed at this new deeper level. Deeper down below "basic" metadata we find:

• Information on document quality and accuracy
• Information relating to document value (from a commercial perspective)
• Information relating to a document's time sensitivity (which could be used in a 'time series' visualisation dashboard view perhaps)
• ... and even, information related to "sentiment" depending on the tone of language used i.e. angry, positive, critical etc. (which could be used to check on customer churn perhaps).

Ken Bisconti, VP of ECM products at IBM points to IBM Content Analytics as an advanced search and analytics platform. According to IBM, "Content analytics solutions can understand the meaning and context of human language and rapidly process information to improve knowledge-driven search and surface new insights from your enterprise content. Content Analytics uses the same Natural Language Processing (NLP) technologies as IBM Watson DeepQA, the world's most advanced question-answering machine."

Although IBM is not the only company operating in this space, this story was generated in response to looking at IBM Cognos Consumer Insight -- a product designed to enable marketing professionals to "listen to" customer demands and opinions expressed through social media by analysing large volumes of publicly available Internet content.

So, next time you complain about an airline or some other service that you have purchased on Twitter or Facebook, now you know how companies might be listening to you.

Document management then -- dull and dreary, or quite interesting and rapidly coming into the social business sphere with deeper more intelligent contextual information?

Leave a comment

About this Entry

This page contains a single entry by Adrian Bridgwater published on January 17, 2012 8:46 AM.

Nobody ever got fired for collaborating with IBM, yet was the previous entry in this blog.

Our software development future: "probabilistic" applications is the next entry in this blog.

Find recent content on the main index or look in the archives to find all content.