As businesses continue to be inundated with data – coming in from various sources – and as we all continue to harp on about AI and how it is going to change the world, I find myself thinking about the role of metadata in all of this – so much so that I even addressed the topic with my colleagues recently.
I know I talk about data often (that’s what keeps you coming back to my blog right ) but there is ‘method in my madness’, especially given how data actually enhances the ability to truly reap the benefits of all these great technological transformations. And metadata is no different.
While we know it to be there, how often do we actually think about the metadata driving results behind the data we are using for the purpose of being able to benefit from technologies such as AI? I came across an article that provides such solid insight into this and couldn’t resist sharing my views, as the focus placed in this industry piece on metadata and the management thereof is something I believe can be used across industries.
Using content providers as an example (TV and video service providers), the author of this piece highlights how metadata can be used by such providers to drive better AI. With so much content available to the user to watch (and subsequently a very competitive environment for these service providers), AI based on solid metadata can support these providers in supplying more intelligent recommendations – ones that fit the end user better based on their preferences. The outcome? It is likely to be a happier and retained user who in my view would be grateful that they don’t have to think or search very hard about what to watch next as recommendations and the ability to search are already there for selection.
Managing metadata for value
Of course, if the metadata being used for this is not managed effectively, the results won’t be that intelligent. Therefore, with so much data to manage, data foundations once again become a key focus. If data from varying sources is not sorted and merged accurately, the data load will only become untidy, difficult to control and therefore impractical for AI implementation.
Digging deeper for further insights
Metadata that is managed efficiently on the other hand can in turn be analysed further to become richer and greatly insightful. Imagine the ability for the service provider to segment the content per user even further and provide the ability for the user to undertake an advanced search to produce even more accurate results, timely.
I believe that applying this type of focus around metadata in general (and not just for the purpose of better AI), can support businesses in truly managing their data more effectively. One area that comes to mind would be applying this thinking and methodology to unstructured data, to get a better handle on it (think social media and emails), for the purpose of extracting deeper and more meaningful insights that an organisation can actually then use. The same would apply for traffic management – just think how utilising metadata in traffic management systems can produce more accurate results to implement changes that can improve these systems. And these are just two quick examples.
What we can learn from the AI hype is the innate focus it is bringing on data and everything data related – something that perhaps has been overlooked as technology continues to evolve. I stand strong in my belief that at the core of most effective technology implementations lies data – but not just any data – quality driven information. And this is directly enabled and managed by good up-to-date and accurate metadata.
