lundi 14 avril 2014

A 360-degree view is the future of Big Data

A lire sur:

When Big Data analytics hits big, true leaders will be synthesizing a complete view of the enterprise and their marketplace.
By Mike Shaw, HP Software Strategic Marketing
While the use of Big Data is not yet mainstream, it has certainly moved out of the bleeding-edge, science project phase. There are now examples of its use for the collection and analysis of lots of structured data. And there are examples of its use for the collection and analysis of human interaction data, too—voice, video, pictures, social media interactions, and so on.
Mike Shaw, HP Software Strategic Marketing 
Mike Shaw
But there are only a few examples of both structured and human interaction data being used at the same time.
Over the last two years, a team in HP has been working on a vision for the enterprise of 2020. This work has resulted in the publication of a series of chapters on topics like the CIO of 2020applications in 2020, and the data center of 2020.
The 2020 team is now working on a chapter on Big Data in 2020, and one of the main predictions is that we will routinely model and analyze both structured and human interaction data at the same time. This will give us a 360-degree view, and it will mean that we have no blind spots in our analysis.

Data in 360 degreesLet’s look at some examples of “360-degree views” in action. Today, retailers like Guess collect and analyze the structured data from sales, and use this to determine what is trending up (and thus what to promote), what is not selling well (and thus what may need to be discounted), and, importantly, what is selling well with what— “affinity” analysis, as data scientists call it.
This is great. It’s a huge step forward for retailers, because it allows them to more optimally market to customers. But structured data only tells you about what youare selling: it’s the human interaction data from social media like Twitter that can tell you what a competitor is selling that you aren’t. Only social media can alert you to the fact that a celebrity is wearing something you sell, and you need to get stock in!
The latest generation of IT management systems takes all log, alert, performance, and service hierarchy data and uses it to better solve problems with today’s complex IT systems. Just as in retail, this is a huge step forward.
In the future, however, these IT operations analytics solutions will also use human interaction data. They will derive meaning from the interaction between the testing team and the app dev team—"This is a defect." "No, it’s a feature."—between the service desk and application maintenance, and between the service desk and the customer. As a former quality manager, I know that these interactions often contain information that can help in quick diagnoses of customer problems.
At the recent HP Discover conference in Barcelona, HP Labs demonstrated a prototype of a Big Data analysis modeling system code-named Project Titan. The project had the subtitle "analysis by the people." Data scientists are expensive, and in increasingly short supply. The project therefore created a modeling system that could be used by domain experts in the creation of 360-degree Big Data analyses: they choose their sources; they choose the most appropriate algorithm, given the data sources and the reason for the analysis; and they "play" with the visualization of the results that will best suit the intended user of the analysis.
We believe that by 2020 such modeling will be routine: domain experts will regularly choose structured and human interaction data sources for a 360-degree view with no blind spots.

Aucun commentaire:

Enregistrer un commentaire