Analytics can detect trends that appear to be normal

Through our work in neonatal intensive care with the University of Toronto, we learned that by applying machine learning to the full data stream, we were able to diagnose some dangerous infections a full day before any symptoms were noticeable to a human being. How did we discover this? By detecting trends that appear normal!

Read more about this ground-breaking and important analytics work in O’Reilly Radar’s Operations, machine learning and premature babies and in the IBM First-of-a-Kind Technology to Help Doctors Care for Premature Babies press release.


New and exciting career opportunities for IT professionals: Data scientist

A data scientist is somebody who is inquisitive, who can stare at data and spot trends. It’s almost like a Renaissance individual who really wants to learn and bring change to an organization.

Does this sound like you? Learn more about the exciting, new data scientist role, and the value that they bring to an enterprise big data initiative, in NetworkWorld’s ‘Big data’ creating big career opportunities for IT pros.

Got questions about big data? My ReadWriteWeb interview has got answers!

In my interview with Scott Fulton from ReadWriteWeb, I cover topics such as:

  • The Data Scientist role
  • Whether big data will alter the enterprise
  • Data silos
  • Big data tools
  • Making data bigger and more consumable

You can see the complete interview in three installments:

“The Data Scientist” on Dataversity

Want more information about The Data Scientist role in organizations with big data? Check out my Dataversity posting:

Welcome to my blog!

In one-on-one discussions across scores of IBM clients, I often see and hear trends about the challenges that businesses face — particularly the IT aspects of enterprises — and develop insights into how these challenges could be addressed to save time and money, improve relationships with customers, and increase the bottom line. Now acting on my desire to share these observations and the resulting insights more broadly, I introduce my new blog! I hope you’ll find these and future thoughts interesting and useful in your pursuit of Big Data nirvana!

Recently when speaking with IBM clients about Big Data, I hear a couple of underlying, recurrent themes. These are not necessarily spoken themes, but they keep “appearing” in the unspoken conversation.

The first theme centers around social media. Determining how your company, brand, and products and services are perceived based on input from the social media universe is an interesting thing, but it’s not the only thing.

The second theme is that there’s no value in unstructured data — or that unstructured data is only interesting in the context of e-mail archiving. This seems to be an especially strong perception among those in the structured data world.

Both of these themes bring to mind the value of the “IT for IT” big data use case — in other words, big data in the context of log analysis. It might not be as sexy as social media analytics, but log analysis provides a wealth of information that you might not be tapping into today due to the seeming enormity and difficulty of it. The beauty of log analysis is that while you might be analyzing logs (like Web logs) for one insight (like customer behavior information), you can leverage those same logs to understand more about your IT infrastructure, like your data center operation. That’s the “IT for IT” connection!

There is an enormous amount of valuable insight currently hidden in logs that contain data about your Web, servers, networked components and equipment, and so on, and you need that insight to make informed decisions about how to manage your operation. It requires robust analytics capability to tap the hidden intelligence in those logs, and IBM has that capability.

To see an example of how IBM InfoSphere BigInsights log analytics works on Web server logs, check out our BigInsights Information Center. The example describes the challenge that many enterprises face of developing new insights into user behavior, preferences, and tendencies, and details the solution using BigInsights to analyze the data in web server logs. (Note that BigInsights can be used for analyzing any type of log. This example demonstrates how log analysis can be performed on web server logs with click stream data. The same approach, however, can be applied to any type of log.)

I look forward to your thoughts on this topic!