A Noisy Algorithm

Here is something I created while having sea food at Pier 39 in San Fransisco-

Creating an algorithm for distorting predictive models by generating random noise ( either amplified or reduced sample).


“If you can not convince them, confuse them”

  1. Generating white noise like signals to fake and distort noise and signal ratios
  2. Aggressive merger and acquisitions negotiations
  3. Media and Entertainment _                                     (Create Marketing Buzz/ Tabloid /Hype/ Fear , Uncertainty Doubt)
  4. National Security -( Kill _all_ the Terrorists with Love –                        black,brown,yellow,olive,white,blue,red …)
  5. Dating                                                                 (as in u2’s sweetest thing- Brown Eyed Boy meets Blue Eyed Girl)

The 0 1-1 1R 1 Algorithm

  1. Define Initial Position (i.e Use 6 sigma Define step)
  2. Take ANY Step 1 (i.e take a walk, make a phone call)
  3. Repeat ANY Step 1 again
  4. Do ANY Step 2 which is an opposite to ANY Step 1 in directional and /or  magnitude ( maybe time, or x,y,z and T ) vector to Any Step 1
  5. Return to Initial Position
  6. Loop the above 5 steps R times.

A detailed work flow would be followed by a simple diagram.

An earlier attempt to mash creativity with science as far back as July 2008 was the now redundant Ohri Framework

at https://decisionstats.wordpress.com/?s=ohri+framework (note WordPress timestamps can be manipulated so Google cache remains the true source of time series analysis of posts except when affected by black hat SEO )

New Edition of SAS.com Magazine q 1 2010

As always a great edition of an excellent online magazine.

The cover story of GE on stopping service fraud is great ( I am an ex GE alumnus- DIS claimer)

Click the screenshot for the real thing itself.

As my friends used to say, a magazine is something that can shoot multiple times.

Cloud MapReduce

Apparently claimed to be much faster than Hadoop, here is the cloud OS flavor for MapReduce.


Cloud MapReduce was initially developed at Accenture Technology Labs. It is a MapReduce implementation on top of the Amazon Cloud OS.

By exploiting a cloud OS’s scalability, Cloud MapReduce achieves three primary advantages over other MapReduce implementations built on a traditional OS:

  • It is faster than other implementations (e.g., 60 times faster than Hadoop in one case. Speedup depends on the application and data.).
  • It is more scalable and more failure resistant because it has no single point of bottleneck.
  • It is dramatically simpler with only 3,000 lines of code (e.g., two orders of magnitude simpler than Hadoop).