Big Data is Useless Without Science. WRONG.
counterexample:
genetic algorithm - no rules - just results.
1.)traditional scientific method. hypothesis, then
prediction test. Deduction using BIG DATA.
I call this quantitative deduction and classify it in
the class of brute force.
2.)induction using quality DATA.
I call this QUALITATIVE induction and classify it in
the class of 'elegance.'
3.)ABduction using quality/quantity DATA.
WTF? Oversimplified explanation:
Black swan - counter-examples or FALSIFY occurs only rarely - need BIG DATA to find the elusive black swan.
example:
Physics: Newton's law works 99% of the time. Need
QUANTITATIVE BIG DATA to get to 'corner cases' and 'quantum
mechanics.'
So, 'BIG DATA' useful in finding COUNTEREXAMPLES.
So, 'Small Quality Data useful for 'testing predictions' - abduction is a mixture.
http://ultimavi.arc.net.my/ave/integra-j.html
key question: does MORE and BIG mean BETTER?
induction involves the classification or generalization of examples for explaining tendencies in observations. - YES?
Abduction involves the adoption of existing or non-existing hypotheses for a given observation explanation. - NO?
PS.due to FLEXIBILITY of human mind versus AI, I prefer
ABDUCTION.
Thought test: IBM Watson.
Method Winner Big or Quantitative DATA
deduction AI sometimes or TRUE/FALSE
induction AI TRUE
abduction Human FALSE
the next post is 'fuzzy logic' Zadeh and 'fuzzy data'?
USA Mortgages. Banking ASSOCIATION creates MERS, a database.
How do you know it is fake?
ruled against in various courts.
replaces the HISTORIC paper trail in local cities, counties
for HOME OWNERSHIP TITLE.
bank of america forecloses on WRONG houses.
no audits made public.
ROBOSIGNING.
ROBO-lawsuits using DCMA.
less than ethical lawyers - get rich, then flee to
non-extradition country.
use the internet as tracking and Big Brother device.
theater show - Potemkin village - crack down and WIN against
high school kids, but let BIG COUNTRIES like China (PRC),
Africa (Nigeria?) 'negotiate.'
Big Data is Useless Without Science. WRONG. counterexample: genetic algorithm - no rules - just results.
prediction test. Deduction using BIG DATA. I call this quantitative deduction and classify it in the class of brute force. 2.)induction using quality DATA. I call this QUALITATIVE induction and classify it in the class of 'elegance.' 3.)ABduction using quality/quantity DATA. WTF? Oversimplified explanation: Black swan - counter-examples or FALSIFY occurs only rarely - need BIG DATA to find the elusive black swan.
example: Physics: Newton's law works 99% of the time. Need QUANTITATIVE BIG DATA to get to 'corner cases' and 'quantum mechanics.' So, 'BIG DATA' useful in finding COUNTEREXAMPLES. So, 'Small Quality Data useful for 'testing predictions' - abduction is a mixture.
http://ultimavi.arc.net.my/ave/integra-j.html key question: does MORE and BIG mean BETTER? induction involves the classification or generalization of examples for explaining tendencies in observations. - YES? Abduction involves the adoption of existing or non-existing hypotheses for a given observation explanation. - NO?
PS.due to FLEXIBILITY of human mind versus AI, I prefer ABDUCTION.
Thought test: IBM Watson. Method Winner Big or Quantitative DATA deduction AI sometimes or TRUE/FALSE induction AI TRUE abduction Human FALSE
the next post is 'fuzzy logic' Zadeh and 'fuzzy data'?