►
Description
Trying to use NuPIC to understand the transmission of Ebola.
http://nupic2015spring.challengepost.com/submissions/37842-vector-transmissions-of-disease-with-nupic
A
So
basically,
what
we
wanted
to
do
is
work
with
some
disease
vectors
that
I
had
worked
on
earlier
this
year
for
the
Ebola
project,
that
we
were
analyzing
data
and
trying
to
basically
locate
how
to
describe
this.
A
couple
of
factors.
One
was
trying
to
normalize
the
data
through
many
many
many
sources
and
many
variations
in
accuracy,
levels
in
rates
of
return
and
the
other
was
simply
trying
to
then
re.
You
know
establish
and
graph
the
data
so
of
it.
A
You
know
this
is
a
data
that
came
up
at
that
time
and
you
can
see
that
there
are
several
anomalies
in
the
construction
of
the
data
so
talking
with
everyone
here,
we
thought
it
might
be
pretty
interesting
to
reap
lot
this
data
using
the
hot
gem
structure
to
see
if
we
can.
Actually,
you
know
physically
see
the
anomalies
in
the
new
mix
system,
how's
it
going
to
move.
I
have
to
now
move
the
screen.
B
A
A
This
is
our
final
plot
on
and
they
can
explain
more
the
technical
details
on
the
Python
and
the
additions
to
the
the
actual
data
that
we
ran
through
and
explain
where
the
data,
how
we
do
the
analysis
on
the
data,
but
this
is
basically,
as
you
can
see
once
again
we're
capturing
the
anomalies
in
the
data
you
can
see.
If
you
want
me
to
move
back
again,
you
can
see
the
areas
where
we
had
the
drop
out
of
actual
physical.
A
One
of
the
problems
is,
if
you'll
notice
in
this
region
right
here,
I
guess
I
can
walk
right.
Is
you
have
actual
physical
dropouts
where
there's
just
no
data
for
whatever
those
transmissions?
You
know
we're
dealing
with
countries
that
have
you
know
very
difficult
transmission.
A
A
Those
are
some
mathematical
models
that
I
was
working
with
and
those
were
the
eye.
I
worked
with
an
an
S,
an
sde
model
which
is
actually
kind
of
closely
representative
to
a
stock
model
in
the
original,
because
the
idea
that
I
took
was
I
was
like
okay.
Well,
how
are
we
going
to
capture
this?
You
know
data.
Is
it's
so
erratic,
but
we
have.
A
We
have
to
have
some
kind
of
common
factor
here,
and
the
common
factor
was
much
like
stock
data
either
is
going
up
or
it's
going
down
well,
unfortunately,
here
basically,
the
initial
premise
is
you're,
either
alive
or
you
no
longer
alive
inside
of
that
data.
There's
about
five
other
tertiary
steps
of
a
probability,
diagnosis
that
it
was
in
fact
Ebola.
A
Oh
there's
a
number
of
other
options,
so
to
speak
of
how,
when
people
they
have
a
21-day
window
that
we
were
trying
to
capture
also
from
the
time
that
they
possibly
were
infected,
and
then
there
is
the
actual
outcome
of
people
of
cases
that
were
in
fact
infected
and
then
there's
a
more
about.
You
know
the
mortality
and
then
their
actual
is
a
recovery
rate.
Also
so
there's
a
huge
scholastic
equations
that
you
use
to
calculate
from
the
status.
Oh,
it's
going
to
random.
A
Take
those
scholastic
differentials
from
the
normalized
data
from
these
log
log
plot
so
anyway.
So
we
got
back
here,
and
this
is
what
we're
showing
is
that
we're?
Definitely
their
system
is
definitely
capture
the
same
drop-off
points
and
the
same
anomalies.
So
it
looks
like
between
the
two
plotting
systems
between
the
mathematical
plotting
system
and
they're
plotting
system
and
their
probability
matrix
we're
getting
very
close,
similar
relationships,
which
is
you
know,
a
very
good
great
thing
to
see.
A
So
the
next
thing
that
I'd
really
like
to
be
able
to
start
working
on-
and
we
just
basically
discuss
this-
is
working
on
the
actual
cell
data,
because
one
of
the
things
that
it's
happening
is
that
we're
getting
many
many
data
points
on
one
day
from
various
areas,
so
we'd
like
to
start
training
and
tracking
the
actual
physical
movement
of
the
disease,
so
that
we
would
know,
oh,
is
probably
going
to
X
next,
it's
probably
going
here
is
probably
going
there.
The
probability
of
it
going
here
is
less
likely
than
their.