►
From YouTube: Numenta HTM Challenge - HTM Models Adelaide
Description
My submission for the Numenta Challenge. My application uses HTM models to detect anomalous traffic flows in Adelaide, Australia. The idea is that anomalous traffic indicates an incident.
Github here: https://github.com/JonnoFTW/htm-models-adelaide
A
So
hi
I'm
Jonathan
Mackenzie-
and
this
is
my
presentation
for
the
new
payment
to
HTM
challenge.
My
presentation
is
on
HTM
models
for
automate
into
the
detection
in
Adelaide
Australia.
So
what
is
automating
to
the
detection?
Firstly,
we
need
to
know
what
congestion
is
and
that's
basically
where
the
capacity
of
the
road
is
exceeded,
and
so
everyone
slows
down.
So
this
can
be
recurrent,
which
you
might
think
of
as
something
like
rush
hour
in
the
morning
when
everyone
goes
to
work
and
wash
your
own
the
afternoon
what
everyone
goes
home.
A
So
our
goal
is
to
detect
non-recurrent
condition
and
that's
where
something
like
an
accident.
A
breakdown,
unskilled
load
burst,
pipe
landslides,
flooding
rights
and
protests
cause
congestion
on
the
road.
So
the
problem
is
basically
that
freeway
autumn
agents
and
detection
is
well-researched
and
there's
lots
of
solutions
for
it.
A
But
we
want
to
try
and
do
the
same
thing
but
on
arterial
roads,
which
is
much
more
difficult
due
to
the
very
nature
of
the
traffic
and
everyone's
turning
going
different
ways
and
we
use
loop
detectors,
which
are
sense
that
underneath
the
road
which
counts
traffic
and
they
come
with
their
own
warts.
So
in
our
solution
we
use
HTM
to
detect
anomalous
traffic
flow.
This
is
an
improvement
on
previous
research,
which
made
mostly
used
simulations
and
supervised
learning
techniques
which
allowed
people
to
easily
set
up
incidents
and
traffic
flow
and
can
really
easily
monitor
everything.
A
A
So
we've
currently
got
three
and
a
half
terabytes
of
data
from
seven
years
and
that's
at
five
minute
intervals,
but
for
now
I'm
only
looking
at
two
month
period,
which
is
about
130,
intersections
or
170,000
data
points
per
intersection,
and
this
is
private
data.
So
you
don't
have
access
to
it.
Sorry,
and
we
also
use
crash
data
to
verify
our
anomaly
prediction
and
there's
142,000
incidents,
but
in
a
period
we're
looking
at
there's
only
a
turn
reach
500
in
this
public
data.
You
get
it
from
DSA
of
today
you
so
here's
my
readings
data.
A
Basically,
it's
just
a
collection
of
objects
which
map
the
sensor
to
the
count
of
the
sensor.
So
we
can
see
that
on
intersection,
3041
at
830am
there
was
58
vehicles
on
sensor,
136
and
here's
all
the
crashes
which
basically
just
say
when
and
where
crash
occurred,
and
what
sort
of
features
that
have
like
raining
for
drivers
etc.
A
So
the
analysis
script
uses
HTM
and
there's
an
option
to
do
it
with
one
model
which
takes
in
every
sensor
as
an
input,
and
then
we
get
the
anomaly
score
from
that,
but
they
didn't
work
very
well.
So
we
use
one
model
/
sensor.
So
if
its
intersection
with
24
sensors
that'll,
be
twenty,
twenty
four
models
24
out
different
outputs,
but
still
this
takes
a
long
time.
So
we
use
zipper
computer.
A
So
here
you
can
see
the
script
running
on
my
machine
using
multi
model,
which
means
that
there's
one
model
/
sensor,
which
run
it
runs
in
its
own
sub
process.
This
is
for
inter
section
3001
and
writing.
The
results
of
the
database
is
also
a
smoothing
option
available
which
we
haven't
used,
but
this
applies
a
median
filter
with
the
window
size
of
your
choosing
to
the
readings
data,
and
you
can
see
down
here
that
this
would
uses
more
processes,
then
cause
inter
section.
3001
uses
twenty
senses,
so
it
won't
run
as
fast
as
it
could.
A
A
And
then
you
can
reveal
it
using
the
ND
web
application.
So
here
you
can
see
the
lovely
city
of
adelaide,
and
this
is
a
central
business
district,
which
is
the
only
area
that
is
traffic
for
so
we
can
click
on
each
section
and
kind
of
button
to
see
the
data
for
the
intersection.
So
here
we've
got
intersection,
3083
to
signalize
t-junction
and
here's
all
the
readings
on
sensor
56-
and
here
we
have
the
anomaly
scores
in
blue
and
lee
likelihood
in
red
incidents
in
green
and
the
orange
dot
indicate
that
a
sensor
has
been
exceeded.
A
A
So
certainly
here
there
was
a
significant
dip
due
to
this
incident
here
and
if
we
mouse
over,
we
can
see
that
it
was
caused
by
inattention
and
was
a
rear
end
that
cost
two
thousand
dollars
and
down
here
we
can
see
on
the
map
exactly
where
it
occurred
for
the
going
north,
but
it
did
think
this
bit
of
traffic
over
here
was
anomalous
on
that
was
on
sensor,
848,
etc.
We
can
click
the
button
to
change
the
sensor.
A
So
here
we
have
a
list
of
incidents
with
the
date
and
which
innocent,
which
intersection
a
decoder
and
if
any
of
its
senses
exceeded
the
point
99
threshold,
then
they'll
be
shown
here
with
the
sensor.
We
can
filter
to
only
show
ones
which
have
exceeded
and
will
click
on
this
one
here,
30
40,
and
so
here
we
can
see
there
was
an
accident
at
that
time.