►
From YouTube: Trustworthy Decision Mgt: Explainable, predictive decision making helps us trust our decision models
Description
The adoption of Machine Learning in conjunction with traditional Decision Management has increased over the last few years: user data can be easily collected and processed so it is crucial now to leverage similar information to build Intelligent Applications where Machine Learning and Decision Management combine.
Similar integrations can also be achieved by using many different technologies, proprietary or based on open standards. Red Hat embraces open source and open standards and the Red Hat Decision Manager platform offers the possibility to use DMN (Decision Model and Notation) and PMML (Predictive Modeling Markup Language) standards to represent decisions and predictive models well integrated together.
During this session, attendees will have the opportunity to learn more about the Trustworthy Decision Management concept and see how it can be applied to Decision Managed and Hyperautomation solutions that run natively in the cloud.
A
A
For
that
reason,
over
the
years
some
standards
have
been
developed
so
that
you
can
just
focus
on
the
model
definition
and
not
on
the
technology
that
runs
it
in
the
in
the
industry.
We
have
some
standards
so,
for
example,
for
processes
we
have
bpmn
for
decisions.
We
have
dmn
and
for
rules
we
have
drl
and
truth
in
general.
A
In
this
talk,
I'm
going
to
focus
on
dmn.
Only
red
hat
strongly
believes
in
open
source
and
open
standards
and
business
automation.
Products
have
been
built
on
top
of
such
standards
and
the
idea
behind
these
these
products
is
to
provide
a
generic
runtime
for
business
automation,
resources,
so
bpmn,
dmm
and
dll.
A
A
A
It
also
supports
the
soft
connection
between
departments
between
the
same
company,
so
the
communication
channel
are
now
the
standards
and
both
part
do
not
need
to
be
technology
dependent
because
the
the
business
analyst
that
gets,
for
example,
a
pnm
model,
is,
doesn't
really
need
to
know
how
that
model
was
trained
and
what
was
the
language?
What
was
the
data,
because
it?
It
is
a
portable
model
at
the
end
of
the
day,
red
art
started
20
years
ago
with
business
automation.
A
So
it
has
a
lot
of
experience
and
few
years
ago,
radat
started
a
new
brand
project
called
cogito
to
build
business
automation.
Microservices
cogit
is
the
next
generation
cloud
native
business,
automation,
solution
provided
by
rada,
and
it's
built
on
top
of
corgus
and
springbook.
If
you
want
to
use
it,
we
support
both.
A
Roughly
speaking,
cogito
is
the
runtime
for
your
business
models.
Bpmn
dnn
and
it's
super
easy
to
use
and
also
to
consume,
because
you
just
create
a
new
cogito
project,
which
is
a
java
project.
At
the
end
of
the
day,
you
take
your
business
models
and
you
put
them
inside
the
project.
You
just
compile
the
project
and
cogito
will
generate
the
microservice
for
you
and
it's
ready
to
be
consumed.
A
A
A
On
the
other
side,
we
have
the
business
analyst,
who
has
a
very
good
knowledge
of
the
use
case,
but
maybe
he
doesn't
have
that
good
knowledge
of
the
machine
learning
model
that
has
been
used
or
anyway,
technical
knowledge.
Of
course,
this
is
not
always
true,
but
in
general
it
is
so
take
it
as
it
is.
A
And,
as
a
matter
of
fact,
it's
easier
to
explain
simple
models
rather
than
complex
ones.
Usually,
complex
models
can
solve
complex
problem
with
very
high
performances,
but
there
are
black
boxes
and
they
are
hard
to
explain.
So
we
have
a
trade-off
between
the
explanation,
explanation
of
a
complex
model
and
its
performances.
A
I've
been
working
for
some
years
for
an
insurance
company
and
we
were
trying
to
find
some
new
variables
to
improve
the
risk
or
calculation
and
we
used
machine
learning
very
complex
machine
learning
models.
But
when
we
talked
to
our
colleagues
that
were
doing
the
traditional
risk
score,
they
talked
they.
They
said
that
they
already
tried
some
complex
machine
learning
models,
but
the
main
problem
was
the
regulator.
A
A
A
The
we
have
another
kind
of
matrix
which
are
the
operational
matrix
and
they
are
more
devops
oriented.
So
they
are
used
to
check
that
the
application
is
running
fine
and
it's
a
thing
for
example,
so
that
the
response
time
of
the
request
are
within
a
particular
range
or,
for
example,
the
number
of
requests
over
one
minute
average
and
so
on.
A
The
second
feature
is
the
accountability
and
and
the
tracing,
so
we
keep
track
of
all
the
executions
of
your
model
and
for
each
for
each
execution,
it
is
possible
to
drill
down
to
each
part
each
specific
input
and
output.
So
for
each
execution
we
provide
the
input,
the
outputs,
the
intermediate
decisions,
the
model
that
was
used
to
evaluate
that
particular
decision
and
the
explainability.
For
that
execution
we
provide
two
kind
of
explanations.
A
A
All
those
execution
are
then
stored
and
the
result
is
stored
and
a
linear
classifier
is
trained
by
labels
so
to
understand
what
were
the
most
influencing
features
that
impacted
the
original
decision.
So,
for
example,
in
this
slide,
we
see
that
the
children
is
a
very
important
feature
for
the
original
for
the
original
execution.
A
A
We
might
wonder
what
do
we
have
to
change
so
to
get
the
mortgage
approved,
so
we
set
the
desired
outcome,
which
is
the
mortgage
is
approved
and
we
also
set
some
constraints
on
the
inputs.
So,
for
example,
I
cannot
change
my
age
of
course,
so
that
one
must
be
fixed.
But,
for
example,
I
can
change
the
total
amount
of
money
that
I
request,
because
maybe
I
cannot
request
100
000,
but
I
can
request,
I
don't
know
80
000
and
with
80
000
it
gets
approved.
A
A
So
we
have
to
start
with
the
dna
model,
of
course,
so
we
have
some
inputs
here,
so
the
age,
the
monthly
salary,
the
total
asset,
the
total
amount
of
money
that
that
are
required
that
are
requested
and
the
number
of
installments
and
those
are
the
inputs
of
this
risk
score
decision,
which
is
basically
calling
the
pml.
The
pml
file
that
we
are
using
here,
which
is
around
for
small.
A
This
random
forest
model,
should
provide
a
number
between
0
to
100
0.
If
the
risk
score
is
very
low
and
100
it's
if
very
high-
and
you
see
here
that
this
risk
score
number
is,
then
the
input
of
this
mortgage
approval,
request
decision
and
it's
basically
a
decision
table.
So
if
there
is
score
is
below
40,
then
the
mortgage
is
approved.
A
There
is
another
repository,
always
under
my
github
account,
which
is
called
from
data
to
code,
and
if
you
would
like
to
do
it
step
by
step,
I
have
another
presentation
at
the
key
live,
which
is
our
community
community
channel
and
it's
45
minutes.
So
if
you
would
like
to
to
have
a
look
and
comment
it,
if
you
need
any
help,
just
feel
free
to
write
in
the
comments
or
to
reach
out
directly
to
us.
A
A
A
The
trusty
ui
is
running
under
locals
1338
and
let's
reload
the
page,
because
of
course
no
executions
were
were
calculated
before
I
executed
it,
and
you
see
here
that
we
have
one
execution
for
of
the
my
mortgage
yemen
model.
The
date
is
one
minute
ago
and
the
execution
status
was
completed
because
it
was
evaluated
successfully
from
an
operational
perspective.
A
We
can
drill
down
to
the
to
the
to
the
execution
and,
let's
see
what
were
the,
what
was
the
outcomes,
the
inputs
and
the
models
and
etc?
And
you
you
remember,
we
had
two
decisions,
so
there
is
score
and
the
mortgage
approval
decision
and
you
see
we
have
two
outcomes.
There
is
a
mortgage
approval
and
we
can
see
the
both
the
results.
A
Then
we
can
drill
down
to
the
details
of
of
the
outcomes.
We
can
select
it
with
this
menu.
So
let's
have
a
look
first
of
all
at
the
recess
code
and
we
see
that
those
were
the
influencing
input.
So
all
the
inputs
were
relevant
for
this
decision
and
it
makes
sense
because
of
course,
I
created
the
the
pmml
model.
A
So
I
know
it
and
all
the
in
all
the
features
were
relevant
for
for
the
risk
score
calculation-
and
here
we
see
also
the
inputs,
of
course,
of
this
decision,
and
if
we
go
to
the
mortgage
approval
you
see
here
that
the
mortgage
of
the
world
was
not
was
not
approved
and
the
input
for
this
particular
decision
was
the
risk
score.
So
we
can
see
all
the
intermediate
inputs
for
all
the
decision,
and
this
is
really
interesting
and
we
can
have
a
look
again
at
the
input
data
of
the
model.