Intelligent information systems,
Год журнала:
2002,
Номер
unknown, С. 23 - 32
Опубликована: Янв. 1, 2002
Discretization
of
continuous
attributes
offers
a
number
benefits
for
machine
learning
process.
Fundamental
are:
significant
decreasing
time,
possible
improvement
knowledge
quality
in
case
noisy
data,
an
increasing
legibility.
In
this
paper
we
show
analysis
multidiscretizer
which
allows
to
apply
wide
range
supervised
discretization
algorithms
and
has
very
good
abilities
tuning.
Designed
generate
new
also
variants
similar
well
known
classic
such
as
e.g.,
ChiMerge.
The
is
based
on
top-down
method
bottom-up
one.
It
uses
many
measures
are
mainly
used
induction
decision
trees
few
stop
criterions.
study
the
generated
by
through
evaluation
trees'
accuracy.
We
focus
comparison
with
one
configuration
presented
below.
For
research
use
tree
C4.5
ID3.
Journal of information and organizational sciences,
Год журнала:
2021,
Номер
45(1), С. 55 - 75
Опубликована: Июнь 15, 2021
This
article
provides
a
brief
discussion
of
ERPs,
including
how
they
can
be
defined,
their
component
parts,
and
the
evolution
technologies
since
1940s.
The
then
offers
general
overview
current
ERP
vendor
market
landscape
based
on
academic
journals
industry
reports.
article’s
methodological
approach
is
review
both
literature
expert
reports,
which
author
performed
to
trace
trends
within
landscape.
Results
analysis
demonstrate
that
rapidly
evolving
dates
back
at
least
six
decades,
there
are
many
large,
longstanding
vendors.
Yet,
opportunities
like
cloud
computing
may
result
in
niche
products
existing
players
or
through
start-ups.
Nonetheless,
such
as
blockchain
emerging.
It
remains
seen
will
impact
market.
International Journal of Uncertainty Fuzziness and Knowledge-Based Systems,
Год журнала:
2015,
Номер
23(06), С. 893 - 908
Опубликована: Дек. 1, 2015
Rough
set
is
mainly
concerned
with
the
approximations
of
objects
through
an
equivalence
relation
on
a
universe.
Matroid
generalization
linear
algebra
and
graph
theory.
Recently,
matroidal
structure
rough
sets
established
applied
to
problem
attribute
reduction
which
important
application
In
this
paper,
we
propose
new
call
it
parametric
matroid.
On
one
hand,
for
universe,
family,
any
subset
universe
as
its
parameter,
defined
lower
approximation
operator.
This
family
proved
satisfy
independent
axiom
matroids,
therefore
matroid
generated,
set.
Through
operator,
three
equivalent
representations
are
obtained.
Moreover,
be
direct
sum
partition-circuit
free
other
matroids
well
studied
number,
then
use
investigate
Several
characteristics
set,
such
sets,
bases,
circuits,
rank
function
closure
expressed
by
number.
International Journal of Engineering Research and Advanced Technology,
Год журнала:
2020,
Номер
06(12), С. 38 - 45
Опубликована: Янв. 1, 2020
Artificial
intelligence
(AI)
is
the
use
of
computer
science
to
develop
machine
that
can
be
trained
learn,
reason,
communicate,
and
make
human-like
decisions.It
a
technology
rapidly
being
adopted
in
many
industries
improve
performance,
precision,
time
efficiency,
cost
reduction.The
artificial
healthcare
an
emerging
scientific
area
aims
generate
by
analyzing
health
data.AIis
becoming
increasingly
attractive
industry
changing
landscape
biomedical
research.This
paper
provides
overview
broad
range
applications
AI
healthcare.
This
paper
develops
a
novel
supervised
method
for
predicting
earthquake
ground
motions
in
the
wavelet
domain.
The
training
input
is
set
of
seismological
predictors
related
to
seismic
source,
path
and
local
site
conditions,
output
consists
weights
from
multiway
analysis
motions.
We
treat
transforms
acceleration
records
as
images
extract
essential
patterns
them
using
tensor
decomposition.
decomposition
these
are
then
linked
variables
general
regression
neural
network
(GRNN).
resulting
nonparametric
model
used
predict
image
an
accelerogram
given
variables.
predicted
can
be
transformed
back
time
domain
inverse
transform
subsequent
processing
match
design
spectrum.
Unlike
conventional
motion
models,
proposed
approach
retains
characteristics
Pearson's
correlation
coefficient
between
vectorized
forms
actual
has
been
similarity
metric
assessing
prediction
capability
model.
Experimental
results
demonstrate
ability
significant
energy
distribution.
ADCAIJ ADVANCES IN DISTRIBUTED COMPUTING AND ARTIFICIAL INTELLIGENCE JOURNAL,
Год журнала:
2017,
Номер
6(2), С. 19 - 32
Опубликована: Июнь 30, 2017
Multi-Agent
Systems
(MAS)
software
has
been
increasing
dramatically
in
last
years.
In
this
context,
the
Software
Agents,
Environments
and
Applications
School
(WESAAC)
is
a
Brazilian
event
of
MAS
which
its
tenth
edition.
When
studying
research
area,
it
important
to
identify
most
active
groups,
topics,
trends
so
forth.
This
study
aims
investigate
how
WESAAC
evolving,
by
analysing
papers
published
ten
editions.
We
adopted
strategy
that
combines
scoping
systematic
review
good
practices.
participating
institutions
authors,
main
topics
discussed,
types
contributions,
conferences
journals
have
referenced
papers,
publications
with
greatest
impact,
MAS.
found
128
over
10
editions,
were
analysed
discussed.
