Entropy,
Journal Year:
2022,
Volume and Issue:
24(4), P. 525 - 525
Published: April 8, 2022
Only
the
smell
perception
rule
is
considered
in
butterfly
optimization
algorithm
(BOA),
which
prone
to
falling
into
a
local
optimum.
Compared
with
original
BOA,
an
extra
operator,
i.e.,
color
rule,
incorporated
proposed
hybrid-flash
(HFBOA),
makes
it
more
line
actual
foraging
characteristics
of
butterflies
nature.
Besides,
updating
strategy
control
parameters
by
logistic
mapping
used
HFBOA
for
enhancing
global
optimal
ability.
The
performance
method
was
verified
twelve
benchmark
functions,
where
comparison
experiment
results
show
that
converges
quicker
and
has
better
stability
numerical
problems,
are
compared
six
state-of-the-art
methods.
Additionally,
successfully
applied
engineering
constrained
problems
(i.e.,
tubular
column
design,
tension/compression
spring
cantilever
beam
etc.).
simulation
reveal
approach
demonstrates
superior
solving
complex
real-world
tasks.
Engineering Applications of Artificial Intelligence,
Journal Year:
2022,
Volume and Issue:
114, P. 105150 - 105150
Published: July 7, 2022
Machine
learning
(ML)
has
been
extensively
applied
to
model
geohazards,
yielding
tremendous
success.
However,
researchers
and
practitioners
still
face
challenges
in
enhancing
the
reliability
of
ML
models.
In
present
study,
a
systematic
framework
combining
k-fold
cross-validation
(CV),
metaheuristics
(MHs),
support
vector
regression
(SVR),
Friedman
Nemenyi
tests
was
proposed
improve
performance
geohazard
modeling.
The
average
normalized
mean
square
error
(NMSE)
from
CV
sets
adopted
as
fitness
metric.
Twenty
most
well-established
MHs
recent
were
tune
hyperparameters
SVR
evaluated
through
nonparametric
post
hoc
identify
significant
differences.
Observations
typical
reservoir
landslide
selected
benchmark
dataset,
accuracy,
robustness,
computational
time,
convergence
speed
compared.
Significant
differences
among
twenty
identified
by
absolute
(MAE),
root
squared
(RMSE),
Kling–Gupta
efficiency
(KGE),
with
p
values
lower
than
0.05.
comparison
results
demonstrated
that
multiverse
optimizer
(MVO)
is
highest-performing,
stable,
computationally
efficient
algorithms,
providing
superior
other
methods,
nearly
optimum
correlation
coefficient
(R),
low
MAE
(23.5086
versus
23.9360),
RMSE
(48.6946
50.1882),
high
KGE
(0.9803
0.9893)
predicting
displacement
Shuping
landslide.
This
paper
considerably
enriches
literature
regarding
hyperparameter
optimization
algorithms
enhancement
their
reliability.
addition,
have
potential
for
evaluating
comparing
various
ML-based
IEEE Access,
Journal Year:
2020,
Volume and Issue:
8, P. 194303 - 194314
Published: Jan. 1, 2020
Feature
selection
represents
an
essential
pre-processing
step
for
a
wide
range
of
Machine
Learning
approaches.
Datasets
typically
contain
irrelevant
features
that
may
negatively
affect
the
classifier
performance.
A
feature
selector
can
reduce
number
these
and
maximise
accuracy.
This
paper
proposes
Dynamic
Butterfly
Optimization
Algorithm
(DBOA)
as
improved
variant
to
(BOA)
problems.
BOA
one
most
recently
proposed
optimization
algorithms.
has
demonstrated
its
ability
solve
different
types
problems
with
competitive
results
compared
other
However,
original
algorithm
when
optimising
high-dimensional
Such
issues
include
stagnation
into
local
optima
lacking
solutions
diversity
during
process.
To
alleviate
weaknesses
BOA,
two
significant
improvements
are
introduced
in
BOA:
development
Local
Search
Based
on
Mutation
(LSAM)
operator
avoid
problem
use
LSAM
improve
diversity.
demonstrate
efficiency
superiority
DBOA
algorithm,
20
benchmark
datasets
from
UCI
repository
employed.
The
classification
accuracy,
fitness
values,
selected
features,
statistical
results,
convergence
curves
reported
competing
These
significantly
outperforms
comparative
algorithms
majority
used
performance
metrics.
Symmetry,
Journal Year:
2020,
Volume and Issue:
12(11), P. 1800 - 1800
Published: Oct. 30, 2020
In
order
to
solve
the
problem
that
butterfly
optimization
algorithm
(BOA)
is
prone
low
accuracy
and
slow
convergence,
trend
of
study
hybridize
two
or
more
algorithms
obtain
a
superior
solution
in
field
problems.
A
novel
hybrid
proposed,
namely
HPSOBOA,
three
methods
are
introduced
improve
basic
BOA.
Therefore,
initialization
BOA
using
cubic
one-dimensional
map
introduced,
nonlinear
parameter
control
strategy
also
performed.
addition,
particle
swarm
(PSO)
hybridized
with
for
global
optimization.
There
experiments
(including
26
well-known
benchmark
functions)
were
conducted
verify
effectiveness
proposed
algorithm.
The
comparison
results
show
HPSOBOA
converges
quickly
has
better
stability
numerical
problems
high
dimension
compared
PSO,
BOA,
other
kinds
algorithms.
