NATURENGS MTU Journal of Engineering and Natural Sciences Malatya Turgut Ozal University,
Год журнала:
2024,
Номер
unknown
Опубликована: Окт. 25, 2024
Image
classification
is
a
critical
area
of
research
with
widespread
applications
across
various
disciplines,
including
computer
vision,
pattern
recognition,
and
artificial
intelligence.
Despite
the
advancements
in
Convolutional
Neural
Networks
(CNNs),
which
have
revolutionized
field
by
providing
powerful
tools
for
image
classification,
many
studies
encountered
challenges
achieving
optimal
performance.
These
often
arise
from
complex
nature
CNN
architectures
multitude
hyperparameters
that
require
fine-tuning.
Among
models,
AlexNet
has
been
widely
recognized
its
contributions
to
deep
learning,
yet
there
remains
significant
potential
improvement
through
optimization
hyperparameters.
In
this
study,
WF-AlexNET
designed
enhance
performance
architecture
optimizing
first
convolutional
layer
using
Equilibrium
Optimization
(EO)
algorithm.
The
EO
algorithm,
was
employed
fine-tune
filter
size,
number,
stride,
padding
parameters,
are
crucial
effective
feature
extraction.
proposed
method
rigorously
tested
on
multi-class
weather
dataset
evaluate
accuracy
robustness.
Experimental
results
demonstrate
significantly
outperforms
standard
model,
10.5%
increase
mean
validation
6.51%
test
accuracy.
Furthermore,
approach
compared
against
other
prominent
architectures,
VGG-16,
GoogleNet,
ShuffleNet,
MobileNet-V2,
VGG-19.
consistently
exhibited
superior
multiple
metrics,
F1-score
maximum
accuracy,
highlighting
efficacy
addressing
associated
hyperparameter
CNNs.
Atmosphere,
Год журнала:
2024,
Номер
15(12), С. 1407 - 1407
Опубликована: Ноя. 22, 2024
Predicting
streamflow
is
essential
for
managing
water
resources,
especially
in
basins
and
watersheds
where
snowmelt
plays
a
major
role
river
discharge.
This
study
evaluates
the
advanced
deep
learning
models
accurate
monthly
peak
forecasting
Gilgit
River
Basin.
The
utilized
were
LSTM,
BiLSTM,
GRU,
CNN,
their
hybrid
combinations
(CNN-LSTM,
CNN-BiLSTM,
CNN-GRU,
CNN-BiGRU).
Our
research
measured
model’s
accuracy
through
root
mean
square
error
(RMSE),
absolute
(MAE),
Nash–Sutcliffe
efficiency
(NSE),
coefficient
of
determination
(R2).
findings
indicated
that
models,
CNN-BiGRU
achieved
much
better
performance
than
traditional
like
LSTM
GRU.
For
instance,
lowest
RMSE
(71.6
training
95.7
testing)
highest
R2
(0.962
0.929
testing).
A
novel
aspect
this
was
integration
MODIS-derived
snow-covered
area
(SCA)
data,
which
enhanced
model
substantially.
When
SCA
data
included,
CNN-BiLSTM
improved
from
83.6
to
71.6
during
108.6
testing.
In
prediction,
outperformed
other
with
(108.4),
followed
by
(144.1).
study’s
results
reinforce
notion
combining
CNN’s
spatial
feature
extraction
capabilities
temporal
dependencies
captured
or
GRU
significantly
enhances
accuracy.
demonstrated
improvements
prediction
accuracy,
extreme
events,
highlight
potential
these
support
more
informed
decision-making
flood
risk
management
allocation.
Scientific Reports,
Год журнала:
2025,
Номер
15(1)
Опубликована: Фев. 11, 2025
Abstract
The
Parrot
Optimizer
(PO)
has
recently
emerged
as
a
powerful
algorithm
for
single-objective
optimization,
known
its
strong
global
search
capabilities.
This
study
extends
PO
into
the
Multi-Objective
(MOPO),
tailored
multi-objective
optimization
(MOO)
problems.
MOPO
integrates
an
outward
archive
to
preserve
Pareto
optimal
solutions,
inspired
by
behavior
of
Pyrrhura
Molinae
parrots.
Its
performance
is
validated
on
Congress
Evolutionary
Computation
2020
(CEC’2020)
benchmark
suite.
Additionally,
extensive
testing
four
constrained
engineering
design
challenges
and
eight
popular
confined
unconstrained
test
cases
proves
MOPO’s
superiority.
Moreover,
real-world
helical
coil
springs
automotive
applications
conducted
depict
reliability
proposed
in
solving
practical
Comparative
analysis
was
performed
with
seven
published,
state-of-the-art
algorithms
chosen
their
proven
effectiveness
representation
current
research
landscape-Improved
Manta-Ray
Foraging
Optimization
(IMOMRFO),
Gorilla
Troops
(MOGTO),
Grey
Wolf
(MOGWO),
Whale
Algorithm
(MOWOA),
Slime
Mold
(MOSMA),
Particle
Swarm
(MOPSO),
Non-Dominated
Sorting
Genetic
II
(NSGA-II).
results
indicate
that
consistently
outperforms
these
across
several
key
metrics,
including
Set
Proximity
(PSP),
Inverted
Generational
Distance
Decision
Space
(IGDX),
Hypervolume
(HV),
(GD),
spacing,
maximum
spread,
confirming
potential
robust
method
addressing
complex
MOO
