Malicious
attack
is
a
major
factor
to
endanger
the
cyberspace
security.
The
accurate
detection
of
abnormal
network
traffic
generated
by
malicious
attacks
can
effectively
detect
potential
and
thus
protecting
However,
scale
relatively
small
(i.e.,
there
data
imbalance
phenomenon),
which
causes
significant
decrease
accuracy.
introduction
CycleGAN
model
deal
with
phenomenon,
but
it
suffers
from
semantic
inconsistency,
image
distortion
lack
diversity.
This
paper
proposes
an
called
CGSA-RNN
that
incorporates
CycleGAN,
self-attention
mechanism
RNN
overcome
disadvantages
model,
thereby
accurately
detecting
traffic.
first
takes
advantage
style
migration
perform
augmentation
for
small-scale
traffic,
then
replaces
ReLU
activation
function
LeakyReLU
in
generator
reduce
effects
artifacts
images.
In
addition,
introduced
into
help
better
capture
important
features,
further
improving
capability.
Finally,
uses
Extensive
experimental
results
on
two
publicly
available
datasets
show
compared
four
advanced
models
based
augmentation,
average
precision,
recall
F1-measure
are
improved
more
than
2%.
International Journal of Electrical Power & Energy Systems,
Год журнала:
2024,
Номер
159, С. 110070 - 110070
Опубликована: Июнь 3, 2024
To
conduct
analysis
on
the
field
of
electricity
management
in
buildings
is
crucial
to
contribute
clean
energy
promotion,
efficiency,
and
resilience
against
climate
change.
This
manuscript
proposes
a
methodology
for
modeling
predictive
calibrated
system
(EMS)
using
hybrid
that
combines
long
short-term
memory
multilayer
perceptron
models
(LSTM-MLP)
optimized
by
non-dominated
sorting
genetic
algorithm
II
(NSGA-II).
The
proposed
approach
utilizes
global
forecast
(GFS)
data
anticipate
consumption
fluctuations
optimize
use
distributed
sources,
such
as
photovoltaic
(PV)
production,
based
knowledge
prices
free
market
one
day
ahead.
trade-off
building
conducted
with
NSGA-II,
guaranteeing
exploration
exploitation
while
minimizing
costs
wastes.
research
carried
out
demonstrates
effectiveness
LSTM-MLP
model
advantages
NSGA-II
hyperparameter
tuning
balance
sustainable
practices.
tested
an
existing
building,
Industrial
Engineering
School
located
Campus
Lagoas-Marcosende
Universidade
de
Vigo,
Spain.
Journal of Computer Security,
Год журнала:
2025,
Номер
unknown
Опубликована: Апрель 13, 2025
The
growing
prevalence
of
encrypted
malicious
network
traffic
poses
significant
challenges
for
cybersecurity,
as
it
conceals
the
content
from
traditional
detection
methods.
Temporal
convolutional
networks
(TCNs)
present
promising
capabilities
extracting
complex
temporal
features
and
patterns
dynamic
flow
data.
However,
unidirectional
nature
TCNs
limits
their
effectiveness
in
capturing
full
context
traffic,
which
often
exhibits
bidirectional
dependencies.
Consequently,
a
few
studies
have
proposed
TCN
(BiTCN)
architectures
to
address
limitations.
these
methods
that
require
amount
parameters
be
learned,
imposes
high
memory
requirements
on
computational
resources
training
such
models.
In
this
study,
we
introduce
efficient
(eBiTCN)
model,
an
BiTCN
requires
fewer
yet
not
at
expense
cost
effective
detection.
eBiTCN
framework
combines
processor,
lightweight
gating
mechanism,
attention,
dropout,
novel
loss
function,
dense
layers.
Extensive
experiments
show
outperforms
eight
state-of-the-art
competing
models
terms
efficacy,
speed,
scalability.
model
showcased
robust
performance
detecting
evolving
attacks
excelled
across
various
real-world
datasets.
Its
efficiency
speed
reduced
usage
translates
lower
infrastructure
costs,
making
accessible
choice
deployment.
These
findings
highlight
eBiTCN’s
practicality
dependability
addressing
contemporary
security
needs.
ACM Transactions on Privacy and Security,
Год журнала:
2023,
Номер
26(4), С. 1 - 21
Опубликована: Авг. 7, 2023
In
recent
years,
the
use
of
TLS
(Transport
Layer
Security)
protocol
to
protect
communication
information
has
become
increasingly
popular
as
users
are
more
aware
network
security.
However,
hackers
have
also
exploited
salient
features
carry
out
covert
malicious
attacks,
which
threaten
security
space.
Currently,
commonly
used
traffic
detection
methods
not
always
reliable
when
applied
problem
encrypted
due
their
limitations.
The
most
significant
is
that
these
do
focus
on
key
traffic.
To
address
this
problem,
study
proposes
an
efficient
model
for
based
transport
layer
and
a
multi-head
self-attention
mechanism
called
TLS-MHSA.
Firstly,
we
extract
during
pre-processing
perform
statistics
filter
redundant
features.
Then,
learning
well
generate
important
combined
construct
model,
thereby
detecting
Finally,
public
dataset
verify
effectiveness
efficiency
TLS-MHSA
experimental
results
show
proposed
high
precision,
recall,
F1-measure,
AUC-ROC
higher
stability
than
seven
state-of-the-art
models.