International Journal of Intelligent Systems,
Год журнала:
2024,
Номер
2024, С. 1 - 21
Опубликована: Апрель 3, 2024
In
the
dynamic
global
trade
environment,
accurately
predicting
values
of
diverse
commodities
is
challenged
by
unpredictable
economic
and
political
changes.
This
study
introduces
Meta-TFSTL
framework,
an
innovative
neural
model
that
integrates
Meta-Learning
Enhanced
Trade
Forecasting
with
efficient
multicommodity
STL
decomposition
to
adeptly
navigate
complexities
forecasting.
Our
approach
begins
partition
value
sequences
into
seasonal,
trend,
residual
elements,
identifying
a
potential
10-month
cycle
through
Ljung–Box
test.
The
employs
dual-channel
spatiotemporal
encoder
for
processing
these
components,
ensuring
comprehensive
grasp
temporal
correlations.
By
constructing
spatial
graphs
leveraging
correlation
matrices
graph
embeddings
introducing
fused
attention
multitasking
strategies
at
decoding
phase,
surpasses
benchmark
models
in
performance.
Additionally,
integrating
meta-learning
fine-tuning
techniques
enhances
shared
knowledge
across
import
export
predictions.
Ultimately,
our
research
significantly
advances
precision
efficiency
forecasting
volatile
scenario.
AIP conference proceedings,
Год журнала:
2024,
Номер
3033, С. 020018 - 020018
Опубликована: Янв. 1, 2024
Today,
companies'
green
competitiveness
is
developing
under
the
influence
of
a
wide
set
economic,
social,
environmental,
corporate
and
marketing
determinants.
The
purpose
paper
to
justify
role
determinants
in
competitiveness.
provided
analysis
semantic
core
information-commercial
scientific
analytics
on
search
queries
related
using
Google
Analytics
Trends
allowed
identifying
following
groups
its
provision:
sustainable
strategic
development
(green
strategies,
supply
chains,
logistics,
pricing);
media
information
networks,
platforms,
digital
tools,
web
tools);
targeted
brand,
advertising,
promotion).
systematic
combination
correlation
Godrick-Prescott
smoothing
filtering
method
time
series
issues
enterprises
three
different
vector
trends:
increasing
level
commercial
interest
field
(at
request
Internet
users)
accompanied
by
community
both
trend
cyclical
components
(positive
statistically
significant
correlation);
an
uneven
growth
part
users
(negative
object
dominant
researchers
are
such
as
brand
promotion
(results
components).
AIP conference proceedings,
Год журнала:
2024,
Номер
3033, С. 020001 - 020001
Опубликована: Янв. 1, 2024
The
main
idea
of
the
article
is
to
draw
attention
necessity
combining
multiple
methods
in
research
process
with
a
view
obtaining
comprehensive
answer
questions
posed.
an
attempt
demonstrate
rationale
for
using
triangulation
processes.
Triangulation
presented
as
methodology
procedure
and
condition
enhancing
reliability
research.
advantages
triangulating
are
also
discussed.
highlights
importance
raising
quality
management
sciences;
it
respond
challenges
civilization,
which
determined
by
science
economy.
AIP conference proceedings,
Год журнала:
2024,
Номер
3033, С. 020025 - 020025
Опубликована: Янв. 1, 2024
In
order
to
ensure
sustainable
development
and
system
management
of
livestock
farming,
the
expansion
scientific
research
boundaries
interdisciplinary
focus
science
is
becoming
increasingly
important.
this
regard,
at
present
stage
in
field
animal
breeding,
especially
intensive
possibility
early
forecasting
long-term
productive
use
animals
with
high
levels
productivity
important,
which
turn
puts
forward
need
combine
efforts
different
branches
science.
For
example,
selection
more
attention
paid
not
only
appearance,
but
also
genotype
animals,
particular
isolation
individual
species
genotype,
a
certain
allelic
state
genes.
last
years,
scientists
agrarian
university
have
been
actively
studying
somatotropin
cascade
genes,
it
was
decided
deepen
prevalence
alleles
gene
breed,
Holstein
breed
cattle,
widely
spread
Kazakhstan.
The
laboratory
DNA-typing
on
polymorphisms
bGH-AluI,
bIGF-1-SnaBI
carried
out
analysis
genetic
structure
researched
populations
out.
genotypes
each
were
established.
aim
article
evaluate
association
pair
combinations
life
expectancy
cows
Kazakh
breeding
studied.
there
observed
strengthening
phenotypic
effects
towards
an
increase
milk
indicators
combination¹
1
bGH-AluILL-bIGF-1-SnaBIAA.
record
diplotype
bGH-AluILV-bIGF-1-SnaBIAA,
combined
effect
relation
separately
taken
by
trait
total
yield.
AIP conference proceedings,
Год журнала:
2024,
Номер
3033, С. 020022 - 020022
Опубликована: Янв. 1, 2024
Based
on
the
materials
of
specific
agricultural
formations
in
Almaty
region,
this
paper
presents
a
methodological
approach
to
assessing
effectiveness
using
resource
potential.
Approved
economic-mathematical
model
optimizing
production
and
industrial
structure,
justifying
optimal
parameters
elements
potential
through
which
there
are
revealed
reserves
for
increasing
efficiency
The
research
results
show
that
use
economic
mathematical
optimization
models
will
significantly
increase
formations.
Practical
recommendations
given
substantiate
all
types
enterprises
region
based
methodology.
International Journal of Intelligent Systems,
Год журнала:
2024,
Номер
2024, С. 1 - 21
Опубликована: Апрель 3, 2024
In
the
dynamic
global
trade
environment,
accurately
predicting
values
of
diverse
commodities
is
challenged
by
unpredictable
economic
and
political
changes.
This
study
introduces
Meta-TFSTL
framework,
an
innovative
neural
model
that
integrates
Meta-Learning
Enhanced
Trade
Forecasting
with
efficient
multicommodity
STL
decomposition
to
adeptly
navigate
complexities
forecasting.
Our
approach
begins
partition
value
sequences
into
seasonal,
trend,
residual
elements,
identifying
a
potential
10-month
cycle
through
Ljung–Box
test.
The
employs
dual-channel
spatiotemporal
encoder
for
processing
these
components,
ensuring
comprehensive
grasp
temporal
correlations.
By
constructing
spatial
graphs
leveraging
correlation
matrices
graph
embeddings
introducing
fused
attention
multitasking
strategies
at
decoding
phase,
surpasses
benchmark
models
in
performance.
Additionally,
integrating
meta-learning
fine-tuning
techniques
enhances
shared
knowledge
across
import
export
predictions.
Ultimately,
our
research
significantly
advances
precision
efficiency
forecasting
volatile
scenario.