Infectious Disease Modelling,
Journal Year:
2023,
Volume and Issue:
8(1), P. 183 - 191
Published: Jan. 7, 2023
Recently
some
of
us
used
a
random-walk
Monte
Carlo
simulation
approach
to
study
the
spread
COVID-19.
The
calculations
were
reasonably
successful
in
describing
secondary
and
tertiary
waves
infection,
countries
such
as
USA,
India,
South
Africa
Serbia.
However,
they
failed
predict
observed
third
wave
for
India.
In
this
work
we
present
more
complete
set
simulations
that
take
into
consideration
two
aspects
not
incorporated
previously.
These
include
stochastic
movement
an
erstwhile
protected
fraction
population,
reinfection
recovered
individuals
because
their
exposure
new
variant
SARS-CoV-2
virus.
extended
now
show
COVID-19
India
was
missing
earlier
calculations.
They
also
suggest
additional
fourth
wave,
which
indeed
during
approximately
same
time
period
model
prediction.
Journal of Geophysical Research Biogeosciences,
Journal Year:
2025,
Volume and Issue:
130(1)
Published: Jan. 1, 2025
Abstract
Long‐term
eddy
covariance
(EC)
data
are
crucial
for
understanding
the
impact
of
global
change
on
ecosystem
functions.
However,
EC
often
contain
long
gaps,
particularly
in
tropical
dry
forests
(TDF)
due
to
seasonality
and
El
Niño‐Southern
Oscillation
(ENSO)
phases.
These
factors
create
high
variability,
complex
dependencies,
dynamic
flux
footprints.
No
current
gap‐filling
method
adequately
addresses
gaps
TDFs.
This
study
introduces
a
novel
framework
addressing
this
issue
by
(a)
defining
gap
sizes
their
relative
percentages,
(b)
training,
tuning,
evaluating
two
machine
learning
(ML)
models:
MissForest
short
Prophet
intermediate
(c)
predicting
half‐hourly
from
2013
2022
six
variables,
where
actual
sets
ranged
26.6%
28.4%,
at
TDF
Costa
Rica.
Results
indicate
that
excelled
filling
(≤5%,
R
2
=
0.76
Nash‐Sutcliffe
efficiency
(NSE)
0.71),
while
performed
exceptionally
well
between
5%
10%
(
0.72
NSE
0.67).
both
models
struggled
with
13%.
Validation
showed
values
0.79,
0.88,
0.77
CO₂
flux,
sensible
heat
latent
respectively,
corresponding
0.78,
0.86,
0.72,
normalized
root
mean
squared
error
(NRMSE)
around
2E‐4.
Additionally,
validate
our
results,
we
applied
approach
three
sites
different
ecological
conditions,
demonstrating
robust
performance.
presents
reliable
ML
imputing
data,
which
can
be
strong
variability.
Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi),
Journal Year:
2023,
Volume and Issue:
7(2), P. 405 - 413
Published: March 28, 2023
Crude
oil
price
fluctuations
affect
the
business
cycle
due
to
affecting
ups
and
downs
of
growth
economy,
which
one
indicators
economic
phenomenon.
The
importance
prediction
requires
a
model
that
can
predict
future
prices
quickly,
easily,
accurately
so
it
be
used
as
reference
in
determining
policies.
Machine
learning
is
an
accurate
method
predicting
makes
easier
because
there
no
need
program
computers
manually.
ARIMA
machine
algorithm
while
uses
seasonal
component
called
SARIMA.
Based
on
background,
research
purpose
modeling
crude
forecasting
by
Forecasting
done
daily
data
taken
from
Yahoo
Finance
January
27,
2020
25,
2023.
evaluation
results
show
RMSE
value
SARIMA
1.905.
forecast
result
7
days
ahead
with
86.230003
86.260002.
are
expected
helpful
for
policy
makers
adopt
policies
make
right
decisions
use
oil.
Advances in computational intelligence and robotics book series,
Journal Year:
2023,
Volume and Issue:
unknown, P. 234 - 242
Published: April 4, 2023
This
research
was
carried
out
in
order
to
conduct
a
sentiment
analysis
on
customer
reviews
for
an
online
store.
It
is
technique
that
makes
use
of
textual
contextual
mining
identify
and
extract
information
subjective.
type
aids
company
understanding
the
attitudes
their
customers
toward
brand,
products,
services.
When
it
comes
making
evidence-based
decisions,
taken
next
level
by
using
count-based
metrics.
The
study
examines
key
aspects
product
are
concerned
about,
as
well
reactions
or
intentions
these
have
brand
product.
machine
learning
approach,
specifically
supervised
approach.
Sentiment
decision
tree
technique.
findings
assist
makers
product,
service.
assists
them
determining
future
business
strategy,
which
will
help
increase
sales
profits.
Sustainability,
Journal Year:
2023,
Volume and Issue:
15(8), P. 6404 - 6404
Published: April 9, 2023
The
COVID-19
outbreak
is
a
disastrous
event
that
has
elevated
many
psychological
problems
such
as
lack
of
employment
and
depression
given
abrupt
social
changes.
Simultaneously,
psychologists
scientists
have
drawn
considerable
attention
towards
understanding
how
people
express
their
sentiments
emotions
during
the
pandemic.
With
rise
in
cases
with
strict
lockdowns,
expressed
opinions
publicly
on
networking
platforms.
This
provides
deeper
knowledge
human
psychology
at
time
events.
By
applying
user-produced
content
platforms
Twitter,
views
are
analyzed
to
assist
introducing
awareness
campaigns
health
intervention
policies.
modern
evolution
artificial
intelligence
(AI)
natural
language
processing
(NLP)
mechanisms
revealed
remarkable
performance
sentimental
analysis
(SA).
study
develops
new
Marine
Predator
Optimization
Natural
Language
Processing
for
Twitter
Sentiment
Analysis
(MPONLP-TSA)
Pandemic.
presented
MPONLP-TSA
model
focused
recognition
exist
data
technique
undergoes
preprocessing
convert
into
useful
format.
Furthermore,
BERT
used
derive
word
vectors.
To
detect
classify
sentiments,
bidirectional
recurrent
neural
network
(BiRNN)
utilized.
Finally,
MPO
algorithm
exploited
optimal
hyperparameter
tuning
process,
it
assists
enhancing
overall
classification
performance.
experimental
validation
approach
can
be
tested
by
utilizing
tweets
dataset
from
Kaggle
repository.
A
wide
comparable
reported
better
outcome
method
over
current
approaches.
