Composting
is
a
promising
strategy
for
manure
treatment
that
recycles
organic
waste.
However,
nitrous
oxide
(N2O),
greenhouse
gas,
generated
during
the
composting
process.
This
causes
nitrogen
loss
and
pollution,
contributes
to
global
warming
which
problem
of
concern.
Therefore,
it
necessary
develop
an
approach
accurately
quantify
N2O
emissions
explore
relationships
between
parameters.
study
employed
model-agnostic
meta-learning
(MAML)
enhance
performance
prediction
based
on
machine
learning.
The
highest
R2
lowest
root
mean
squared
error
(RMSE)
values
achieved
were
0.939
18.42
mg
d-1,
respectively.
Five
learning
methods
adopted
comparison
further
demonstrate
effectiveness
MAML
model.
Feature
analysis
showed
moisture
content
co-composting
material
ammonia
concentration
process
two
most
significant
features
affecting
emissions.
serves
as
proof
application
method
waste
management,
giving
new
insights
into
effects
properties
data
can
help
determine
optimum
conditions
mitigating
in
composting.
IEEE Access,
Journal Year:
2023,
Volume and Issue:
11, P. 101055 - 101086
Published: Jan. 1, 2023
The
degradation
of
water
quality
has
become
a
critical
concern
worldwide,
necessitating
innovative
approaches
for
monitoring
and
predicting
quality.
This
paper
proposes
an
integrated
framework
that
combines
the
Internet
Things
(IoT)
machine
learning
paradigms
comprehensive
analysis
prediction.
IoT-enabled
comprises
four
modules:
sensing,
coordinator,
data
processing,
decision.
IoT
is
equipped
with
temperature,
pH,
turbidity,
Total
Dissolved
Solids
(TDS)
sensors
to
collect
from
Rohri
Canal,
SBA,
Pakistan.
acquired
preprocessed
then
analyzed
using
models
predict
Water
Quality
Index
(WQI)
Class
(WQC).
With
this
aim,
we
designed
learning-enabled
Preprocessing
steps
such
as
cleaning,
normalization
Z-score
technique,
correlation,
splitting
are
performed
before
applying
models.
Regression
models:
LSTM
(Long
Short-Term
Memory),
SVR
(Support
Vector
Regression),
MLP
(Multilayer
Perceptron)
NARNet
(Nonlinear
Autoregressive
Network)
employed
WQI,
classification
SVM
Machine),
XGBoost
(eXtreme
Gradient
Boosting),
Decision
Trees,
Random
Forest
WQC.
Before
that,
Dataset
used
evaluating
split
into
two
subsets:
1
2.
600
values
each
parameter,
while
2
includes
complete
set
6000
parameter.
division
enables
comparison
evaluation
models'
performance.
results
indicate
regression
model
strong
predictive
performance
lowest
Mean
Absolute
Error
(MAE),
Squared
(MSE),
Root
(RMSE)
values,
along
highest
R-squared
(0.93),
indicating
accurate
precise
predictions.
In
contrast,
demonstrates
weaker
performance,
evidenced
by
higher
errors
lower
(0.73).
Among
algorithms,
achieves
metrics:
accuracy
(0.91),
precision
recall
(0.92),
F1-score
(0.91).
It
also
conceived
perform
better
when
applied
datasets
smaller
numbers
compared
larger
values.
Moreover,
comparisons
existing
studies
reveal
study's
improved
consistently
For
classification,
outperforms
others,
exceptional
accuracy,
precision,
recall,
metrics.
Water Research X,
Journal Year:
2024,
Volume and Issue:
24, P. 100256 - 100256
Published: Sept. 1, 2024
Due
to
the
large
amounts
of
pharmaceuticals
and
personal
care
products
(PPCPs)
currently
being
consumed
released
into
environment,
this
study
provides
a
comprehensive
analysis
pharmaceutical
pollution
in
both
raw
treated
water
from
full-scale
drinking
treatment
plants
nationwide.
Our
investigation
revealed
that
30
out
37
PPCPs
were
present
with
mean
concentrations
ranging
0.01-131
ng/L.
The
sources,
surface
(ND
-
147
ng/L),
subsurface
123
ng/L)
reservoir
sources
135
exhibited
higher
concentration
levels
residues
compared
groundwater
1.89
ng/L).
Meanwhile,
water,
17
analyzed
carbamazepine,
clarithromycin,
fluconazole,
telmisartan,
valsartan,
cotinine
most
common
(detection
frequency
>
40
%),
having
1.22,
0.12,
3.48,
40.1,
6.36,
3.73
ng/L,
respectively.
These
findings
highlight
that,
while
processes
are
effective,
there
some
persistent
compounds
prove
challenging
fully
eliminate.
Using
Monte
Carlo
simulations,
risk
assessment
indicated
these
likely
have
negligible
impact
on
human
health,
except
for
antihypertensives.
Telmisartan
was
identified
as
posing
highest
ecological
(RQ
1),
warranting
further
investigation,
monitoring.
concludes
by
prioritizing
specific
14
pharmaceuticals,
including
lamotrigine,
cotinine,
lidocaine,
tramadol,
others,
future
monitoring
safeguard
health.
Ozone
has
demonstrated
high
efficacy
in
depredating
emerging
contaminants
(ECs)
during
drinking
water
treatment.
However,
traditional
quantitative
structure-activation
relationship
(QSAR)
models
often
fall
short
effectively
normalizing
and
characterizing
diverse
molecular
structures,
thereby
limiting
their
predictive
accuracy
for
the
removal
of
various
ECs.
This
study
uses
embedded
structure
vectors
generated
by
a
graph
neural
network
(GNN),
combined
with
functional
group
prompts,
as
inputs
to
feedforward
network.
A
data
set
28
ECs
542
points,
representing
structures
physiochemical
properties,
was
built
predict
residual
rate
(REC)
ozonation
oxidation.
Compared
QSAR
models,
GNN-based
methods
significantly
improve
prediction
accuracy.
The
resulting
KANO-EC
model
achieved
an
R2
0.97
REC,
demonstrating
its
ability
capture
complex
structural
features.
Moreover,
maintains
exceptional
interpretability,
elucidating
key
groups
(e.g.,
carbonyls,
hydroxyls,
aromatic
rings,
amines)
involved
oxidation
mechanism.
presents
novel
approach
predicting
efficiency
current
potential
also
provides
valuable
insights
developing
efficient
control
strategies
ensuring
long-term
safety
sustainability
supplies.
