Small,
Journal Year:
2024,
Volume and Issue:
20(29)
Published: Feb. 11, 2024
Functional
nanostructures
build
up
a
basis
for
the
future
materials
and
devices,
providing
wide
variety
of
functionalities,
possibility
designing
bio-compatible
nanoprobes,
etc.
However,
development
new
nanostructured
via
trial-and-error
approach
is
obviously
limited
by
laborious
efforts
on
their
syntheses,
cost
manpower.
This
one
reasons
an
increasing
interest
in
design
novel
with
required
properties
assisted
machine
learning
approaches.
Here,
dataset
synthetic
parameters
optical
important
class
light-emitting
nanomaterials
-
carbon
dots
are
collected,
processed,
analyzed
transitions
red
near-infrared
spectral
ranges.
A
model
prediction
characteristics
these
based
multiple
linear
regression
established
verified
comparison
predicted
experimentally
observed
synthesized
three
different
laboratories.
Based
analysis,
open-source
code
provided
to
be
used
researchers
procedures.
Hybrid Advances,
Journal Year:
2023,
Volume and Issue:
2, P. 100026 - 100026
Published: Feb. 4, 2023
Reinforced
composite
is
a
preferred
choice
of
material
for
the
design
industrial
lightweight
structures.
As
late,
materials
analysis
and
development
utilizing
machine
learning
algorithms
have
been
getting
expanding
consideration
accomplished
extraordinary
upgrades
in
both
time
productivity
expectation
exactness.
This
review
encapsulates
recent
advances
learning-based
reinforced
during
last
half-decade.
It
summarizes
limitations
traditional
methods
presents
detailed
protocol
technology;
implementation
was
covered,
with
an
emphasis
on
importance
data
hygiene.
Machine
integration
process
selection,
sourcing
techniques
were
also
examined.
The
evaluation
looked
at
emerging
digital
tools
platforms
implementing
algorithms.
In
addition,
essential
effort
made
to
identify
research
gaps
define
areas
further
research.
indeed
designed
provide
some
direction
future
into
use
design.
ACS Applied Bio Materials,
Journal Year:
2023,
Volume and Issue:
7(2), P. 510 - 527
Published: Jan. 26, 2023
Polymers,
with
the
capacity
to
tunably
alter
properties
and
response
based
on
manipulation
of
their
chemical
characteristics,
are
attractive
components
in
biomaterials.
Nevertheless,
potential
as
functional
materials
is
also
inhibited
by
complexity,
which
complicates
rational
or
brute-force
design
realization.
In
recent
years,
machine
learning
has
emerged
a
useful
tool
for
facilitating
via
efficient
modeling
structure–property
relationships
domain
interest.
this
Spotlight,
we
discuss
emergence
data-driven
polymers
that
can
be
deployed
biomaterials
particular
emphasis
complex
copolymer
systems.
We
outline
developments,
well
our
own
contributions
takeaways,
related
high-throughput
data
generation
polymer
systems,
methods
surrogate
learning,
paradigms
property
optimization
design.
Throughout
discussion,
highlight
key
aspects
successful
strategies
other
considerations
will
relevant
future
polymer-based
target
properties.
Materials,
Journal Year:
2023,
Volume and Issue:
16(17), P. 5977 - 5977
Published: Aug. 31, 2023
Material
innovation
plays
a
very
important
role
in
technological
progress
and
industrial
development.
Traditional
experimental
exploration
numerical
simulation
often
require
considerable
time
resources.
A
new
approach
is
urgently
needed
to
accelerate
the
discovery
of
materials.
Machine
learning
can
greatly
reduce
computational
costs,
shorten
development
cycle,
improve
accuracy.
It
has
become
one
most
promising
research
approaches
process
novel
material
screening
property
prediction.
In
recent
years,
machine
been
widely
used
many
fields
research,
such
as
superconductivity,
thermoelectrics,
photovoltaics,
catalysis,
high-entropy
alloys.
this
review,
basic
principles
are
briefly
outlined.
Several
commonly
algorithms
models
their
primary
applications
then
introduced.
The
predicting
properties
guiding
synthesis
discussed.
Finally,
future
outlook
on
materials
science
field
presented.
Journal of Chemical Information and Modeling,
Journal Year:
2024,
Volume and Issue:
64(3), P. 799 - 811
Published: Jan. 18, 2024
The
pursuit
of
designing
smart
and
functional
materials
is
paramount
importance
across
various
domains,
such
as
material
science,
engineering,
chemical
technology,
electronics,
biomedicine,
energy,
numerous
others.
Consequently,
researchers
are
actively
involved
in
the
development
innovative
models
strategies
for
design.
Recent
advancements
analytical
tools,
experimentation,
computer
technology
additionally
enhance
design
possibilities.
Notably,
data-driven
techniques
like
artificial
intelligence
machine
learning
have
achieved
substantial
progress
exploring
applications
within
science.
One
approach,
ChatGPT,
a
large
language
model,
holds
transformative
potential
addressing
complex
queries.
In
this
article,
we
explore
ChatGPT's
understanding
science
by
assigning
some
simple
tasks
subareas
computational
findings
indicate
that
while
ChatGPT
may
make
minor
errors
accomplishing
general
tasks,
it
demonstrates
capability
to
learn
adapt
through
human
interactions.
However,
issues
output
consistency,
probable
hidden
errors,
ethical
consequences
should
be
addressed.
Molecular Systems Design & Engineering,
Journal Year:
2021,
Volume and Issue:
6(12), P. 1066 - 1086
Published: Jan. 1, 2021
In
Bayesian
optimization,
we
efficiently
search
for
an
optimal
material
by
iterating
between
(i)
conducting
experiment
on
a
material,
(ii)
updating
our
knowledge,
and
(iii)
selecting
the
next
experiment.
Molecular Systems Design & Engineering,
Journal Year:
2022,
Volume and Issue:
7(6), P. 661 - 676
Published: Jan. 1, 2022
In
this
work,
we
present,
evaluate,
and
analyze
strategies
for
representing
polymer
chemistry
to
machine
learning
models
the
advancement
of
data-driven
sequence
or
composition
design
macromolecules.
Digital Discovery,
Journal Year:
2022,
Volume and Issue:
1(4), P. 355 - 374
Published: Jan. 1, 2022
Given
the
large
number
of
known
and
hypothetical
nanoporous
materials,
high-throughput
computational
screening
is
an
efficient
method
to
identify
current
best-performing
materials
guide
design
future
materials.
Journal of Materiomics,
Journal Year:
2022,
Volume and Issue:
8(5), P. 937 - 948
Published: April 28, 2022
A
material's
electronic
properties
and
technological
utility
depend
on
its
band
gap
value
the
nature
of
(i.e.
direct
or
indirect).
This
gaps
is
notoriously
difficult
to
compute
from
first
principles.
In
fact
it
computationally
intense
approximate
also
rather
time
consuming.
Hence
prediction
represents
a
challenging
problem.
Machine
learning
based
approach
offers
promising
efficient
means
address
this
Here
we
predict
for
perovskite
oxides
(ABO3)
with
elemental
composition,
ionic
radius,
character
electronegativity.
We
do
by
training
machine
models
generated
datasets.
Knowing
(whether
indirect)
plays
pivotal
role
in
determining
whether
can
be
used
photovoltaic
photocatalytic
applications.
total
5329
are
considered
study.
Here,
determine
correlation
between
composition
oxide.
Random
Forest
algorithm
predicting
same
since
yielded
higher
accuracy
(∼91%)
compared
other
Learning
models.
The
suggested
here
bandgap
aid
novel
materials
discovery
within
family
perovskites.
robust,
quick,
low-cost
strategy
find
light
harvesting
applications
particular.
Also
present
feature
ranking
as
pertains
discuss
features.
show
importance
graphs
SHapley
Additive
exPlanations
(SHAP)
relevant
gaps.
Using
reported,
NaPuO3
VPbO3
discovered
good
candidates
solar
cell
(direct
gap∼1.5
eV).
Novel
predictions
targeted
future
our
model
step
ahead
direction.
