Graph
Neural
Networks
(GNNs)
have
emerged
as
a
powerful
tool
for
analyzing
and
modeling
graph-structured
data.
In
recent
years,
GNNs
gained
significant
attention
in
various
domains.
This
review
paper
aims
to
provide
an
overview
of
the
state-of-the-art
graph
neural
network
techniques
their
industrial
applications.First,
we
introduce
fundamental
concepts
architectures
GNNs,
highlighting
ability
capture
complex
relationships
dependencies
We
then
delve
into
variants
evolution
graphs,
including
directed
heterogeneous
dynamic
hypergraphs.
Next,
discuss
interpretability
GNN,
GNN
theory
augmentation,
expressivity,
over-smoothing.Finally,
specific
use
cases
settings,
finance,
biology,
knowledge
recommendation
systems,
Internet
Things
(IoT),
distillation.
highlights
immense
potential
solving
real-world
problems,
while
also
addressing
challenges
opportunities
further
advancement
this
field.
ACM Computing Surveys,
Journal Year:
2024,
Volume and Issue:
56(9), P. 1 - 42
Published: May 8, 2024
Due
to
underlying
privacy-sensitive
information
in
user-item
interaction
data,
the
risk
of
privacy
leakage
exists
centralized-training
recommender
system
(RecSys).
To
this
issue,
federated
learning,
a
privacy-oriented
distributed
computing
paradigm,
is
introduced
and
promotes
crossing
field
“Federated
Recommender
System
(FedRec).”
Regarding
data
distribution
characteristics,
there
are
horizontal,
vertical,
transfer
variants,
where
horizontal
FedRec
(HFedRec)
occupies
dominant
position.
User
devices
can
personally
participate
architecture,
making
user-level
feasible.
Therefore,
we
target
point
summarize
existing
works
more
elaborately
than
surveys.
First,
from
model
perspective,
group
them
into
different
learning
paradigms
(e.g.,
deep
meta
learning).
Second,
privacy-preserving
techniques
systematically
organized
homomorphic
encryption
differential
privacy).
Third,
fundamental
issues
communication
fairness)
discussed.
Fourth,
each
perspective
has
detailed
subcategories,
specifically
state
their
unique
challenges
with
observation
current
progress.
Finally,
figure
out
potential
promising
directions
for
future
research.
