Journal of Robotics and Automation Research,
Год журнала:
2023,
Номер
4(2)
Опубликована: Июль 3, 2023
This
survey
discusses
the
concept
of
knowledge
graphs,
including
their
construction,
extraction,
and
applications.Various
tools
such
as
Zotero,
Web
Science,
Google
Scholar,
EndNote,
VosViewer
are
used
to
analyze
visualize
collected
data.A
Boolean
query
mechanism
ensures
gathered
material
is
relevant
study.The
discussion
includes
studies
on
relation
extraction
using
graph
neural
networks,
application
graphs
in
biomedical
research,
use
embedding
healthcare.These
highlight
growing
importance
managing
representing
complex
information.Notable
discussed
include
role
connecting
related
medical
information,
technology
healthcare,
potential
benefits
limitations
data
analysis.This
paper
provides
valuable
insights
into
information
how
they
can
help
provide
new
various
fields.It
suggests
future
directions
for
research
this
area,
highlighting
continued
exploration
innovation
realize
fully.
Journal of Robotics and Automation Research,
Год журнала:
2023,
Номер
4(2)
Опубликована: Июнь 23, 2023
This
survey
discusses
the
concept
of
knowledge
graphs,
including
their
construction,
extraction,
and
applications.
Various
tools
such
as
Zotero,
Web
Science,
Google
Scholar,
EndNote,
VosViewer
are
used
to
analyze
visualize
collected
data.
A
Boolean
query
mechanism
ensures
gathered
material
is
relevant
study.
The
discussion
includes
studies
on
relation
extraction
using
graph
neural
networks,
application
graphs
in
biomedical
research,
use
embedding
healthcare.
These
highlight
growing
importance
managing
representing
complex
information.
Notable
discussed
include
role
connecting
related
medical
information,
technology
healthcare,
potential
benefits
limitations
data
analysis.
paper
provides
valuable
insights
into
information
how
they
can
help
provide
new
various
fields.
It
suggests
future
directions
for
research
this
area,
highlighting
continued
exploration
innovation
realize
fully
Concurrency and Computation Practice and Experience,
Год журнала:
2024,
Номер
36(21)
Опубликована: Июнь 25, 2024
Summary
Existing
research
on
recommender
systems
primarily
focuses
improving
a
single
objective,
such
as
prediction
accuracy,
often
ignoring
other
crucial
aspects
of
recommendation
performance
temporal
factor,
user
satisfaction,
and
acceptance.
To
solve
this
problem,
we
proposed
an
explicable
model
using
many‐objective
optimization
time‐assisted
knowledge
graph,
which
utilizes
interaction
times
within
the
graph
to
prioritize
recommending
recently
frequently
visited
items
is
further
optimized
algorithm.
In
model,
weight
actions
at
different
first
determined
through
time
decay
function.
Additionally,
if
clicks
same
item
again,
current
action's
set
one.
This
strategy
prioritizes
recent
items,
reflecting
interests
preferences
better.
Next,
created
used
create
list
potential
recommendations.
Embedding
methods
obtain
vectors
for
entities
relations
in
path.
These
vectors,
combined
with
actions,
quantify
explainability
Optimizing
rest
many
objective
algorithms
while
focusing
user's
frequent
visits
item.
Finally,
outcomes
study
indicate
that,
compared
recommended
methods,
our
considering
improved
average
accuracy
by
11%,
diversity
1%,
21%
Useraction1
data
set.
Results
sets
also
that
maintains
diversity,
novelty
enhancing
explainability.
<p>This
survey
discusses
the
concept
of
knowledge
graphs,
including
their
construction,
extraction,
and
applications.
Various
tools
such
as
Zotero,
Web
Science,
Google
Scholar,
EndNote,
VosViewer
are
used
to
analyze
visualize
collected
data.
A
Boolean
query
mechanism
ensures
gathered
material
is
relevant
study.
The
discussion
includes
studies
on
relation
extraction
using
graph
neural
networks,
application
graphs
in
biomedical
research,
use
embedding
healthcare.
These
highlight
growing
importance
managing
representing
complex
information.
Notable
discussed
include
role
connecting
related
medical
information,
technology
healthcare,
potential
benefits
limitations
data
analysis.
This
paper
provides
valuable
insights
into
information
how
they
can
help
provide
new
various
fields.
It
suggests
future
directions
for
research
this
area,
highlighting
continued
exploration
innovation
realize
fully.</p>
This
survey
discusses
the
concept
of
knowledge
graphs,
including
their
construction,
extraction,
and
applications.
Various
tools
such
as
Zotero,
Web
Science,
Google
Scholar,
EndNote,
VosViewer
are
used
to
analyze
visualize
collected
data.
A
Boolean
query
mechanism
ensures
gathered
material
is
relevant
study.
The
discussion
includes
studies
on
relation
extraction
using
graph
neural
networks,
application
graphs
in
biomedical
research,
use
embedding
healthcare.
These
highlight
growing
importance
managing
representing
complex
information.
Notable
discussed
include
role
connecting
related
medical
information,
technology
healthcare,
potential
benefits
limitations
data
analysis.
paper
provides
valuable
insights
into
information
how
they
can
help
provide
new
various
fields.
It
suggests
future
directions
for
research
this
area,
highlighting
continued
exploration
innovation
realize
fully.
A
Knowledge
Graph-based
Recommendation
System
(KG-RS)
employs
a
knowledge
graph
to
represent
data
and
generate
precise
recommendations
for
customers
based
on
the
given
information.
In
this
paper,
we
first
investigate
various
filtering
techniques
commonly
utilized
in
recommendation
systems
analyze
distinctions
between
Heterogeneous
Information
Networks
(HINs)
Graphs
(KGs).
Then,
classify
models
their
embedding
methods,
loss
functions,
entity
representations,
integration
of
additional
Also,
classified
extra
they
used
facilitate
research.
Our
research
demonstrates
that
item
information
is
consistently
included
graphs,
while
user
not.
Additionally,
KG-RSs
are
progressing
by
incorporating
more
advanced
into
process,
rather
than
complicating
itself.
Engineering Reports,
Год журнала:
2024,
Номер
unknown
Опубликована: Янв. 28, 2024
Abstract
In
recent
years,
graph
neural
networks
(GNNs)
have
showcased
a
strong
ability
to
learn
representations
and
been
widely
used
in
various
practical
applications.
However,
many
currently
proposed
GNN‐based
representation
learning
methods
do
not
retain
neighbor‐based
node
similarity
well,
this
structural
information
is
crucial
cases.
To
address
issue,
drawing
inspiration
from
generative
adversarial
(GANs),
we
propose
PNS‐AGNN
(i.e.,
Preserving
Node
Similarity
Adversarial
Graph
Neural
Networks),
novel
framework
for
acquiring
representations,
which
can
preserve
of
the
original
efficiently
extract
nonlinear
features
graph.
Specifically,
new
positive
sample
allocation
strategy
based
on
index,
where
generator
generate
vector
that
satisfy
through
training.
addition,
also
adopt
an
improved
GNN
as
discriminator,
utilizes
structure
recursive
neighborhood
aggregation
maintain
local
feature
nodes,
thereby
enhancing
representation's
ability.
Finally,
experimentally
demonstrate
significantly
improves
tasks,
including
link
prediction,
classification,
visualization.
<p>This
survey
discusses
the
concept
of
knowledge
graphs,
including
their
construction,
extraction,
and
applications.
Various
tools
such
as
Zotero,
Web
Science,
Google
Scholar,
EndNote,
VosViewer
are
used
to
analyze
visualize
collected
data.
A
Boolean
query
mechanism
ensures
gathered
material
is
relevant
study.
The
discussion
includes
studies
on
relation
extraction
using
graph
neural
networks,
application
graphs
in
biomedical
research,
use
embedding
healthcare.
These
highlight
growing
importance
managing
representing
complex
information.
Notable
discussed
include
role
connecting
related
medical
information,
technology
healthcare,
potential
benefits
limitations
data
analysis.
This
paper
provides
valuable
insights
into
information
how
they
can
help
provide
new
various
fields.
It
suggests
future
directions
for
research
this
area,
highlighting
continued
exploration
innovation
realize
fully.</p>
