2021 14th International Conference on Developments in eSystems Engineering (DeSE),
Год журнала:
2023,
Номер
unknown, С. 870 - 875
Опубликована: Дек. 18, 2023
Large
Language
Models
(LLMs)
have
advanced
significantly
in
Natural
Processing
(NLP)
over
the
past
few
years.
Ongoing
research
continues
exploring
their
capabilities
recommendation
systems,
aiming
to
enhance
user-tailored
content
delivery
efficiency,
accuracy,
and
personalisation.
The
investigation
introduces
a
novel
approach
integration
possibilities
of
open-source
Model
(LLM)
technology—FLAN-T5,
Falcon,
Vicuna,
UL2,
LLAMA—into
anime
systems.
delves
into
creating
personalised
recommendations
by
inputting
titles,
genres,
descriptions
these
LLMs.
Furthermore,
it
harnesses
LLMs
explain
recommendations,
bolstering
user
engagement
amplifying
transparency
process.
findings
clearly
show
that
using
for
works
well.
It
proves
techniques
great
potential
make
suggestions
better.
Artificial Intelligence Review,
Год журнала:
2024,
Номер
58(1)
Опубликована: Ноя. 16, 2024
Abstract
Recommender
systems
are
software
mechanisms
whose
usage
is
to
offer
suggestions
for
different
types
of
entities
like
products,
services,
or
contacts
that
could
be
useful
interesting
a
specific
user.
Other
ways
have
been
explored
in
the
field
enhance
power
these
by
integrating
context
as
an
additional
attribute.
This
inclusion
tries
extract
user
preferences
more
accurately
taking
into
account
multiple
components
such
temporal,
spatial,
social
ones.
Notwithstanding
magnitude
context-awareness
this
area,
research
community
agreement
with
lack
framework
information
and
how
integrate
it
recommender
systems.
Under
premise,
paper
focuses
on
comprehensive
systematic
literature
review
state-of-the-art
recommendation
techniques
their
characteristics
benefit
from
contextual
information.
The
following
survey
presents
contributions
outcomes
our
study:
(i)
determine
where
aspects
taken
clear
definition
representation,
(ii)
used
incorporate
context,
(iii)
evaluation
methods
terms
reproducibility
effectiveness.
Our
also
covers
some
crucial
topics
about
integration,
classification
contexts,
application
domains,
datasets,
metrics,
code
implementations,
we
observed
shiftings
algorithmic
trends
towards
Neural
Network
approaches
ranking
respectively.
Just
importantly,
future
opportunities
directions
exposed
final
closure,
standing
out
exploitation
various
data
sources
scalability
customization
existing
solutions.
Generative
artificial
intelligence
(AI),
in
particular
large
language
models
such
as
ChatGPT
have
reached
public
consciousness
with
a
wide-ranging
discussion
of
their
capabilities
and
suitability
for
various
professions.
Following
the
printing
press
internet,
generative
AI
are
third
transformative
technological
invention
truly
cross-sectoral
impact
on
knowledge
transmission.
While
allowed
transmission
that
is
independent
physical
presence
holder
publishers
acting
gatekeepers,
internet
added
levels
democratization
allowing
anyone
to
publish,
along
global
immediacy.
The
development
social
media
resulted
an
increased
fragmentation
tribalization
on-line
communities
ways
knowing,
resulting
alternative
truths
propagated
echo
chambers.
It
against
this
background
entered
consciousness.
Using
strategic
foresight
methodology,
paper
will
examine
polemic
proposition
age
emerge
ignorance.
Environment Technology Resources Proceedings of the International Scientific and Practical Conference,
Год журнала:
2024,
Номер
2, С. 179 - 182
Опубликована: Июнь 22, 2024
The
release
of
ChatGPT
technology
identified
the
large
language
models
as
a
new
disruptive
technology,
which
changes
behaviours
society
and
its
attitude
towards
presence
artificial
intelligence
in
everyday
life.
tourism
industry
is
one
economic
sectors,
will
be
impacted
by
through
personalized
marketing
advertisements.
A
common
approach
to
capture
attention
AI-centric
tourists,
who
want
get
answers
their
questions
without
manually
researching
topic
or
using
services
travel
advisors,
integrate
chatbot
virtual
assistant
information
system.
We
applied
this
promotion
East
Latvia
(Latgale)
rapidly
developing
prompt
method
with
context-oriented
material.
Two
were
prepared
for
Latgale.
evaluated
pilot
survey
understand
satisfaction
target
users.
data
analysis
was
applied.
study
importance
trustworthy
answer
saturation.
trade-off
between
dialog
freedom
trustworthiness
can
achieved
development
microservices,
are
grouped
system
direct
conversation
chatbot.
appropriate
conceptual
presented
article.
In
the
realm
of
e-commerce
customer
support,
adoption
chatbots
is
on
rise,
driven
by
a
quest
for
heightened
user
interactions.
This
study
introduces
an
inventive
approach
harnessing
advanced
capabilities
GPT-4
to
construct
chatbot
interfaces
that
are
both
context-aware
and
personalized.
The
main
aim
this
work
revolutionize
ecommerce
support
developing
intelligent
adaptable
context-aware,
personalized,
significantly
enhance
experiences.
Existing
models
struggle
with
context
retention
personalization,
limiting
effectiveness
in
dynamic
environments
despite
promise
automation.
innovative
leverages
GPT-4's
robust
language
processing,
integrating
it
profiling
systems
personalized
responses
grounded
preferences
historical
method
gives,
diverse
dataset
undergoes
collection
pre-processing,
followed
fine-tuning
emphasis
context-awareness
responses.
model
seamlessly
integrated
backend
real-time
information,
multimodal
input
caters
varied
preferences.
Results
exhibit
significant
95%
accuracy
improvement,
affirming
chatbot's
enhanced
ability
comprehend
queries.
By
personalization
retention,
enhances
engagement,
paving
way
revolutionary
support.
