2021 18th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP),
Год журнала:
2023,
Номер
unknown, С. 1 - 5
Опубликована: Дек. 15, 2023
Machine
and
deep
learning
techniques
are
now
essential
for
many
different
kinds
of
solutions
in
areas,
applications,
businesses.
Analyzing
e-commerce
reviews
using
machine
can
motivate
a
company
to
make
more
wise
decisions
regarding
the
quality
services
it
offers.
But
issues
occur
if
there
is
insufficient
high-quality
data
training,
scalability,
upkeep
models.
To
address
these
important
issues,
we
suggest
data-centric
architecture
that
makes
use
blockchain
Interplanetary
File
System
(IPFS)-based
storage.
Our
provides
secure
cost-effective
storage
an
incentive
mechanism.
experiment
shows
as
dataset
gets
bigger,
Naïve
Bayes
(NB)
Support
Vector
(SVM)
models
get
2%
accurate,
Convolutional
Neural
Network
(CNN)
model
4%
Long
Short-Term
Memory
(LSTM)
3
%
accurate.
The
exhibited
higher
accuracy
rates
than
because
hidden
layers
possess
ability
extract
intricate
syntactic
features
from
review
text.
Journal of theoretical and applied electronic commerce research,
Год журнала:
2024,
Номер
19(3), С. 1984 - 1999
Опубликована: Авг. 1, 2024
E-businesses
often
face
challenges
related
to
customer
service
and
communication,
leading
increased
dissatisfaction
among
customers
potential
damage
the
brand.
To
address
these
challenges,
data-driven
AI-based
approaches
have
emerged,
including
predictive
analytics
for
optimizing
interactions
chatbots
powered
by
AI
NLP
technologies.
This
study
focuses
on
developing
a
hybrid
rule-based
extractive-based
chatbot
e-business,
which
can
handle
both
routine
complex
inquiries,
ensuring
quick
accurate
responses
improve
communication
problems.
The
QA
method
used
in
demonstrated
high
precision
accuracy
providing
answers
user
queries.
approach
achieved
impressive
98%
97%
rates
1684
received
positive
feedback,
with
91%
of
users
rating
it
as
“good”
or
“excellent”
an
average
satisfaction
score
4.38.
General
was
notably
high,
Likert
4.29,
54%
participants
gave
highest
5.
Communication
time
significantly
improved,
reduced
response
times
41
s,
compared
previous
20-min
inquiries.
Advances in hospitality, tourism and the services industry (AHTSI) book series,
Год журнала:
2024,
Номер
unknown, С. 283 - 298
Опубликована: Окт. 18, 2024
The
integration
of
blockchain
technology
into
the
hospitality
industry
represents
a
significant
opportunity
to
enhance
operational
efficiency
and
improve
guest
experiences.
Blockchain,
decentralized
digital
ledger
offers
transparency
in
security
immutability
data
making
it
well-suited
for
applications
hospitality.
chapter
explores
various
including
booking
reservation
systems,
payment
solutions,
identity
verification,
supply
chain
management,
loyalty
programs
reviews.
By
leveraging
businesses
can
streamline
processes,
reduce
costs
provide
more
personalized
services
guests.
However,
adoption
also
presents
challenge
technical
scalability
issues,
regulatory
uncertainties
complexities.
Collaborative
efforts
between
stakeholders,
providers
bodies
will
be
essential
overcoming
these
challenges
unlocking
transformative
power
Brilliance Research of Artificial Intelligence,
Год журнала:
2023,
Номер
3(2), С. 252 - 261
Опубликована: Ноя. 17, 2023
The
increasing
trend
of
e-commerce
users
has
not
been
matched
by
customer
satisfaction
in
the
shopping
process.
Indonesia
highest
level
dissatisfaction
compared
to
other
ASEAN
countries.
Although
chatbot
technology
used
as
an
aid
optimize
services,
still
occurs
with
regard
agility,
service
assurance,
reliability,
scalability
and
security.
purpose
this
study
is
determine
services
providing
satisfaction.
research
approach
uses
quantitative
explantory
survey
method.
population
online
shop
using
rondom
sampling,
175
respondents
were
collected.
Assisted
PLS
SEM
analysis
tool.
results
show
that
social
orientation
contribute
Likewise,
personification
makes
a
positive
contribution
2021 18th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP),
Год журнала:
2023,
Номер
unknown, С. 1 - 5
Опубликована: Дек. 15, 2023
Machine
and
deep
learning
techniques
are
now
essential
for
many
different
kinds
of
solutions
in
areas,
applications,
businesses.
Analyzing
e-commerce
reviews
using
machine
can
motivate
a
company
to
make
more
wise
decisions
regarding
the
quality
services
it
offers.
But
issues
occur
if
there
is
insufficient
high-quality
data
training,
scalability,
upkeep
models.
To
address
these
important
issues,
we
suggest
data-centric
architecture
that
makes
use
blockchain
Interplanetary
File
System
(IPFS)-based
storage.
Our
provides
secure
cost-effective
storage
an
incentive
mechanism.
experiment
shows
as
dataset
gets
bigger,
Naïve
Bayes
(NB)
Support
Vector
(SVM)
models
get
2%
accurate,
Convolutional
Neural
Network
(CNN)
model
4%
Long
Short-Term
Memory
(LSTM)
3
%
accurate.
The
exhibited
higher
accuracy
rates
than
because
hidden
layers
possess
ability
extract
intricate
syntactic
features
from
review
text.
