Published June 13, 2023, 10:20 a.m. by Courtney
You may also like to read about:
this is the 2012 film ibm project
demonstration i'm your chainsaw a phd
student from kdec trained college dublin
rob brennan is also involved in this
work
in this year's demonstration our
semantic application engine is located
with open flow management to support the
quality of experience based iptv network
monitoring through our visual widgets
our semantic uplift approach com
combines network logs
network models and domain expert
knowledge together to simplify the iptv
service delivery network monitoring
we use semantic 10 technologies to
capture and model the dome expert
knowledge
in this year's demo we made three
enhancements on our approach
compose event temporal reasoning and
enhanced widgets
this figure shows how our semantic
athlete uplift approach works
our approach takes multi-network log
inputs
annotate systematic concepts to the raw
data
maintain annotated information into an
entity port
and then aggregate this information to
support root cause analysis for network
anomalies
this is the visual interface of
last year's demo in which we support the
status and real-time event monitoring
and also the knowledge-driven anomaly
root cause analysis
the three enhancements this year we
applied on our layered visualization
framework
in the
information uplifting layer semantic
concepts are annotated to the
streamlocked data
according to the domain experts insight
this annotated information is maintained
and upgraded
in the semantic processing layer
we enhance the reasoning capability for
compose and temporal events here
this uplifted information is presented
through our enhanced widgets
in the virtual dish visual
representation layer to spot better
understanding for iptv network
monitoring
we will also use a diamond to show all
these
new enhancements with the scenario too
in this scenario a central router in
connect work was offline
which made
network traffic overflow on the other
two routers
and causes the low quality of the
experience problem for all ftm network
users
this is our visual interface
it's a web-based in
interface made by flags and its
communicates with our information
eclipse engine through blades middleware
the expert knowledge is modeled by
semantic anthologies and rules the
reasoning is driven by gender
this panel is quality of
experience panel
this is all
in this panel we can see the status of
different user groups
red label means bad condition and green
label means
normal condition
if we click the source domain user group
we can see the users in this group and
its network context
and after we click a particular user
we can know the activity service details
and the network contacts for this
service
here our approach enables the reasoning
cross composing event from
quality of experience iptv service and
network status
after we click the contacts label
the context of this iptv service is
showing in the network context panel
we can see this activity service goes
through video server code network
dislike and gateway
this is a very abstract view if we want
to know more detail we can drill down
particular node and
like a code network
and its
sub content is displayed through our
multi-resolution network content
contacts panel
the real time normally is reported as a
red bubble in the anomaly panel
but normally here means the event
affects the
quality of experience for network users
if we click particularly normally
in the anomaly analysis panel we can see
the analysis results of this anomaly
in this scenario the low quality of
experience normally is caused by an iptv
locality problem happens on all iptv
services
and our next step we performed a
track back analysis
because code network is the last node in
the network has bad status
and all nodes after it has
has ballistic
has bad condition too
we informed the code network probably is
the reason caused the low quality
problem
this reasoning across the network logs
and service logs
which is benefited by our new composing
event reasoning capability
for the code network
there are two
problems
one router is offline and the other two
router has
overflow problem
but the overflow problem happened after
the offline problem so according to our
new temporal reasoning capability we can
infer the root reason for this low qe
problem is caused by central rotor 3
it's offline
the suggested solution is to restart
this router
this is the demo of this year's visual
interface in this demo we showed how our
semantic information uplift approach and
its three new features
composing that temporal reasoning and
enhanced widget
supports better qe based fptv network
monitoring
2CUTURL
Created in 2013, 2CUTURL has been on the forefront of entertainment and breaking news. Our editorial staff delivers high quality articles, video, documentary and live along with multi-platform content.
© 2CUTURL. All Rights Reserved.