sentinel
DESCRIPTION
Sentinel. Alun Preece Irena Spasi ć David Rogers Cardiff School of Computer Science & Informatics. Data-to-decisions. Data source. Analytic services. Decision maker. Data sources. - PowerPoint PPT PresentationTRANSCRIPT
SentinelAlun PreeceIrena SpasićDavid Rogers
Cardiff School of Computer Science & Informatics
Analytic services Decision maker
Data-to-decisions
Data sourceData sources
An open, flexible, scalable suite of technologies intended to support situation understanding and provide actionable intelligence from social data. Sentinel applies semantic models of crime and social reaction to data collected in real-time from a variety of social media sources. The data is analyzed using text mining techniques, enabling Sentinel to deliver interpretations of events via a customizable set of apps. Sentinel is a result of collaboration between the Universities Police Science Institute and the School of Computer Science & Informatics at Cardiff University.
Sentinel
Sentinel core services & models
Data collection servicesExpression & term recognition
Customizable apps
Semantic APIs: who, what, when, where, why
Signal Crimes Conflict
Extremist Narrative
s
5 Ws
Analytic servicesidentify significant
terms
Decision maker
“Bottom-up” issue identification
Data sourceData sources
Analytic servicesmatch data toontology terms
Decision maker
“Top-down” issue identification
Data sourceData sources
data + knowledge = information
we interpret text data using our knowledge of both language & world
data
unprocessed facts
no context or purposeful meaning
information
organized collection of facts
processed data that have meaning & context
information is a joint function of data & knowledge
Ontology
How can we represent knowledge?
ontology
machine readable knowledge representation
models concepts in a domain & their relationships
supports shared understanding between both humans & computers
supports reasoning about the domain
Concepts
a concept represents a class of entities within a domain
each concept is represented by:
ID name(s) definition type
Types
concepts are organized into a hierarchy using is–a (or kind–of) relationship
we can now search by type
... and navigate up & down the hierarchy
Relationships
associative relationships relate concepts across the type hierarchy
we can now search by associations
Current state
448 concepts
357 additional synonyms
121 associations
ontology will continue to evolve in order to:
expand the coverage of the domain
reflect the changes in the domain
Applications
semantic search
keyword to concept mapping
generalization
e.g.
query automatically expanded:improvised explosive device OR car bomb OR truck bomb OR explosive belt OR suicide vest ORpetrol bomb OR Molotov cocktail OR Molotov OR fire bomb OR pipe bomb
Applications
semantic interpretation
annotation
classification
qualitative analysis
Wonderful. Jewish and Muslim folks get together to protect Stoke Newington mosque from hate crime
SENTINEL:0000089:hate crime
SENTINEL:0000119:MuslimSENTINEL:0000121:Jewish
SENTINEL:0000211:mosque
Applications
inference through machine learning
ontology supports features based on meaning (not just words)
infer meaning based on annotated concepts
@Official_EDL: EDL leader Tommy Robinson on way to Woolwich now, Take to the streets peeps ENOUGH IS ENOUGH
stance: hard support
subject side: far right extremism
routine: reacting
dynamic: mobilising
Thanks for listening!