Transcript progress report

TRAKS: Terrorist Related
Assessment using Knowledge
Similarity
Boanerges Aleman-Meza
Chris Halaschek
Satya Sanket Sahoo
CSCI 8350 – Enterprise Integration
Semantic Web Course
Large Scale Distributed Information Systems
(LSDIS) Lab
1
Motivation
• identification of contributing factors to
terrorist activities
– money laundering
– identity theft
– terrorist planning
• 9/11 terrorist attacks
– claimed around 3000 lives [1],
– caused estimated “$120 billion of damage”
[2]
2
Current Approaches
• Data mining techniques
• Rule-based mechanisms
• Structural match of known rules/patters
3
Our Approach
• Semantic Similarity Template Match
– Capture the knowledge with OWL
– Define an ontology for money laundering, …
– Case Scenario modeled as a template
– “fill-in” the template with instance data 
scenario found ! notify the police !
4
Where are the
Semantics?
• Our system uses OWL
• The template is defined with OWL and our
‘template’ and ‘core-template’ classes
• Instance data is defined in OWL
• This allows our code to work:
– with anyone's ontology/instance/template
5
Where is the Semantic
similarity?
• Template example:
– [CEO] transfers Money To [InternationalBank]
• Similarity based on the ontology:
– [CTO] deposits Check In [DomesticBank]
6
Three components:
– Ontology:
– Data Set:
– Template:
7
Templates
[Person]
[Company]
[Country]
[City]
[Country]
[RealState]
[Loan]
[Bank]
8
So far…
•
•
•
•
•
•
•
Website
Templates
Servlet engine setup
Servlet implementation
Data Sets
ontology
Code prototype – step 1 (core match)
9
References
[1] The Economic Cost of Terrorism
[2] September 11, 2001: A day of terror
[3] Financial Action Task Force on Money Laundering Homepage
10
Responsabilities
•
•
•
•
•
•
Website: Chris
Templates: Aleman, Chris, Satya
Servlet engine setup: Aleman
Servlet implementation: Satya
Testbed, ontology: Aleman
Code prototype: Chris
11