Download 23)Dr-Bhavani-presentation-AFCEA2006

Data Mining
for
Security Applications
Dr. Bhavani Thuraisingham
The University of Texas at Dallas
January 2006
5/22/2017 15:33
2
Aspects of Counterterrorism
0 Types of threats
- Non-real-time Threats / Real-time Threats
- Cyber terrorism
- Bio terrorism
0 What needs to be protected?
- Services
= Transportation, Financial, Medical, - - - Infrastructures
= Telecommunication networks, Power systems, water
supply/tanks/reservoirs
- Information related
= Computing systems and networks, National
databases, Financial databases., Medical databases, - -
5/22/2017 15:33
3
What is Data Mining?
The process of discovering meaningful new correlations, patterns
and trends, often previously unknown, by sifting through large
amounts of data, using pattern recognition, statistical and
mathematical techniques
Information Harvesting
Data Mining
Siftware
Knowledge Discovery
in Databases
Data Pattern Processing
Knowledge Mining
Data Archaeology
Knowledge Extraction
Data Dredging
Database Mining
5/22/2017 15:33
4
Steps to Data Mining
Integrate
data
sources
Data sources
Clean/
modify
data
sources
Report/
evaluate
results
Mine
the data
Examine/
prune
results
The cycle may continue; add new data, use different algorithms
5/22/2017 15:33
5
What’s going on in data mining?
0 What are the technologies for data mining?
- Database management, machine learning, statistics, pattern
recognition, visualization, parallel processing, . . .
0 What can data mining do for you?
- Data mining outcomes: Classification, Clustering, Association,
Anomaly detection, Prediction, Estimation, . . .
0 How do you carry out data mining?
- Data mining techniques: Decision trees, Neural networks,
Market-basket analysis, Genetic algorithms, . . .
0 What is the current status?
- Many commercial products mine relational databases
0 What are some of the challenges?
- Mining unstructured data, extracting useful patterns, web mining
5/22/2017 15:33
6
Data Mining Needs for Counterterrorism:
Non-real-time Data Mining
0 Gather data from multiple sources
- Information on terrorist attacks: who, what, where, when, how
- Personal and business data: place of birth, ethnic origin,
religion, education, work history, finances, criminal record,
relatives, friends and associates, travel history, . . .
- Unstructured data: newspaper articles, video clips, speeches,
emails, phone records, . . .
0 Integrate the data, build warehouses and federations
0 Develop profiles of terrorists, activities/threats
0 Mine the data to extract patterns of potential terrorists and predict
future activities and targets
0 Find the “needle in the haystack” - suspicious needles?
0 Data integrity is important
0 Techniques have to SCALE
5/22/2017 15:33
7
Data Mining Needs for Counterterrorism:
Real-time Data Mining
0 Nature of data
- Data arriving from sensors and other devices
= Continuous data streams
- Breaking news, video releases, satellite images
- Some critical data may also reside in caches
0 Rapidly sift through the data and discard unwanted data for later use
and analysis (non-real-time data mining)
0 Data mining techniques need to meet timing constraints
0 Quality of service (QoS) tradeoffs among timeliness, precision and
accuracy
0 Presentation of results, visualization, real-time alerts and triggers
5/22/2017 15:33
8
Data Mining Needs for Counterterrorism:
Cybersecurity
0 Determine nature of threats and vulnerabilities
- e.g., emails, trojan horses and viruses
0 Classify and group the threats
0 Profiles of potential cyberterrorist groups and their capabilities
0 Data mining for intrusion detection
- Real-time/ near-real-time data mining
- Limit the damage before it spreads
0 Data mining for preventing future attacks
0 Data mining for Digital forensics and Biometrics
5/22/2017 15:33
9
Are general data mining
techniques sufficient?
0 Does one size fit all?
- Non real-time, real-time, cyber, bio?
0 What are the major differences
- e.g., develop models ahead of time for real-time data mining?
- What happens in a very dynamic environment?
0 Data mining tasks/outcomes
- Classification, clustering, associations, anomaly detection,
prediction - - - -?
0 Data mining techniques
- Which techniques are good for which problems?
5/22/2017 15:33
10
Current Status and Challenges
0 Where are we now?
- building data warehouses from structured data
- integrating structured heterogeneous databases
- mining structured data
- forming some links and associations
- image processing and analysis
0 What are our Challenges?
- Scalability for petabyte sized databases?
- Integrating structured data with unstructured data
- Mining unstructured data
- Extracting useful patterns from knowledge-directed data mining
- Rapidly forming links and associations
- Mining the web
5/22/2017 15:33
11
Form a Research Agenda
0 Immediate action (0 - 1 year)
- We’ve got to know what our current capabilities are
- Do the commercial tools scale? Do they work only on special
data and limited cases? Do they deliver what they promise?
- Need an unbiased objective study with demonstrations
0 At the same time, work on the big picture
- What do we want? What are our end results for the foreseeable
future? What are the criteria for success? How do we evaluate
the data mining algorithms? What testbeds do we build?
0 Near-term (1 - 3 years)
- Leverage current research
- Fill the gaps in a goal-directed way
0 Long-term research (3 - 5 years and beyond)
- 5-year basic research plan for data mining for counterterrorism
5/22/2017 15:33
12
IN SUMMARY:
0 Data Mining is very useful to solve Security Problems
- Data mining tools could be used to examine audit data
and flag abnormal behavior
- Much recent work in Intrusion detection
= e.g., Neural networks to detect abnormal patterns
- Tools are being examined to determine abnormal patterns
for national security
= Classification techniques, Link analysis
- Fraud detection
= Credit cards, calling cards, identity theft etc.
BUT CONCERNS FOR PRIVACY
5/22/2017 15:33
13
Some Privacy concerns
0 Medical and Healthcare
- Employers, marketers, or others knowing of private medical
concerns
0 Security
- Allowing access to individual’s travel and spending data
- Allowing access to web surfing behavior
0 Marketing, Sales, and Finance
- Allowing access to individual’s purchases
5/22/2017 15:33
14
Data Mining as a Threat to Privacy
0 Data mining gives us “facts” that are not obvious to human analysts
of the data
0 Can general trends across individuals be determined without
revealing information about individuals?
0 Possible threats:
- Combine collections of data and infer information that is private
= Disease information from prescription data
= Military Action from Pizza delivery to pentagon
0 Need to protect the associations and correlations between the data
that are sensitive or private
5/22/2017 15:33
15
Some Privacy Problems
and Potential Solutions
0 Problem: Privacy violations that result due to data mining
- Potential solution: Privacy-preserving data mining
0 Problem: Privacy violations that result due to the Inference problem
- Inference is the process of deducing sensitive information from
the legitimate responses received to user queries
- Potential solution: Privacy Constraint Processing
0 Problem: Privacy violations due to un-encrypted data
- Potential solution: Encryption at different levels
0 Problem: Privacy violation due to poor system design
- Potential solution: Develop methodology for designing privacyenhanced systems
5/22/2017 15:33
16
Data Mining and Privacy: Friends or Foes?
0 They are neither friends nor foes
0 Need advances in both data mining and privacy
0 Need to design flexible systems
- For some applications one may have to focus entirely on “pure”
data mining while for some others there may be a need for
“privacy-preserving” data mining
- Need flexible data mining techniques that can adapt to the
changing environments
0 Technologists, legal specialists, social scientists, policy makers and
privacy advocates MUST work together
5/22/2017 15:33
17
Ideas and Directions?
Prof. Bhavani Thuraisingham
- Director Cyber Security Center
- Department of Computer Science
- Erik Jonsson School of Engineering and Computer Science
- The University of Texas at Dallas
- Richardson, Texas
- [email protected]
http://www.utdallas.edu/~bxt043000/
President
Bhavani Security Consulting
Dallas, TX
www.dr-bhavani.org