Handouts
– 1 –
Data Mining for Analysis of Rare Events:
A Case Study in Security, Financial and
Medical Applications
Aleksandar Lazarević,
Jaideep Srivastava, Vipin Kumar
Army High Performance Computing Research Center
Department of Computer Science
University of Minnesota
PAKDD-2004 Tutorial
IntroductionIntroduction
We are drowning in the deluge of
data that are being collected
world-wide, while starving for
knowledge at the same time*
Despite the enormous amount of
data, particular events of interest
are still quite rare
Rare events are events that occur
very infrequently, i.e. their
frequency ranges from 0.1% to
less than 10%
However, when they do occur, their consequences can be
quite dramatic and quite often in a negative sense
“Mining needle in a haystack.
So much hay and so little time”
* - J. Naisbitt, Megatrends: Ten New Directions Transforming Our Lives. New York: Warner Books, 1982.
Handouts
– 2 –
Related problemsRelated problems
Chance discovery
“chance is defined as some event which is significant
for decision-making in a specified domain” Yukio Ohsawa
chance event may be either positive (opportunity), or
negative (potential risk)
Chance Discovery – learning, explaining and
discovering such chance events, typically rare to find
Novelty Detection
Exception Mining
Black Swan* [Taleb’04]
* N. Talleb, The Black Swan: Why Don’t We Learn that We Don’t Learn?, draft 2004
Applications of Rare ClassesApplications of Rare Classes
Network intrusion detection
number of intrusions
on the network is typically
a very small fraction of
the total network traffic
Credit card fraud detection
Millions of regular transactions are
stored, while only a very small
percentage corresponds to fraud
Medical diagnostics
When classifying the pixels in mammogram images, cancerous
pixels represent only a very small fraction of the entire image
Computer
Network
Compromised
Machine
Attacker
Handouts
– 3 –
Industrial Applications of Rare ClassesIndustrial Applications of Rare Classes
Insurance Risk Modeling (e.g. Pednault, Rosen, Apte ’00)
Claims are rare but very costly
Web mining
Less than 3% of all people visiting Amazon.com make a purchase
Targeted Marketing (e.g. Zadrozny, Elkan ’01)
Response is typically rare but can be profitable
Churn Analysis (e.g. Mamitsuka and Abe ’00)
Churn is typically rare but quite costly
Hardware Fault Detection (e.g. Apte, Weiss, Grout 93)
Faults are rare but very costly
Airline No-show Prediction (e.g. Lawrence, Hong, et al ’03)
Disease is typically rare but can be deadly
Limitations of Standard Data Mining SchemesLimitations of Standard Data Mining Schemes
Standard approaches for feature selection and
construction do not work well for rare class analysis
While most normal events are similar to each other, rare
events are quite different from one another
regular credit transaction are fairly standard, while fraudulent
ones vary from the standard ones in many different ways
Metrics used to evaluate normal event detection methods
Accuracy is not appropriate for evaluating methods for rare event
detection
In many applications data keeps arriving in an ongoing
stream, and there is a need to detect rare events on the
fly, with models built only on the events seen so far
Handouts
– 4 –
Evaluation of Rare Class ProblemsEvaluation of Rare Class Problems –– FF--valuevalue
Accuracy is not sufficient metric for evaluation
Example: network traffic data set with 99.9% of normal data and
0.1% of intrusions
Trivial classifier that labels everything with the normal class can
achieve 99.9% accuracy !!!!!
P r e d ic te d
c la ss
C o n fu sio n
m a tr ix
N C C
N C T N F PA c tu a l
c la ss C F N T P
rare class – C
normal class – NC
• Focus on both recall and precision
– Recall (R) = TP/(TP + FN)
– Precision (P) = TP/(TP + FP)
• F – measure = 2*R*P/(R+P)
Evaluation of Rare Class ProblemsEvaluation of Rare Class Problems -- ROCROC
Standard measures for evaluating rare class problems:
Detection rate (Recall) - ratio between the number of correctly
detected rare events and the total number of rare events
False alarm (false positive) rate – ratio
between the number of data records
from majority class that are misclassified
as rare events and the total number of
data records from majority class
ROC Curve is a trade-off between
detection rate and false alarm rate
P r e d ic te d
c la s s
C o n fu sio n
m a tr ix
N C C
N C T N F PA c tu a l
c la s s C F N T P
rare class – C
normal class – NC
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
ROC curves for different outlier detection techniques
False alarm rate
Detectionrate
Ideal
ROC
curve
Random prediction
Handouts
– 5 –
Evaluation of Rare Class ProblemsEvaluation of Rare Class Problems -- AUCAUC
P r e d ic te d
c la s s
C o n fu sio n
m a tr ix
N C C
N C T N F PA c tu a l
c la s s C F N T P
rare class – C
normal class – NC
0 0.02 0.04 0.06 0.08 0.1 0.12
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Detectiorate
False alarm rate
Area under Curve (AUC)
Area under the ROC curve
(AUC) is computed using a
form of the trapezoid rule.
Equivalent Mann-Whitney
two-sample statistics:
,),(
1ˆ
1 1
∑∑= =
+−
⋅
=
m
i
n
j
ji rrI
nm
A







=+−
0
2
1
1
),( rrI
if r - > r+
if r - = r+
if r - < r+
m ratings of negative cases r –
n ratings of positive cases r +
Example: in naïve Bayes, rating may be the
posterior probability of the positive class
Major Techniques for Detecting Rare EventsMajor Techniques for Detecting Rare Events
Unsupervised techniques
Deviation detection, outlier analysis, anomaly detection,
exception mining
Analyze each event to determine how similar (or dissimilar) it is
to the majority, and their success depends on the choice of
similarity measures, dimension weighting
Supervised techniques
Mining rare classes
Build a model for rare events based on labeled data (the
training set), and use it to classify each event
Advantage: they produce models that can be easily understood
Drawback: The data has to be labeled
Other techniques – association rules, clustering
Handouts
– 6 –
Unsupervised TechniquesUnsupervised Techniques ––
Anomaly DetectionAnomaly Detection
Build models of “normal” behavior and
detect anomalies as deviations from it
Possible high false alarm rate - previously unseen (yet
legitimate) data records may be recognized as
anomalies
Two types of techniques
with access to normal data
with NO access to normal data (not known what is “normal”)
False
alarm
Missed
rare events
Anomalous data records
Normal profile
Outlier Detection SchemesOutlier Detection Schemes
Outlier is defined as a data point which is very different from the
rest of the data based on some measure
Detect novel attacks/intrusions by identifying them as deviations
from “normal” behavior
Identify normal behavior
Construct useful set of features
Define similarity function
Use outlier detection algorithm
Statistics based approaches
Distance based approaches
Nearest neighbor approaches
Clustering based approaches
Density based schemes
Model based schemes
Handouts
– 7 –
Statistics Based Outlier Detection SchemesStatistics Based Outlier Detection Schemes
Statistics based approaches – data points are modeled
using stochastic distribution ⇒ points are determined
to be outliers depending on their relationship with this
model
With high dimensions, difficult to estimate distributions
Major approaches
Finite Mixtures
BACON
Using probability distribution
Information Theory measures
Statistics Based Outlier Detection SchemesStatistics Based Outlier Detection Schemes
Using Finite Mixtures – SmartSifter (SS)*
SS uses a probabilistic model as a representation of
underlying mechanism of data generation.
Histogram density used to represent a probability density for
categorical attributes
SDLE (Sequentially Discounting Laplace Estimation) for learning
histogram density for categorical domain
Finite mixture model used to represent a probability density
for continuous attributes
SDEM (Sequentially Discounting Expectatioan and Maximizing)
for learning finite mixture for continuous domain
SS gives a score to each example xi (input into the
model) on the basis of the learned model, measuring
how large the model has changed after the learning
* K. Yamanishi, On-line unsupervised outlier detection using finite mixtures with
discounting learning algorithms, KDD 2000
Handouts
– 8 –
Statistics Based Outlier Detection SchemesStatistics Based Outlier Detection Schemes
BACON* Basic Steps:
1. Select an initial basic subset of size m free of outliers
Initial subset selected based on Mahalanobis distances
Initial subset selected based on distances from the medians
2. Fit the model to the subset and for ∀xi compute the discrepancies
3. Set the new basic subset to all points with discrepancy less than
cnpr where is 1 − α percentile of the χ2 distribution with
p degrees of freedom, cnpr = cnp + chr is a correction factor,
chr = max{0, (h−r)/(h+r)}; h=[(n+p + 1)/2]; r is the size of the current
basic subset
4. Iterate steps 2&3 until the the size of basic subset no longer changes
5. List observations excluded by the final basic subset as outliers
* N. Billor, A. Hadi, P. Velleman, BACON: blocked adaptive computationally efficient
outlier nominators, Computational Statistics & Data Analysis, 34, 279-298, 2000.
)()(),( 1
bib
T
bibbi xxd µµµ −⋅Σ⋅−=Σ −
2
,αχp
2
,αχ p
phnpn
p
cnp
−−
+
−
+
+=
11
1
Statistics Based Outlier Detection SchemesStatistics Based Outlier Detection Schemes
Using Probability Distributions*
Basic Assumption: # of normal elements in the
data is significantly larger then # of anomalies
Distribution for the data D is given by:
D = (1-λ)·M + λ·A
M - majority distribution, A - anomalous distribution
Mt, At sets of normal, anomalous elements
respectively
Compute likelihood Lt(D) of distribution D at time t
Measure how likely each element xt is outlier:
Mt = Mt-1 \ {xt}, At = At-1 ∪ {xt}
Measure the difference (Lt – Lt-1)
* E. Eskin, Anomaly Detection over Noisy Data using Learned Probability
Distributions, ICML 2000
Handouts
– 9 –
Statistics Based Outlier Detection SchemesStatistics Based Outlier Detection Schemes
Using InformationUsing Information--Theoretic Measures*Theoretic Measures*
Entropy measures the uncertainty (impurity) of data items
The entropy is smaller when the class distribution is skewer
Each unique data record represents a class => the smaller the entropy
the fewer the number of different records (higher redundancies)
If the entropy is large, data is partitioned into more regular subsets
Any deviation from achieved entropy indicates potential intrusion
Anomaly detector constructed on data with smaller entropy will be
simpler and more accurate
Conditional entropy H(X|Y) tells how much uncertainty
remains in sequence of events X after we have seen
subsequence Y (Y ∈ X)
Relative Conditional Entropy
* W. Lee, et al, Information-Theoretic Measures for Anomaly Detection, IEEE
Symposium on Security 2001
Distance based Outlier Detection SchemesDistance based Outlier Detection Schemes
Nearest Neighbor (NN) approach1,2
For each data point d compute the distance to the k-th nearest
neighbor dk
Sort all data points according to the distance dk
Outliers are points that have the largest distance dk and therefore
are located in the more sparse neighborhoods
Usually data points that have top n% distance dk are identified as
outliers
n – user parameter
Not suitable for datasets that have modes with varying density
1. Knorr, Ng,Algorithms for Mining Distance-Based Outliers in Large Datasets, VLDB98
2. S. Ramaswamy, R. Rastogi, S. Kyuseok: Efficient Algorithms for Mining Outliers from
Large Data Sets, ACM SIGMOD Conf. On Management of Data, 2000.
Handouts
– 10 –
Distance based Outlier Detection SchemesDistance based Outlier Detection Schemes
Mahalanobis-distance based approach
Mahalanobis distance is more appropriate for computing distances
with skewed distributions
dM =
Example:
In Euclidean space, data point p1 is closer to the origin than data point p2
When computing Mahalanobis distance, data points p1 and p2 are equally
distant from the origin
**
* *
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
•
♦
x’y’
p1p2
)p()p( T
µµ −⋅Σ⋅− −1
Density based Outlier Detection SchemesDensity based Outlier Detection Schemes
Local Outlier Factor (LOF) approach*
For each data point O compute compute the distance to the k-th
nearest neighbor (k-distance)
Compute reachability distance (reach-dist) for each data example O
with respect to data example p as:
reach-dist(O,p) = max{k-distance(p), d(O,p)}
Compute local reachability density (lrd) of data example O as inverse
of the average reachabaility distance based on the MinPts nearest
neighbors of data example O
lrd(O) =
Compute LOF of example p as the average of the ratios of the density
of example p and the density of its nearest neighbors
LOF(O) =
∑
p
MinPts pOdistreach
MinPts
),(_
∑⋅
p Olrd
plrd
MinPts )(
)(1
*- Breunig, et al, LOF: Identifying Density-Based Local Outliers, KDD 2000.
Handouts
– 11 –
Advantages of Density based SchemesAdvantages of Density based Schemes
Local Outlier Factor (LOF) approach
Example:
p2
× p1
×
In the NN approach, p2 is
not considered as outlier,
while the LOF approach
find both p1 and p2 as
outliers
NN approach may
consider p3 as outlier, but
LOF approach does not
×p3
Distance from p3 to
nearest neighbor
Distance from p2 to
nearest neighbor
Clustering based outlier detection schemes*Clustering based outlier detection schemes*
Radius ω of proximity is specified
Two points x1 and x2 are “near” if d(x1, x2) ≤ ω
Define N(x) – number of points that are within ω of x
Time Complexity O(n2) ⇒ approximation of the algorithm
Fixed-width clustering is first applied
The first point is a center of a cluster
If every subsequent point is “near” add to a cluster
Otherwise create a new cluster
Approximate N(x) with N(c)
Time Complexity – O(cn), c - # of clusters
Points in small clusters - anomalies
* E. Eskin et al., A Geometric Framework for Unsupervised Anomaly Detection:
Detecting Intrusions in Unlabeled Data, 2002
Handouts
– 12 –
Clustering based outlier detection schemesClustering based outlier detection schemes
K-nearest neighbor + canopy clustering approach *
Compute the sum of distances to the k nearest
neighbors (k-NN) of each point
Points in dense regions – small k-NN score
k has to exceed the frequency of any given attack type
Time complexity O(n2)
Speed up with canopy clustering that is used to split
the entire space into small subsets (canopies) and then
to check only the nearest points within the canopies
Apply fixed width clustering and compute distances
within clusters and to the centers of other clusters
* E. Eskin et al., A Geometric Framework for Unsupervised Anomaly Detection:
Detecting Intrusions in Unlabeled Data, 2002
Clustering based outlier detection schemesClustering based outlier detection schemes
FindOut algorithm* by-product of WaveCluster
Main idea: Remove the clusters from original data and
then identify the outliers
Transform data into multidimensional signals using
wavelet transformation
High frequency of the signals correspond to regions where is
the rapid change of distribution – boundaries of the clusters
Low frequency parts correspond to
the regions where the data is
concentrated
Remove these high and low
frequency parts and all remaining
points will be outliers
* D. Yu, G. Sheikholeslami, A. Zhang,
FindOut: Finding Outliers in Very Large Datasets, 1999.
Handouts
– 13 –
Model based outlier detection schemesModel based outlier detection schemes
Use a prediction model to learn the normal
behavior
Every deviation from learned prediction model
can be treated as anomaly or potential intrusion
Recent approaches:
Neural networks
Unsupervised Support Vector Machines (SVMs)
Neural networks for outlier detection*Neural networks for outlier detection*
Use a replicator 4-layer feed-forward neural network
(RNN) with the same number of input and output nodes
Input variables are the output variables so that RNN
forms a compressed model of the data during training
A measure of outlyingness is the reconstruction error
of individual data points. Target
variablesInput
* S. Hawkins, et al. Outlier detection using replicator neural networks,
DaWaK02 2002.
Handouts
– 14 –
Unsupervised Support Vector Machines forUnsupervised Support Vector Machines for
Outlier DetectionOutlier Detection
Unsupervised SVMs attempt to separate the entire set of
training data from the origin, i.e. to find a small region
where most of the data lies and label data points in this
region as one class
Parameters
Expected number of outliers
Variance of rbf kernel
As the variance of the rbf kernel
gets smaller, the number of
support vectors is larger and
the separating surface gets
more complex
origin
push the hyper plane away from
origin as much as possible
* E. Eskin et al., A Geometric Framework for Unsupervised Anomaly Detection:
Detecting Intrusions in Unlabeled Data, 2002.
* A. Lazarevic, et al., A Comparative Study of Anomaly Detection Schemes in Network
Intrusion Detection, SIAM 2003
Supervised Classification for Rare ClassesSupervised Classification for Rare Classes
Standard classification models are not suitable for rare classes
Models must be able to handle skewed class distributions
Learning from data streams - sequences of events
Key approaches:
Manipulating data records (oversampling / undersampling / generating
artificial examples)
Design of new algorithms (SHRINK, PN-rule, CREDOS)
Case specific feature/rule weighting
Boosting based algorithms (SMOTEBoost, RareBoost)
Cost sensitive classification (MetaCost, AdaCost, CSB, SSTBoost)
Emerging Patterns
Query/Active Learning Approach
Internally bias discrimination
Clustering based classification
Handouts
– 15 –
Manipulating Data RecordsManipulating Data Records
Over-sampling the rare class*
Make the duplicates of the rare events until the data set contains
as many examples as the majority class => balance the classes
Does not increase information but increase misclassification cost
Down-sizing (undersampling) the majority class**
Sample the data records from majority class
Randomly
Near miss examples
Examples far from minority class examples (far from decision boundaries)
Introduce sampled data records into the original data set instead
of original data records from the majority class
Usually results in a general loss of information and potentially
overly general rules
* Ling, C., Li, C. Data mining for direct marketing: Problems and solutions, KDD-98.
** Kubat M., Matwin, S., Addressing the Curse of Imbalanced Training Sets: One-Sided
Selection, ICML 1997.
Manipulating Data RecordsManipulating Data Records –– generating examplesgenerating examples
SMOTE (Synthetic Minority Over-sampling TEchnique)*
- over-sampling the rare (minority) class by
synthetically generating the minority class examples
When generating artificial minority class example,
distinguish two types of features
Continuous features
Nominal (Categorical) features
Generating artificial anomalies** [Fan 2001]
Artificial anomalies are generated around the edges of the
sparsely populated data regions
* N. Chawla, K. Bowyer, L. Hall, P. Kegelmeyer, SMOTE: Synthetic Minority OverSampling
Technique, JAIR, vol. 16, 321-357, 2002.
** W. Fan et al, Using Artificial Anomalies to Detect Unknown and Known Network
Intrusions, IEEE ICDM 2001
Handouts
– 16 –
i jk
l
m
n
Blue points corresponds to minority class examples
SMOTE Technique for Continuous FeaturesSMOTE Technique for Continuous Features
* N. Chawla, K. Bowyer, L. Hall, P. Kegelmeyer, SMOTE: Synthetic Minority OverSampling
Technique, JAIR, vol. 16, 321-357, 2002.
i jk
l
m
n
For each minority example k compute 5 nearest
minority class examples {i, j, l, n, m}
SMOTE Technique for Continuous FeaturesSMOTE Technique for Continuous Features
Handouts
– 17 –
i
jk
l
m
n
SMOTE Technique for Continuous FeaturesSMOTE Technique for Continuous Features
Randomly choose an example out of 5 closest points
Distance between the randomly chosen point i and the
current point k is diff
diff
i
jk
l
m
n
Randomly generate a number, such that it represents a
vector V that has length that is less then diff
SMOTE Technique for Continuous FeaturesSMOTE Technique for Continuous Features
diff
V
Handouts
– 18 –
i
jk
l
m
n
Synthetically generate event k1 based on the vector V,
such that k1 lies between point k and point i
SMOTE Technique for Continuous FeaturesSMOTE Technique for Continuous Features
diff
k1
i
jk
l
m
n
After applying SMOTE 3 times (SMOTE parameter = 300%)
data set may look like as the picture above
SMOTE Technique for Continuous FeaturesSMOTE Technique for Continuous Features
diff
k1 k3
k2
Handouts
– 19 –
SMOTE Technique for Nominal FeaturesSMOTE Technique for Nominal Features
For each minority example compute k nearest
neighbors using Value Difference Metric (VDM)
C1 – total number of occurrences of V1
C1i – total number of occurrences of V1 for class i
n – number of classes, p – constant (usually 1)
Create a synthetic example by taking a majority
vote among the k nearest neighbors
pn
i
ii
C
C
C
C
VV ∑
=
−=
1 2
2
1
1
21 ),(δ V1, V2 – corresponding feature values
Example: E1 = A B C D E
2 nearest neighbors are:
E2 = A F C G N
E3 = H B C D N
Esmote = A B C D N
Generating artificial anomalies*Generating artificial anomalies*
For sparse regions of data generate more artificial
anomalies than for the dense data regions
For each attribute value v generate
[(# of occurrence for most frequent value)
– (# of occurrences for this value)]
anomalies (va ≠ v, other attributes random
from data set)
Filter artificial anomalies to avoid collision with known
instance
Use RIPPER to discover rule sets
Pure anomaly detection vs. combined misuse and anomaly
detection
* W. Fan et al, Using Artificial Anomalies to Detect Unknown and Known Network
Intrusions, IEEE ICDM 2001.
Values of
Attribute i
Number of
occurrences
Number of
generated
examples
A 1000 B
100 900
C 10 990
Handouts
– 20 –
Design of New Algorithms: SHRINK*Design of New Algorithms: SHRINK*
SHRINK insists that a mixed region is labeled as positive
(rare) class, whether positives examples prevail in that
region or not
Focus on searching best positive region (with maximum ratio
positive examples to negative examples)
System is restricted to search for a single region to be labeled as positive
Induce classifier with low complexity
Classifier will be represented by the network of tests
Tests on numeric attributes have the form xi ∈ [min ai, max ai], Boolean xi = 0 v 1
hi - output of i-th test, hi = 1 if the test suggests a positive label, hi = -1 otherwise.
Example is classified positive if hi·wi > θ, wi - weight of i-th test
Remove min ai or max ai whichever reduces more radically # of negative examples
and results in better g-mean score
Find the interval with the maximum g-mean score and select it as the test
Tests with gi > 0.5 are discarded
* M. Kubat,R. Holte, S. Matwin, Learning when Negative Examples Abound, ECML-97
∑
−+
⋅=− accaccmeang
T
ii
i
i
i ple
e
e
⋅=
−
= ,
1
logω
Design of New Algorithms: PNDesign of New Algorithms: PN--rulerule Learning*Learning*
P-phase:
cover most of the positive examples with high support
seek good recall
N-phase:
remove FP from examples covered in P-phase
N-rules give high accuracy and significant support
Existing techniques can possibly
learn erroneous small signatures for
absence of C
C
NC
PNrule can learn strong signatures
for presence of NC in N-phase
C
NC
* M. Joshi, et al., PNrule, Mining Needles in a Haystack: Classifying Rare Classes via
Two-Phase Rule Induction, ACM SIGMOD 2001
Handouts
– 21 –
Design of New Algorithms: CREDOS*Design of New Algorithms: CREDOS*
Ripple Down Rules (RDRs) offer a unique tree based
representation that generalizes the decision tree and DNF
rule list models and specializes a generic form of multiphase
PNrule model
First use ripple down rules to overfit the training data
Generate a binary tree where each node is characterized by the rule
Rh, a default class and links to two child subtrees
Grow the RDS structure in a recursive manner
Induces rules at a node
Prune the structure in one pass to improve generalization
using Minimum Description Length (MDL) principle
Different mechanism from decision trees
* M. Joshi, et al., CREDOS: Classification Using Ripple Down Structure (A Case for
Rare Classes), SIAM International Conference on Data Mining, (SDM'04), 2004.
Case specific feature/rule weightingCase specific feature/rule weighting
Case specific feature weighting*
Information gain case-based learning (IG-CBL) algorithm
Create decision tree for the learning task
Compute the feature weights
wf = IG(f) if f is in the generated decision tree (IG – information gain), otherwise wf = 0
Testing phase:
For each rare class test example replace global weight vector
with dynamically generated weight vector that depends on the
path taken by that test example
Case specific rule weighting**
LERS (Learning from Examples based on Rough Sets) algorithm uses modified
“bucket brigade” algorithm to create certain and possible rules
Strength – how well the rule performed during the training phase
Support – sum of scores of all the matching rules from the concept
Increase the rule strength for all rules describing the rare class
* C. Cardie, N. Howe, Improving Minority Class Prediction Using Case specific feature
weighting, ICML-1997.
** J. Grzymala et al, An Approach to Imbalanced Data Sets Based on Changing Rule
Strength, AAAI Workshop on Learning from Imbalanced Data Sets, 2000.
Handouts
– 22 –
Manipulates training examples to generate
multiple classifiers
Proceeds in a series of rounds
Maintains a Dt - distribution of weights wt over
the training examples
- importance weight
where Zt is a normalization constant chosen
such that Dt+1 is a distribution
In each round t, the learning algorithm is invoked
to output a classifier Ct
Standard BoostingStandard Boosting -- BackgroundBackground
))((
1 )/)(()( itit xhy
ttt eZiDiD α−
+ ⋅= 







−
+
⋅=
t
t
t
r
r
1
1
ln
2
1
α
BoostingBoosting –– Combining ClassifiersCombining Classifiers
The error of classifier Ct (ht) is used to update the
sampling weights of the training examples
Misclassified examples – higher weights
Correctly classified examples – smaller weights
The final classifier is constructed as a weighted
vote of individual classifiers
Each classifier Ct (ht) is weighted by αt according
to its accuracy on the entire training set








= ∑
=
M
t
tt xhsignxH
1
)()( α
Handouts
– 23 –
Boosting Based AlgorithmsBoosting Based Algorithms -- SMOTEBoostSMOTEBoost
SMOTEBoost embeds SMOTE technique in the
boosting procedure
Utilize SMOTE for improving the prediction of the
minority class
Utilize boosting for improving general performance
After each boosting round, SMOTE is used to
create new synthetic examples from the
minority class
SMOTEBoost indirectly gives higher weights to
misclassified minority (rare) class examples
Constructed classifiers are more diversed
* N. Chawla, A. Lazarevic, et al, SMOTEBoost: Improving the Prediction of Minority
Class in Boosting, PKDD 2003.
Boosting Based AlgorithmsBoosting Based Algorithms -- SLIPPERSLIPPER
SLIPPER builds rules only for positive class
The only rule that predicts the negative class is a single
default rule
The weak hypotheses ht(x) used here are rules
Force the weak hypothesis based on a rule R to abstain
(vote with confidence 0) on all instances not satisfied
by R (x ∉R) by setting the prediction h(x) for x ∉R to 0
Force the rule to predict with the same confidence CR
on every x ∈ R, for t-th rule Rt , αtht(x) = CRt








= ∑
∈ tt
t
RxR
RCsignxH
:
)(



=
0
)( tR
t
C
xh
if x ∈ Rt
otherwise
* W. Cohen, Y. Singer, A Simple, Fast and Effective Rule Learner, Annual Conference
of American Association for Artificial Intelligence, 1999.
Handouts
– 24 –
Boosting Based AlgorithmsBoosting Based Algorithms -- RareBoostRareBoost
RareBoost-1
Modify boosting by choosing different weight update
factors for positives (TP, FP)
and negatives (TN, FN)
To achieve good recall for class C, distinguish between FN from TP
To achieve good precision for class C, distinguish between TP from FP
RareBoost-2
Build rules for both the classes in every iteration
Modify SLIPPER by choosing between C and NC model
based on whichever minimizes corresponding Zt value
SLIPPER-C chooses between C and default
SLIPPER-NC chooses between NC and default
)/ln()2/1( tt
p
t FPTP⋅=α
)/ln()2/1( tt
n
t FNTN⋅=α








+= ∑∑
<≥ 0)(0)(
)()()(
xh
t
n
t
xh
t
p
t
tt
xhxhsignxH αα
* M. Joshi, et al., Predicting Rare Classes: Can Boosting Make Any Weak Learner
Strong?, KDD 2002.
Cost Sensitive ClassificationCost Sensitive Classification
Learning to minimize the expected cost of
misclassifications
Most classification learning algorithms attempt to
minimize expected number of misclassification errors
In many applications, different kinds of classification
errors have different costs, so we need cost-sensitive
methods
Intrusion Detection
False positive: security analysts waste time to analyze normal behavior
False negative: failure to detect intrusion could cause big damages
Fraud Detection
False positive: resources wasted investigating non-fraud
False negative: failure to detect fraud could be very expensive
Medical Diagnosis: Cost of false negative error - Postponed
treatment or failure to treat; death or injury
Handouts
– 25 –
Cost MatrixCost Matrix
C(i,j) = cost of predicting class i when the true class is j
Example: Misclassification Costs Diagnosis of Cancer
If M is the confusion matrix for a classifier: M(i,j) is the
number of test examples that are predicted to be in
class i when their true class is j
Expected misclassification cost is Hadamard product of
M and C divided by the number of test examples N:
True State of PatientPredicted State
of Patient
C(1,1) = 0C(1,0) = 100Negative - 1
C(0,1) = 1C(0,0) = 1Positive – 0
Negative - 1Positive - 0
∑
=
⋅
N
ji
jiCjiM
N 1,
),(),(
1
Cost Sensitive Classification TechniquesCost Sensitive Classification Techniques
Two Learning Problems
Cost C known at learning time
Cost C not known at learning time (only becomes
available at classification time)
Learned classifier should work well for a wide range of costs
Learning with known cost C
Find a classifier h that minimizes the expected
misclassification cost on new data set points
Two basic strategies
Modify the inputs to the learning algorithm to reflect cost C
Incorporate cost C into the learning algorithm
Handouts
– 26 –
Cost Sensitive Classification TechniquesCost Sensitive Classification Techniques
Modifying inputs of learning algorithm
If there are 2 classes and the cost of a false positive is
λ times larger than the cost of a false negative, put a
weight of λ on each negative training example
λ = C(1,0) / C(0,1)
Then apply the learning algorithm as before
Setting λ by class frequency (less frequent class has higher cost)
λ ~ 1/nk, nk - number of training examples from class k
Setting λ by cross-validation
Setting λ according to class frequency is cheaper
and gives the same results as setting λ by cross
validation
Cost Sensitive Learning:Cost Sensitive Learning: MetaCostMetaCost
MetaCost wraps a “cost-minimizing” procedure around
an arbitrary classifier, so that the classifier effectively
minimizes cost, while seeking to minimize zero-one loss
Bayes optimal prediction for x is class i that minimizes
R(i|x) = Σ P(j|x) C(i,j)
Conditional risk R(i|x) is expected cost of predicting that x belongs
to class i
Both C(i,j) and P(j|x) together with the above rule mean that the
example space X can be partitioned into j regions, such that class j
is the best (optimal) prediction in region j
MetaCost is based on estimating class probabilities P(j|x)
Standard probability estimation methods can determine
these probabilities, but require machine learning bias of
both the classifier and the density
* P. Domingos, MetaCost: A general Method for making Classifiers Cost-Sensitive,
KDD 1999.
Handouts
– 27 –
Cost Sensitive Learning:Cost Sensitive Learning: MetaCostMetaCost
MetaCost estimates the class probabilities by learning
multiple classifiers and for each example it uses the
class’s fraction of the total vote as an estimate
Method for estimating probabilities for modern unstable learners
Bagging – given a training set S, a “bootstrap” sample of
it is constructed by taking |S| samples with replacement
MetaCost is different from bagging, where it uses smaller
number of examples in each sample, which allows it to be
more efficient
While estimating class probabilities, MetaCost allows
taking all the models generated (samples) into
consideration, or only those samples
The first type has a lower variance, because of its larger
samples, and the second has lower statistical bias
* P. Domingos, MetaCost: A general Method for making Classifiers Cost-Sensitive,
KDD 1999.
Cost Sensitive Classification TechniquesCost Sensitive Classification Techniques
Modifying learning algorithm - Incorporating cost
C into the learning algorithm
Cost-Sensitive Boosting
AdaCost*
CSB**
SSTBoost***
Cost C can be incorporated directly into the error
criterion when training neural networks (Kukar &
Kononenko, 1998)
* W. Fan, S. Stolfo, J. Zhang, and P. Chan, Adacost: Misclassification cost-sensitive
boosting, ICML 1999.
** K. M. Ting, A Comparative Study of Cost-Sensitive Boosting Algorithms, ICML 2000.
*** S. Merler, C. Furlanello, B. Larcher, and A. Sboner. Automatic model selection in
cost-sensitive boosting, Information Fusion, 2003
Handouts
– 28 –
Cost Sensitive BoostingCost Sensitive Boosting
Training examples of the form (xi, yi, ci), where ci
is the cost of misclassifying xi
Boosting: wi := wi * exp(–αt yi ht(xi))/Zt
AdaCost [Fan et al., 1999]
wi := wi * exp(–αt yi ht(xi) βi)/Zt
βi = ½ * (1 + ci) if error
= ½ * (1 – ci) otherwise
Cost-Sensitive Boosting (CSB) [Ting, 2000]
wi := βi wi * exp(–αt yi ht(xi))/Zt
βi = ci if error
= 1 otherwise
SSTBoost [Merler et al., 2003]
wi := wi * exp(–αt yi ht(xi) βi)/Zt
βi = ci if error
βi = 2 – ci otherwise
ci = w for positive examples; 1 – w for negative examples
Learning with Unknown CLearning with Unknown C
Construct a classifier h(x,C) that can accept the
cost function at run time and minimize the
expected cost of misclassification errors with
respect to cost C
Approach: Learning to estimate P(y|x)
Probability Estimation Trees [Provost, Domingos 2000]
Bagged Probability Estimation Trees [Provost,
Domingos 2000]
Lazy Option Trees [Friedman96, Kohavi97, Margineantu &
Dietterich, 2001]
Bagged Lazy Option Trees [Margineantu, Dietterich 2001]
Handouts
– 29 –
Internally bias discrimination*Internally bias discrimination*
A rule learning algorithm is
used to uncover indicators of
fraudulent behavior from a data
set of customer transactions
Indicators are used to create a
set of monitors, which profile
legitimate customer behavior
and indicate anomalies
Outputs of the monitors are used as features in a system
that learns to combine evidence to generate high
confidence alarms
The system has been applied to the problem of detecting
cellular cloning fraud
* T. Fawcett, F. Provost, Adaptive Fraud Detection, DMKD 1997.
Selective Sampling based on Query LearningSelective Sampling based on Query Learning
Active/Query Learning [Angluin 88]
Learner chooses examples on which the labels are requested
Uncertainty Sampling [Lewis and Gale 94]
Learner queries examples for which its prediction so far is
uncertain in order to maximize information gain
Example: Query by committee [Seung et al 92], which queries
examples on which models obtained so far disagree on
Successfully applied to getting labeled data for text
classification
Selective sampling by query learning [Freund et al 93]
Given large number of (possibly unlabeled) data, uses only
small subset of labeled data for learning
Successfully applied to mining very large data set [Mamitsuka
and Abe 2000], even when labeled data are abound
Handouts
– 30 –
Query learning for imbalanced data setQuery learning for imbalanced data set
Use query learning to get more data for rare classes
Use selective sampling by query learning to get data
near the decision boundary
Class 2Class 1
Class 2Class 1
Class 2Class 1
Example 1
Example 3
+
+
+
+
++
+
+
+
+
-
-
-
-
-
-
-
Example
2
* N. Abe, Sampling Approaches to Learning from Imbalanced Datasets, ICML
Workshop on Imbalanced Data Sets II, 2003.
Sampling for CostSampling for Cost--sensitive Learning*sensitive Learning*
Cost C depends on particular example x
Presents reduction of cost-sensitive learning to classification
Altering the original example distribution by multiplying it by a factor
proportional to the relative cost of each example, makes any error
minimizing classifier accomplish expected cost minimization on the
original distribution
Proposes Costing (cost-sensitive ensemble learning)
Empirical evaluation using benchmark data sets from targeted
marketing domain
Costing has excellent predictive performance (w.r.t. cost minimization)
Costing is computationally efficient
* B. Zadrozny, J. Langford, N. Abe, Cost-sensitive learning by cost-proportionate
example weighting, IEEE ICDM 2003.
C(1,1,x)C(1,0,x)Predict = 1
C(0,1,x)C(0,0,x)Predict = 0
True = 1True = 0
Handouts
– 31 –
Association Rules for Rare ClassAssociation Rules for Rare Class
Emerging patterns*
For 2 classes C1 and C2 and support si(X) of itemset X
in class Ci growth rate is defined as s2(X)/s1(X)
Given a threshold p >1, itemset X is p-emerging pattern
(EP) if growth rate ≥ p
Find EPs in rare class
Find values that have the highest growth rate from the
major class to the rare class
Replace different attribute values in the original rare
class EPs with the highest growth rate values
New generated EPs have a strong power to
discriminate rare class from the major class
* H. Alhammady, K. Rao, The Application of Emerging Patterns for Improving the
Quality of Rare-class Classification, PAKDD 2004.
Temporal Analysis of Rare EventsTemporal Analysis of Rare Events
Surprising patterns [Keogh et al, KDD02]
A time series pattern P, extracted from database X is surprising
relative to a database R, if the probability of its occurrence is
greatly different to that expected by chance, assuming that R and
X are created by the same underlying process.
Steps (TARZAN algorithm)
Discretizing time-series into symbolic strings
Fixed sized sliding window
Slope of the best-fitting line
Calculate probability of any pattern, including ones we have
never seen before using Markov models
For maintaining linear time and space property, they use suffix
tree data structure
Computing scores by comparing trees between reference data
and incoming information stream
* E. Keogh, et al, Finding Surprising Patterns in a Time Series Database in Linear
Time and Space, KDD 2002.
Handouts
– 32 –
Temporal Analysis of Rare EventsTemporal Analysis of Rare Events
Learning to Predict Extremely Rare Events*
Genetic-based machine learning system that predicts events by
identifying temporal and sequential patterns in data
Temporal Sequence Associations for Rare Events**
Predicting Rare Events In Temporal Domains***
Transform event prediction problem into a search for all patterns
(event-sets) on the rare class exclusively
Discriminative power of patterns is validated against other
classes
Patterns are then combined into a rule-based model for prediction
* G. Weiss, H. Hirsh, Learning to Predict Extremely Rare Events, AAAI Workshop on
Learning from Imbalanced Data Sets, 2000.
** J. Chen, et al, Temporal Sequence Associations for Rare Events, PAKDD 2004
*** R. Vilalta, Predicting Rare Events In Temporal Domains, IEEE ICDM 2002.
Using Clustering for Rare Class Problems*Using Clustering for Rare Class Problems*
Distinguish between regular “between class imbalances”
and “within-class imbalance”, where a single class is
composed of various sub-classes of different size
The procedure
Use clustering to identify subclasses (subclusters)
Each subcluster of the majority class is resampled until it
reaches the size of the biggest subcluster in the majority class
Denote size of the new majority class as:
Size_of_new_majority_class
Each subcluster of the rare class is resampled until it reaches
the size
Size_of_majority_class / Nsubclusters_in_rare_class
* N. Japkowicz, Class Imbalances: Are We Focusing on the Right Issues, ICML
Workshop on Imbalanced Data Sets II, 2003.
Handouts
– 33 –
Case StudiesCase Studies
Intrusion Detection
Network Intrusion Detection
Host based Intrusion Detection
Fraud Detection
Credit Card Fraud
Insurance Fraud Detection
Cell Fraud Detection
Medical Diagnostics
Mammogramy images
Health Care Fraud
Case Study: Data Mining in Intrusion DetectionCase Study: Data Mining in Intrusion Detection
Due to the proliferation of Internet,
more and more organizations are
becoming vulnerable to cyber attacks
Sophistication of cyber attacks as well
as their severity is also increasing
Security mechanisms always have
inevitable vulnerabilities
Firewalls are not sufficient to ensure
security in computer networks
Insider attacks
Incidents Reported to Computer Emergency Response
Team/Coordination Center (CERT/CC)
Attack sophistication vs. Intruder technical knowledge,
source: www.cert.org/archive/ppt/cyberterror.ppt
The geographic spread of Sapphire/Slammer Worm
30 minutes after release (Source: www.caida.org)
0
20000
40000
60000
80000
100000
120000
1 2 3 4 5 6 7 8 9 10 11 12 13 141990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Handouts
– 34 –
What are Intrusions?What are Intrusions?
Intrusions are actions that attempt to bypass security mechanisms
of computer systems. They are usually caused by:
Attackers accessing the system from Internet
Insider attackers - authorized users attempting to gain and misuse nonauthorized
privileges
Typical intrusion scenario
Scanning
activity
Computer
Network
Attacker Machine with
vulnerability
What are Intrusions?What are Intrusions?
Intrusions are actions that attempt to bypass security mechanisms
of computer systems. They are caused by:
Attackers accessing the system from Internet
Insider attackers - authorized users attempting to gain and misuse nonauthorized
privileges
Typical intrusion scenario
Computer
Network
Attacker Compromised
Machine
Handouts
– 35 –
Basic IDS ModelBasic IDS Model
Information Source - Monitored System
Detector – ID Engine
Response
Component
Data gathering (sensors)
Raw data
Events
Knowledge base Configuration
Alarms
Actions
System State
System
State
IDS TaxonomyIDS Taxonomy
IDS
Information
source
Analysis
strategy
Architecture
Time Aspects
Activeness
Continuality
Host based
Network based
Wireless network
Application Log
Anomaly Detection
Misuse Detection
Unsupervised
Supervised
Data Mining
State-transition
Expert systemsReal-time prediction
Off-line prediction
Centralized
Distributed & heterogeneous
Active response
Passive reaction
Continuous monitoring
Periodic analysis
…..
Sensor Alerts
Handouts
– 36 –
IDSIDS -- Analysis StrategyAnalysis Strategy
Misuse detection is based on extensive knowledge of patterns
associated with known attacks provided by human experts
Existing approaches: pattern (signature) matching, expert systems,
state transition analysis, data mining
Major limitations:
Unable to detect novel & unanticipated attacks
Signature database has to be revised for each new type of discovered attack
Anomaly detection is based on profiles that represent normal
behavior of users, hosts, or networks, and detecting attacks as
significant deviations from this profile
Major benefit - potentially able to recognize unforeseen attacks.
Major limitation - possible high false alarm rate, since detected
deviations do not necessarily represent actual attacks
Major approaches: statistical methods, expert systems, clustering,
neural networks, support vector machines, outlier detection schemes
Intrusion DetectionIntrusion Detection
Intrusion Detection System
combination of software
and hardware that attempts
to perform intrusion detection
raises the alarm when possible
intrusion happens
Traditional intrusion detection system IDS tools (e.g.
SNORT) are based on signatures of known attacks
Example of SNORT rule (MS-SQL “Slammer” worm)
any -> udp port 1434 (content:"|81 F1 03 01 04 9B 81 F1 01|";
content:"sock"; content:"send")
Limitations
Signature database has to be manually revised for each new type
of discovered intrusion
They cannot detect emerging cyber threats
Substantial latency in deployment of newly created signatures
across the computer system
Data Mining can alleviate these limitations
www.snort.org
Handouts
– 37 –
Data Mining for Intrusion DetectionData Mining for Intrusion Detection
Increased interest in data mining based intrusion detection
Attacks for which it is difficult to build signatures
Attack stealthiness
Unforeseen/Unknown/Emerging attacks
Distributed/coordinated attacks
Data mining approaches for intrusion detection
Misuse detection
Building predictive models from labeled labeled data sets (instances
are labeled as “normal” or “intrusive”) to identify known intrusions
High accuracy in detecting many kinds of known attacks
Cannot detect unknown and emerging attacks
Anomaly detection
Detect novel attacks as deviations from “normal” behavior
Potential high false alarm rate - previously unseen (yet legitimate) system
behaviors may also be recognized as anomalies
Summarization of network traffic
Data MiningData Mining for Intrusion Detectionfor Intrusion Detection
Misuse
Detection –
Building
Predictive
Models
categorical
tem
poral
continuous
class
Test
Set
Training
Set Model
Learn
Classifier
Tid SrcIP
Start
time
Dest IP Dest
Port
Number
of bytes
Attack
1 206.135.38.95 11:07:20 160.94.179.223 139 192 No
2 206.163.37.95 11:13:56 160.94.179.219 139 195 No
3 206.163.37.95 11:14:29 160.94.179.217 139 180 No
4 206.163.37.95 11:14:30 160.94.179.255 139 199 No
5 206.163.37.95 11:14:32 160.94.179.254 139 19 Yes
6 206.163.37.95 11:14:35 160.94.179.253 139 177 No
7 206.163.37.95 11:14:36 160.94.179.252 139 172 No
8 206.163.37.95 11:14:38 160.94.179.251 139 285 Yes
9 206.163.37.95 11:14:41 160.94.179.250 139 195 No
10 206.163.37.95 11:14:44 160.94.179.249 139 163 Yes
10
Tid SrcIP
Start
time
Dest Port
Number
of bytes
Attack
1 206.163.37.81 11:17:51 160.94.179.208 150 ?
2 206.163.37.99 11:18:10 160.94.179.235 208 ?
3 206.163.37.55 11:34:35 160.94.179.221 195 ?
4 206.163.37.37 11:41:37 160.94.179.253 199 ?
5 206.163.37.41 11:55:19 160.94.179.244 181 ?
categorical
Anomaly Detection
Rules Discovered:
{Src IP = 206.163.37.95,
Dest Port = 139,
Bytes ∈ [150, 200]} --> {ATTACK}
Rules Discovered:
{Src IP = 206.163.37.95,
Dest Port = 139,
Bytes ∈ [150, 200]} --> {ATTACK}
Summarization of
attacks using
association rules
Handouts
– 38 –
Data MiningData Mining for Intrusion Detectionfor Intrusion Detection
Misuse
Detection –
Building
Predictive
Models
categorical
tem
poral
continuous
class
Test
Set
Training
Set Model
Learn
Classifier
Tid SrcIP
Start
time
Dest IP Dest
Port
Number
of bytes
Attack
1 206.135.38.95 11:07:20 160.94.179.223 139 192 No
2 206.163.37.95 11:13:56 160.94.179.219 139 195 No
3 206.163.37.95 11:14:29 160.94.179.217 139 180 No
4 206.163.37.95 11:14:30 160.94.179.255 139 199 No
5 206.163.37.95 11:14:32 160.94.179.254 139 19 Yes
6 206.163.37.95 11:14:35 160.94.179.253 139 177 No
7 206.163.37.95 11:14:36 160.94.179.252 139 172 No
8 206.163.37.95 11:14:38 160.94.179.251 139 285 Yes
9 206.163.37.95 11:14:41 160.94.179.250 139 195 No
10 206.163.37.95 11:14:44 160.94.179.249 139 163 Yes
10
Tid SrcIP
Start
time
Dest IP
Number
of bytes
Attack
1 206.163.37.81 11:17:51 160.94.179.208 150 No
2 206.163.37.99 11:18:10 160.94.179.235 208 No
3 206.163.37.55 11:34:35 160.94.179.221 195 Yes
4 206.163.37.37 11:41:37 160.94.179.253 199 No
5 206.163.37.41 11:55:19 160.94.179.244 181 Yes
categorical
Anomaly Detection
Rules Discovered:
{Src IP = 206.163.37.95,
Dest Port = 139,
Bytes ∈ [150, 200]} --> {ATTACK}
Rules Discovered:
{Src IP = 206.163.37.95,
Dest Port = 139,
Bytes ∈ [150, 200]} --> {ATTACK}
Summarization of
attacks using
association rules
Data MiningData Mining for Intrusion Detectionfor Intrusion Detection
Misuse
Detection –
Building
Predictive
Models
categorical
tem
poral
continuous
class
Test
Set
Training
Set Model
Learn
Classifier
Tid SrcIP
Start
time
Dest IP Dest
Port
Number
of bytes
Attack
1 206.135.38.95 11:07:20 160.94.179.223 139 192 No
2 206.163.37.95 11:13:56 160.94.179.219 139 195 No
3 206.163.37.95 11:14:29 160.94.179.217 139 180 No
4 206.163.37.95 11:14:30 160.94.179.255 139 199 No
5 206.163.37.95 11:14:32 160.94.179.254 139 19 Yes
6 206.163.37.95 11:14:35 160.94.179.253 139 177 No
7 206.163.37.95 11:14:36 160.94.179.252 139 172 No
8 206.163.37.95 11:14:38 160.94.179.251 139 285 Yes
9 206.163.37.95 11:14:41 160.94.179.250 139 195 No
10 206.163.37.95 11:14:44 160.94.179.249 139 163 Yes
10
categorical
Anomaly Detection
Rules Discovered:
{Src IP = 206.163.37.95,
Dest Port = 139,
Bytes ∈ [150, 200]} --> {ATTACK}
Rules Discovered:
{Src IP = 206.163.37.95,
Dest Port = 139,
Bytes ∈ [150, 200]} --> {ATTACK}
Summarization of
attacks using
association rules
Tid SrcIP
Start
time
Dest IP
Number
of bytes
Attack
1 206.163.37.81 11:17:51 160.94.179.208 150 No
2 206.163.37.99 11:18:10 160.94.179.235 208 No
3 206.163.37.55 11:34:35 160.94.179.221 195 Yes
4 206.163.37.37 11:41:37 160.94.179.253 199 No
5 206.163.37.41 11:55:19 160.94.179.244 181 Yes
Handouts
– 39 –
Data Preprocessing for Data Mining in IDData Preprocessing for Data Mining in ID
Converting the data from monitored system (computer
network, host machine, …) into data (features) that will
be used in data mining models
For misuse detection, labeling data examples into normal or
intrusive may require enormous time for many human experts
Building data mining models
Misuse detection models
Anomaly detection models
Analysis and summarization of results
Feature
construction
Building data mining models
features
Analysis
of results
Data Sources in Network Intrusion DetectionData Sources in Network Intrusion Detection
Network traffic data is usually collected using “network sniffers”
Tcpdump
08:02:15.471817 0:10:7b:38:46:33 0:10:7b:38:46:33 loopback 60:
0000 0100 0000 0000 0000 0000 0000 0000
0000 0000 0000 0000 0000 0000 0000 0000
0000 0000 0000 0000 0000 0000 0000
08:02:19.391039 172.16.112.100.3055 > 172.16.112.10.ntp: v1 client strat 0 poll 0 prec 0
08:02:19.391456 172.16.112.10.ntp > 172.16.112.100.3055: v1 server strat 5 poll 4 prec -16 (DF)
net-flow tools
Source and destination IP address, Source and destination ports, Type of
service, Packet and byte counts, Start and end time, Input and output
interface numbers, TCP flags, Routing information (next-hop address,
source autonomous system (AS) number, destination AS number)
0624.12:4:39.344 0624.12:4:48.292 211.59.18.101 4350 160.94.179.138 1433 6 2 3 144
0624.9:1:10.667 0624.9:1:19.635 24.201.13.122 3535 160.94.179.151 1433 6 2 3 132
0624.12:4:40.572 0624.12:4:49.496 211.59.18.101 4362 160.94.179.150 1433 6 2 3 152
Collected data are in the form of network connections or network
packets (a network connection may contain several packets)
Handouts
– 40 –
Feature Construction in Intrusion DetectionFeature Construction in Intrusion Detection
Basic set of features (start/end time, srcIP, dstIP, srcport,
dstport, protocol, TCP flags, # of bytes, packets, …)
Three groups of features may be constructed (KDDCup 99):
“content-based” features within a connection
Intrinsic characteristics of data packets
number of failed logins, root logins, acknowledgments, directory creations, …)
time-based traffic features included number of connections or
different services from the same source or to the same destination
considering recent time interval (e.g.a few seconds)
Useful for detecting scanning activities
connection based features included number of connections from
same source or to same destination or with the same service
considering in last N connections
Useful for detecting SLOW scanning activities
MADAM IDMADAM ID -- Feature Construction ExampleFeature Construction Example
Example: “syn flood” patterns
(service=http, flag=SYN, victim),
(service=http, flag=SYN, victim)
→ (service = http, SYN, victim)
[0.9, 0.02, 2s]
After 2 http packets with set SYN flag are sent to victim, in 90% of cases
third connection is made within 2s. This happens in 2% of all connections
(support = 2%)
Add features based on frequent episodes learned separately for normal
data and attack data
count the connections in the past 2 seconds
count the connections with the same characteristics (http, SYN, victim) in the past 2
seconds
Different features are found for different classes of attacks =>
different models for the classes
flagdst … service …
h1 http S0
h1 http S0
h1 http S0
h2 http S0
h4 http S0
h2 ftp S0
syn flood
normal
existing featuresexisting features
uselessuseless
flagdst … service …
h1 http S0
h1 http S0
h1 http S0
h2 http S0
h4 http S0
h2 ftp S0
syn flood
normal
flagdst … service …
h1 http S0
h1 http S0
h1 http S0
h2 http S0
h4 http S0
h2 ftp S0
syn flood
normal
existing featuresexisting features
uselessuseless
dst … service …
h1 http S0
h1 http S0
h1 http S0
h2 http S0
h4 http S0
h2 ftp S0
flag %S0
70
72
75
0
0
0
construct features withconstruct features with
high information gainhigh information gain
dst … service …
h1 http S0
h1 http S0
h1 http S0
h2 http S0
h4 http S0
h2 ftp S0
flag %S0
70
72
75
0
0
0
dst … service …
h1 http S0
h1 http S0
h1 http S0
h2 http S0
h4 http S0
h2 ftp S0
flag %S0
70
72
75
0
0
0
construct features withconstruct features with
high information gainhigh information gain
Handouts
– 41 –
MADAM ID Workflow*MADAM ID Workflow*
Association rules and frequent episodes are applied to
network connection records to obtain additional
features for data mining algorithms
Apply RIPPER to labeled data sets and learn the
intrusions
Raw audit
data
network
packets/events
Connection
records
RIPPER
Model
patterns features
Feature
constructor Evaluation feedback
* W. Lee,S. Stolfo, Adaptive Intrusion Detection: a Data Mining Approach, Artificial
Intelligence Review, 2000
MADAM IDMADAM ID -- CostCost--sensitive Modelingsensitive Modeling
A multiple-model approach:
Certain features are more costly to compute than
others
Build multiple rule-sets, each with features of different
cost levels
Cost factors: damage cost, response cost, operational
cost (level 1-4 features)
Level 1: beginning of an event
Level 2: middle to end of an event
Level 3: at the end of an event
Level 4: at the end of an event, multiple events in a time window
Use cheaper rule-sets first, costlier ones later only for
required accuracy
* W. Lee,et al., Toward Cost-Sensitive Modeling for Intrusion Detection and
Response, Journal of Computer Security, 2002
Handouts
– 42 –
ADAM*ADAM*
Data mining testbed that uses combination of
association rules and classification to discover attacks
I phase: ADAM builds a repository of profiles for
“normal frequent itemsets” by mining “attack-free” data
II phase: ADAM runs a sliding window, incremental
algorithm that finds frequent itemsets in the last N
connections and compare them to “normal” profile
Tunable: different thresholds can be set for different
types of rules.
Anomaly detection: first characterize normal behavior
(profile), then flag abnormal behavior
Reduced false alarm rate: using a classifier.
* D. Barbara, et al., ADAM: A Testbed for Exploring the Use of Data Mining in
Intrusion Detection. SIGMOD Record 2001.
Attack-free
data
ADAM: Training phaseADAM: Training phase
Database of frequent itemsets for attack-free data is made
For entire training data, find suspicious frequent itemsets
that are not in the “attack-free” database
Train a classifier to classify itemset as known attack,
unknown attack or normal event
Data
preprocessing
tcpdump
data
Training
data
Classifier
Handouts
– 43 –
ADAM: Phase of Discovering IntrusionsADAM: Phase of Discovering Intrusions
Dynamic mining module produces suspicious itemsets
from test data
Along with features from feature selection module,
itemsets are fed to classifier
Feature
selection
Test data Dynamic
mining
profile
Classifier
Attacks, False alarms,
Unknown
Data
preprocessing
tcpdump
data
The MINDS Project*The MINDS Project*
MINDS – Minnesota Intrusion Detection System
M
I
N
D
S
network
Data capturing
device
Anomaly
detection
…
…
Anomaly
scores
Human
analyst
Detected novel
attacks
Summary and
characterization
of attacks
MINDS system
Known attack
detection
Detected
known attacks
Labels
Feature
Extraction
Association
pattern analysis
MINDSAT
Filtering
Net flow tools
tcpdump
MINDS has been incorporated into the Interrogator architecture at the Army
Research Labs (ARL) Center for Intrusion Monitoring and Protection (CIMP),
where network data from multiple sensors is collected and analyzed.
The MINDS is being used at University of Minnesota and at the ARL-CIMP to
help analysts to detect attacks and intrusive behavior that cannot be
detected using widely used intrusion detection systems, such as SNORT.
* - Ertoz, L., Eilertson, E., Lazarevic, et al, The MINDS - Minnesota Intrusion Detection System, review
for the book “Data Mining: Next Generation Challenges and Future Directions”, AAAI/MIT Press.
Handouts
– 44 –
Alternative Classification ApproachesAlternative Classification Approaches
Fuzzy Data Mining in network intrusion detection (MSU)*
Create fuzzy association rules only from normal data to
learn “normal behavior”
For new audit data, create the set of fuzzy association rules and
compute its similarity to the “normal” one
If similarity low ⇒ for new data generate alarm
Genetic algorithms (GA) used to tune membership
function of the fuzzy sets
Fitness – rewards for high similarity between normal and
reference data, penalizing – high similarity between intrusion
and reference data
Use GA to select most relevant features
* S. Bridges, R. Vaughn, Intrusion Detection Via Fuzzy Data Mining, 2000
Alternative Classification Approaches*Alternative Classification Approaches*
Scalable Clustering Technique*
Apply supervised clustering
For each point find nearest point and if belong to the same class,
append to the same cluster, else create a new
Classification
Class dominated in k nearest clusters
Weighted sum of distances to k nearest clusters
Incremental clustering
Distances: weighted Euclidean, Chi-square, Canbera
(d(x, L)) =
* N. Ye, X. Li, A Scalable Clustering for Intrusion Signature
Recognition, 2001.
∑= +
−d
i
i
ii
ii
C
Lx
Lx
1
2|| Li – i-th attribute of centroid of the cluster L
Ci – correlation between i-th attribute and the class
Handouts
– 45 –
Neural Networks Classification ApproachesNeural Networks Classification Approaches
Neural networks (NNs) applied to host-based
intrusion detection
Building profiles of users according to used
commands
Building profiles of software behavior
Neural networks for network-based intrusion
detection
Hierarchical network intrusion detection
Multi-layer perceptrons (MLP)*
Self organizing maps (SOMs)*
* J. Canady, J. Mahaffey, The Application of Artificial Neural Networks to Misuse
Detection:Initial Results, 1998.
Statistics Based Outlier Detection SchemesStatistics Based Outlier Detection Schemes
Packet level (PHAD) and Application level (ALAD) anomaly detectiPacket level (PHAD) and Application level (ALAD) anomaly detection*on*
PHAD (packet header anomaly detection) monitors Ethernet, IP and
transport layer packet headers
It builds profiles for 33 different fields from these headers by looking
attack free traffic and performs clustering (prespecifed # of clusters)
A new value that does not fit into any of the clusters (intervals), it is
treated as a new cluster and closest two clusters are merged
The number of updates, r, is maintained for each field as well as the
number of observations, n
Testing: For each new observed packet, if the value for some attribute
does not fit into the clusters, anomaly score for that attribute is
proportional to n/r
ALAD uses the same method for anomaly scores, but it works only on
TCP data and build TCP streams
It build profiles for 5 different features
* M. Mahoney, P. Chan: Learning Nonstationary Models of Normal Network Traffic
for Detecting Novel Attacks, 8th ACM KDD, 2002
Handouts
– 46 –
NATE*NATE*
Low-cost anomaly detection algorithm
It uses only header information
5 features are used:
Counts of Push and Ack TCP flags (Syn, Fin, Reset and Urgent TCP flags
are dropped due to their small variability)
Total number of packets
Number of bytes transferred for each packet (bytes / packet)
Percent of control packets per session
Builds clusters for normal behavior
Multivariate normal distribution to model clusters
For each test point compute Mahalanobis distance to
every cluster
* C. Taylor and J. Alves-Foss. NATE - Network Analysis of Anomalous Traffic Events,
A Low-Cost Approach, Proc. New Security Paradigms Workshop. 2001.
Summarization of Anomalous Connections*Summarization of Anomalous Connections*
MINDS example: January 26, 2003 (48 hours after the Slammer worm)
score srcIP sPort dstIP dPort protocoflagspackets bytes
37674.69 63.150.X.253 1161 128.101.X.29 1434 17 16 [0,2) [0,1829)
26676.62 63.150.X.253 1161 160.94.X.134 1434 17 16 [0,2) [0,1829)
24323.55 63.150.X.253 1161 128.101.X.185 1434 17 16 [0,2) [0,1829)
21169.49 63.150.X.253 1161 160.94.X.71 1434 17 16 [0,2) [0,1829)
19525.31 63.150.X.253 1161 160.94.X.19 1434 17 16 [0,2) [0,1829)
19235.39 63.150.X.253 1161 160.94.X.80 1434 17 16 [0,2) [0,1829)
17679.1 63.150.X.253 1161 160.94.X.220 1434 17 16 [0,2) [0,1829)
8183.58 63.150.X.253 1161 128.101.X.108 1434 17 16 [0,2) [0,1829)
7142.98 63.150.X.253 1161 128.101.X.223 1434 17 16 [0,2) [0,1829)
5139.01 63.150.X.253 1161 128.101.X.142 1434 17 16 [0,2) [0,1829)
4048.49 142.150.Y.101 0 128.101.X.127 2048 1 16 [2,4) [0,1829)
4008.35 200.250.Z.20 27016 128.101.X.116 4629 17 16 [2,4) [0,1829)
3657.23 202.175.Z.237 27016 128.101.X.116 4148 17 16 [2,4) [0,1829)
3450.9 63.150.X.253 1161 128.101.X.62 1434 17 16 [0,2) [0,1829)
3327.98 63.150.X.253 1161 160.94.X.223 1434 17 16 [0,2) [0,1829)
2796.13 63.150.X.253 1161 128.101.X.241 1434 17 16 [0,2) [0,1829)
2693.88 142.150.Y.101 0 128.101.X.168 2048 1 16 [2,4) [0,1829)
2683.05 63.150.X.253 1161 160.94.X.43 1434 17 16 [0,2) [0,1829)
2444.16 142.150.Y.236 0 128.101.X.240 2048 1 16 [2,4) [0,1829)
2385.42 142.150.Y.101 0 128.101.X.45 2048 1 16 [0,2) [0,1829)
2114.41 63.150.X.253 1161 160.94.X.183 1434 17 16 [0,2) [0,1829)
2057.15 142.150.Y.101 0 128.101.X.161 2048 1 16 [0,2) [0,1829)
1919.54 142.150.Y.101 0 128.101.X.99 2048 1 16 [2,4) [0,1829)
1634.38 142.150.Y.101 0 128.101.X.219 2048 1 16 [2,4) [0,1829)
1596.26 63.150.X.253 1161 128.101.X.160 1434 17 16 [0,2) [0,1829)
1513.96 142.150.Y.107 0 128.101.X.2 2048 1 16 [0,2) [0,1829)
1389.09 63.150.X.253 1161 128.101.X.30 1434 17 16 [0,2) [0,1829)
1315.88 63.150.X.253 1161 128.101.X.40 1434 17 16 [0,2) [0,1829)
1279.75 142.150.Y.103 0 128.101.X.202 2048 1 16 [0,2) [0,1829)
1237.97 63.150.X.253 1161 160.94.X.32 1434 17 16 [0,2) [0,1829)
1180.82 63.150.X.253 1161 128.101.X.61 1434 17 16 [0,2) [0,1829)
1107.78 63.150.X.253 1161 160.94.X.154 1434 17 16 [0,2) [0,1829)
Potential Rules:
1.
{Dest Port = 1434/UDP
#packets ∈ [0, 2)} -->
Highly anomalous behavior
(Slammer Worm)
2.
{Src IP = 142.150.Y.101,
Dest Port = 2048/ICMP
#bytes ∈ [0, 1829]} -->
Highly anomalous behavior
(ping – scan)
Potential Rules:
1.
{Dest Port = 1434/UDP
#packets ∈ [0, 2)} -->
Highly anomalous behavior
(Slammer Worm)
2.
{Src IP = 142.150.Y.101,
Dest Port = 2048/ICMP
#bytes ∈ [0, 1829]} -->
Highly anomalous behavior
(ping – scan)
* - Ertoz, L., Eilertson, E., Lazarevic, et al, The MINDS - Minnesota Intrusion Detection System, review
for the book “Data Mining: Next Generation Challenges and Future Directions”, AAAI/MIT Press.
Handouts
– 47 –
Profiling based anomaly detectionProfiling based anomaly detection
Profiling methods are usually applied to host based
intrusion detection where users, programs, etc. are
profiled
Profiling sequences of Unix shell command lines*
Set of sequences (user profiles) reduced and filtered to reduce data
set for analysis
Build Instance Based Learning (IBL) model that stores historic
examples of “normal” data
Compares new data stream
Distance measure that favors long
temporal similar sequences
Event sequences are segmented
Profiling users’ behavior
Using neural networks
a b c d
a g c e
a b c d
g b c d
* T. Lane, C. Brodley, Temporal Sequence Learning and Data Reduction
for Anomaly detection, 1998.
Case Study: Data Mining in Fraud DetectionCase Study: Data Mining in Fraud Detection
Fraud detection in E-commerce/ business world
Credit Card Fraud
Internet Transaction Fraud / E-Cash fraud
Insurance Fraud and Health Care Fraud
Money Laundering
Telecommunications Fraud
Subscription Fraud / Identity Theft
All major data miming techniques applicable to intrusion
detection are also applicable to fraud detection domains
mentioned above
Handouts
– 48 –
Data Mining in Credit Fraud DetectionData Mining in Credit Fraud Detection
Credit Card Companies Turn To Artificial
Intelligence - “Credit card fraud costs the industry
about a $billion a year, or 7 cents out of every
$100 spent. But that is down significantly from its
peak about a decade ago, Sorrentino says, in
large part because of powerful technology that
can recognize unusual spending patterns.“
Data Mining in Insurance Fraud DetectionData Mining in Insurance Fraud Detection
Smart Tools
Banks, brokerages, and insurance companies have been relying
on various AI tools for two decades. One variety, called a neural
network, has become the standard for detecting credit-card fraud.
Since 1992, neural nets have slashed such incidents by 70% or
more for the likes of U.S. Bancorp and Wachovia Bank. Now, even
small credit unions are required to use the software in order to
qualify for debit-card insurance from Credit Union National Assn.
Innovative Use of Artificial Intelligence Monitoring NASDAQ for
Potential Insider Trading and Fraud (September 17, 2003)
Handouts
– 49 –
Data Mining in Cell Fraud DetectionData Mining in Cell Fraud Detection
Phone Friend - "More than 15,000 mobile phones are
stolen each month in Britain alone. According to Swedish
cellphone maker Ericsson, the fraudulent use of stolen
mobiles means a loss of between two and five per cent of
revenue for the network operators.“
Ericsson Enlists AI To Fight Wireless Fraud
Mobile network operators claim that 2 to 5 percent of total
revenues are lost to fraud, and the problem is expected to worsen
Data Mining Techniques in Fraud DetectionData Mining Techniques in Fraud Detection
Basic approaches
JAM project [Chan99]
Neural networks [Brauser99, Maes00, Dorronsoro97]
Adaptive fraud detection [Fawcett97]
Account signatures [Cahill02]
Statistical Fraud Detection [Bolton 2002]
Fraud detection in mobile telecommunication
networks [Burge96]
User profiling and classification for fraud detection in
mobile telecommunication networks [Hollmen00]
Handouts
– 50 –
JAM Project*JAM Project*
JAVA agents for meta-learning
Launch learning agents to
remote data sites
Exchange remote classifiers
Local meta-learning agents
produce meta-classifier
Launch meta-classifier to remote data sites
Chase credit card (20% fraud), First Union credit card
(15% fraud) – 500,000 data records
5 base classifiers (Bayes, C4.5, CART, ID3, RIPPER)
Select the classifier with the highest TP-FP rate on the
validation set
* JAM Project, http://www1.cs.columbia.edu/~sal/JAM/PROJECT/
JAM ProjectJAM Project -- ExampleExample
TP-FP rate vs. the number of classifiers
• Input base classifiers:
First Union
• Test data set:
First Union
• Best meta classifier:
Naïve Bayes with 10-17 base
classifiers.
Handouts
– 51 –
Neural Networks in Fraud Detection*Neural Networks in Fraud Detection*
Use an expert net for each feature group
(e.g. time,money) and group the experts
together to form a common vote*
Standard application of neural networks
in credit card fraud detection **
Use a non-linear version of Fisher’s
discriminant analysis, which
separates a good proportion
of fraudulent operations away
from other closer to normal traffic
* R. Brause,T. Langsdorf, M. Hepp, Neural Data Mining for Credit Card Fraud Detection,
11th IEEE International Conference on Tools with Artificial Intelligence, 1999.
** S. Maes, et al, Credit Card Fraud Detection Using Bayesian and Neural Networks,
NAISO Congress on NEURO FUZZY THECHNOLOGIES, 2002
*** J. Dorronsoro, F. Ginel, C. Sánchez, C. Santa Cruz, Neural Fraud Detection in Credit
Card Operations, IEEE Trans. Neural Networks, 8 (1997), pp. 827-834
Fraud Detection in Telecommunication NetworksFraud Detection in Telecommunication Networks
Identify relevant user groups based on call data
and then assign a user to a relevant group
call data is subsequently used in describing
behavioral patterns of users
Neural networks and probabilistic models are
employed in learning these usage patterns from
call data
These models are used either to detect abrupt
changes in established usage patterns or to
recognize typical usage patterns of fraud
• J. Hollmen, User profiling and classification for fraud detection in mobile
communications networks, PhD thesis, Helsinki University of Technology, 2000.
• P. Burge, J. Shawe-Taylor, Frameworks for Fraud Detection in mobile
communications networks, 1996.
Handouts
– 52 –
Case Study: Medical DiagnosisCase Study: Medical Diagnosis
Cost of false positive error: Unnecessary
treatment; unnecessary worry
Cost of false negative error: Postponed
treatment or failure to treat; death or injury
Event Rate Reference
Bleeding in outpatients on warfarin 6.7%/yr Beyth 1998
Medication errors per order 5.3% (n=10070) Bates 1995
Medication error per order 0.3% (n=289000) Lesar 1997
Adverse drug events per admission 6.5% Bates 1995
Adverse drug events per order 0.05% (n=10070) Bates 1995
Adverse drug events per patient 6.7% Lazarou 1998
Adverse drug events per patient 1.2% Bains 1999
Adverse drug events in nursing home patients 1.89/100 pt-months Gurwitz
Nosocomial infections in
older hospitalized patients
5.9 to 16.9 per 1000 days Rothchild 2000
Pressure ulcers 5% Rothschild 2000
Data Mining in Medical DiagnosisData Mining in Medical Diagnosis
Introducing noise to rare class examples*
For each attribute i, noisy example = xi + εi,
where , Σq is qxq diagonal matrix
diag{s1
2, …, sq
2}, si
2 – sample variance of xi
Hierarchical neural nets (HNNs)**
HNNs are designed according to a
divide-and-conquer approach
Triage networks are able to discriminate
supersets that contain the infrequent
pattern
These supersets are then used by specialized networks
),,0( 2
qnoisei N Σ≈ σε
* S. Lee, Noisy replication in skewed binary classification, Computational Statistics &
Data Analysis August, 2000.
** L.Machado, Identification of Low Frequency Patterns in Backpropagation Neural
Networks, Annual Symposium on Computer Applications in Medical Care, 1994.
Handouts
– 53 –
Data Mining in Medical DiagnosisData Mining in Medical Diagnosis
Spectrum and response analysis for neural nets*
Data: polyproline type II (PPII) secondary structure
PPII structure is rare (1.26%)
Around an individual case (sequence) there is a hyper-sphere
with a high probability area for the same secondary structure
type
Sample stratification schemes for neural networks (NNs)**
1. stratifies a sample by adding up the weighted sum of the
derivatives during the backward pass of training
2. After training NNs with multiple sets of bootstrapped examples of
the rare event classes and subsampled examples of common
event classes, multiple voting for classification is performed
* M. Siermala, Local Prediction of Secondary Structures of Proteins from Viewpoints of
Rare Structure, PhD Thesis, University of Tampere, May 2002.
** W. Choe et al, Neural network schemes for detecting rare events in human genomic
DNA, Bioinformatics. 2000.
ConclusionsConclusions
Data mining analysis of rare events requires
special attention
Many real world applications exhibit “needle-inthe-haystack”
type of problem
Current “state of the art” data mining techniques
are still insufficient for efficient handling rare
events
Need for designing better and more accurate
data mining models
Handouts
– 54 –
LinksLinks
Intrusion Detection bibliography
www.cs.umn.edu/~aleks/intrusion_detection.html
www.cs.fit.edu/~pkc/id/related/index.html
www.cc.gatech.edu/~wenke/ids-readings.html
www.cerias.purdue.edu/coast/intrusion-detection/welcome.html
http://cnscenter.future.co.kr/security/ids.html
www.cs.purdue.edu/homes/clifton/cs590m/
www.cs.ucsb.edu/~rsg/STAT/links.html
Fraud detection bibliography
www.hpl.hp.com/personal/Tom_Fawcett/fraud-public.bib.gz
http://dinkla.net !!!!!
http://www.aaai.org/AITopics/html/fraud.html
Fraud detection solutions
www.kdnuggets.com/solutions/fraud-detection.html
Questions?Questions?
Thank You!
Contact: aleks@cs.umn.edu
srivasta@cs.umn.edu
kumar@cs.umn.edu