Title: On Appropriate Assumptions to Mine Data Streams: Analyses and Solutions
1On Appropriate Assumptions to Mine Data Streams
Analyses and Solutions
- Jing Gao Wei Fan Jiawei Han
- University of Illinois at Urbana-Champaign
- IBM T. J. Watson Research Center
2Introduction (1)
- Data Stream
- Continuously arriving data flow
- Applications network traffic, credit card
transaction flow, phone calling records, etc.
3Introduction (2)
- Stream Classification
- Construct a classification model based on past
records - Use the model to predict labels for new data
- Help decision making
Classification model
Fraud?
Labeling
Fraud
4Framework
?
Classification Model
Predict
5Existing Stream Mining Methods
- How to use old examples?
- Throw away or fade out old examples
- Select old examples or models which match the
current concepts - How to update the model?
- Real Time Update
- Batch Update
Match the training distribution!
6Existing Stream Mining Methods
- Shared distribution assumption
- Training and test data are from the same
distribution P(x,y) x-feature vector, y-class
label - Validity of existing work relies on the shared
distribution assumption - Difference from traditional learning
- Both distributions evolve
training
test
7Appropriateness of Shared Distribution
- An example of stream data
- KDDCUP99 Intrusion Detection Data
- P(y) evolves
- Shift or delay inevitable
- The future data could be different from current
data - Matching the current distribution to fit the
future one is a wrong way - The shared distribution assumption is
inappropriate
8Appropriateness of Shared Distribution
- Changes in P(y)
- P(y) P(x,y)P(yx)P(x)
- The change in P(y) is attributed to changes in
P(yx) and P(x)
Time Stamp 1
Time Stamp 11
Time Stamp 21
9Realistic and relaxed assumption
The training and test distributions are similar
to the degree that the model trained from the
training set D has higher accuracy on the test
set T than both random guessing and predicting
the same class label.
Model
Training set
Test set
Random Guessing
Fixed Guessing
10Realistic and relaxed assumption
- Strengths of this assumption
- Does not assume any exact relationship between
training and test distribution - Simply assume that learning is useful
- Develop algorithms based on this assumption
- Maximize the chance for models to succeed on
future data instead of match current data
11A Robust and Extensible Stream Mining Framework
C1
Training set
Test set
C2
Ck
Simple Voting(SV)
Averaging Probability(AP)
12Why ensemble?
- Ensemble
- Reduce variance caused by single models
- Is more robust than single models when the
distribution is evolving - Expected error analysis
- Single model
- Ensemble
13Why simple averaging?
- Combining outputs
- Simple averaging uniform weights wi1/k
- Weighted ensemble non-uniform weights
- wi is inversely proportional to the training
errors - wi should reflect P(M), the probability of model
M after observing the data - Uniform weights are the best
- P(M) is changing and we could never estimate the
true P(M) and when and how it changes - Uniform weights could minimize the expected
distance between P(M) and weight vector
14An illustration
- Single models (M1, M2, M3) have huge variance.
- Simple averaging ensemble (AP) is more stable and
accurate. - Weighted ensemble (WE) is not as good as AP since
training errors and test errors may have
different distributions.
Single Models
Weighted Ensemble
Average Probability
15Experiments
- Set up
- Data streams with chunks T1, T2, , TN
- Use Ti as the training set to classify Ti1
- Measures
- Mean Squared Error, Accuracy
- Number of Wins, Number of Loses
- Normalized Accuracy, MSE
16Experiments
- Methods
- Single models Decision tree (DT), SVM, Logistic
Regression (LR) - Weighted ensemble weights reflect the accuracy
on training set (WE) - Simple ensemble voting (SV) or probability
averaging (AP)
17Experimental Results (1)
Time 40
Time 100
Comparison on Synthetic Data
18Experimental Results (2)
Comparison on Intrusion Data Set
19Experimental Results (3)
Classification Accuracy Comparison
20Experimental Results (4)
Mean Squared Error Comparison
21Conclusions
- Realistic assumption
- Take into account the difference between training
and test distributions - Overly matching the training distribution is thus
unsatisfactory - Model averaging
- Robust and accurate
- Theoretically proved the effectiveness
- Could give the best predictions on average
22Thanks!