Midterm: Review

Lecturer: Dr. Nguyen Thi Ngoc Anh

Email: [email protected]

Seminar: Support Vector Machines

— Massive Data Mining via Support Vector
Machines
— Support Vector Machines for:
◦ classifying from large datasets
◦ single-class classification
◦ discriminant feature combination discovery

1
Data Mining: Classification Schemes
— General functionality
◦ Descriptive data mining
◦ Predictive data mining
— Different views, different classifications
◦ Kinds of data to be mined
◦ Kinds of knowledge to be discovered
◦ Kinds of techniques utilized
◦ Kinds of applications adapted

Knowledge Discovery in Databases:

Process
Interpretation/
Evaluation

Data Mining Knowledge

Preprocessing
Patterns

Selection
Preprocessed
Data
Data
Target
Data

adapted from:
U. Fayyad, et al. (1995), “From Knowledge Discovery to Data
Mining: An Overview,” Advances in Knowledge Discovery and
Data Mining, U. Fayyad et al. (Eds.), AAAI/MIT Press

2
 What Can Data Mining Do?
— Cluster
— Classify
◦ Categorical, Regression
— Summarize
◦ Summary statistics, Summary rules
— Link Analysis / Model Dependencies
◦ Association rules
— Sequence analysis
◦ Time-series analysis, Sequential associations
— Detect Deviations

What is Data Warehouse?

— Defined in many different ways, but not rigorously.
◦ A decision support database that is maintained separately from the
organization’s operational database

◦ Support information processing by providing a solid platform of consolidated,

historical data for analysis.

— “A data warehouse is a subject-oriented, integrated, time-variant, and

nonvolatile collection of data in support of management’s decision-making
process.”—W. H. Inmon
— Data warehousing:
◦ The process of constructing and using data warehouses

3
 Example of Star Schema
time
time_key item
day item_key
day_of_the_week Sales Fact Table item_name
month brand
quarter time_key type
year supplier_type
item_key
branch_key
branch location
location_key
branch_key location_key
branch_name units_sold street
branch_type city
dollars_sold state_or_province
country
avg_sales
Measures
7

From Tables and Spreadsheets to Data

Cubes
— A data warehouse is based on a multidimensional data model which views
data in the form of a data cube
— A data cube, such as sales, allows data to be modeled and viewed in
multiple dimensions
◦ Dimension tables, such as item (item_name, brand, type), or time(day, week,
month, quarter, year)
◦ Fact table contains measures (such as dollars_sold) and keys to each of the
related dimension tables
— In data warehousing literature, an n-D base cube is called a base cuboid.
The top most 0-D cuboid, which holds the highest-level of summarization,
is called the apex cuboid. The lattice of cuboids forms a data cube.

4
 Cube: A Lattice of Cuboids
all
0-D(apex) cuboid

time item location supplier

1-D cuboids

time,location item,location location,supplier

time,item 2-D cuboids
time,supplier item,supplier

time,location,supplier
3-D cuboids
time,item,location
time,item,supplier item,location,supplier

4-D(base) cuboid
time, item, location, supplier

A Sample Data Cube

Total annual sales
Date of TVs in U.S.A.
1Qtr 2Qtr 3Qtr 4Qtr sum
TV
PC U.S.A
VCR
Country

sum
Canada

Mexico

sum

10

5
 Warehouse Summary
— Data warehouse
— A multi-dimensional model of a data warehouse
◦ Star schema, snowflake schema, fact constellations
◦ A data cube consists of dimensions & measures
— OLAP operations: drilling, rolling, slicing, dicing and
pivoting
— OLAP servers: ROLAP, MOLAP, HOLAP
— Efficient computation of data cubes
◦ Partial vs. full vs. no materialization
◦ Multiway array aggregation
◦ Bitmap index and join index implementations
— Further development of data cube technology
◦ Discovery-drive and multi-feature cubes
◦ From OLAP to OLAM (on-line analytical mining)

11

Data Preprocessing
— Data in the real world is dirty
◦ incomplete: lacking attribute values, lacking certain
attributes of interest, or containing only aggregate
data
– e.g., occupation=“”
◦ noisy: containing errors or outliers
– e.g., Salary=“-10”
◦ inconsistent: containing discrepancies in codes or
names
– e.g., Age=“42” Birthday=“03/07/1997”
– e.g., Was rating “1,2,3”, now rating “A, B, C”
– e.g., discrepancy between duplicate records

12

6
Multi-Dimensional Measure of Data
Quality
— A well-accepted multidimensional view:
◦ Accuracy
◦ Completeness
◦ Consistency
◦ Timeliness
◦ Believability
◦ Value added
◦ Interpretability
◦ Accessibility
— Broad categories:
◦ intrinsic, contextual, representational, and accessibility.

14

Major Tasks in Data Preprocessing

— Data cleaning
◦ Fill in missing values, smooth noisy data, identify or remove
outliers, and resolve inconsistencies
— Data integration
◦ Integration of multiple databases, data cubes, or files
— Data transformation
◦ Normalization and aggregation
— Data reduction
◦ Obtains reduced representation in volume but produces the
same or similar analytical results
— Data discretization
◦ Part of data reduction but with particular importance, especially
for numerical data

15

7
How to Handle Missing Data?
— Ignore the tuple: usually done when class label is missing
(assuming the tasks in classification—not effective when
the percentage of missing values per attribute varies
considerably.
— Fill in the missing value manually: tedious + infeasible?
— Fill in it automatically with
◦ a global constant : e.g., “unknown”, a new class?!
◦ the attribute mean
◦ the attribute mean for all samples belonging to the same class:
smarter
◦ the most probable value: inference-based such as Bayesian
formula or decision tree

16

How to Handle Noisy Data?

— Binning method:
◦ first sort data and partition into (equi-depth) bins
◦ then one can smooth by bin means, smooth by bin
median, smooth by bin boundaries, etc.
— Clustering
◦ detect and remove outliers
— Combined computer and human inspection
◦ detect suspicious values and check by human (e.g.,
deal with possible outliers)
— Regression
◦ smooth by fitting the data into regression functions

17

8
Data Transformation
— Smoothing: remove noise from data
— Aggregation: summarization, data cube
construction
— Generalization: concept hierarchy climbing
— Normalization: scaled to fall within a small,
specified range
◦ min-max normalization
◦ z-score normalization
◦ normalization by decimal scaling
— Attribute/feature construction
◦ New attributes constructed from the given ones

18

Data Reduction Strategies

— A data warehouse may store terabytes of data
◦ Complex data analysis/mining may take a very long time to run
on the complete data set
— Data reduction
◦ Obtain a reduced representation of the data set that is much
smaller in volume but yet produce the same (or almost the
same) analytical results
— Data reduction strategies
◦ Data cube aggregation
◦ Dimensionality reduction — remove unimportant attributes
◦ Data Compression
◦ Numerosity reduction — fit data into models
◦ Discretization and concept hierarchy generation

19

9
Principal Component Analysis
— Given N data vectors from k-dimensions, find c
≤ k orthogonal vectors that can be best used
to represent data
◦ The original data set is reduced to one consisting of
N data vectors on c principal components (reduced
dimensions)
— Each data vector is a linear combination of the c
principal component vectors
— Works for numeric data only
— Used when the number of dimensions is large

20

Discretization
— Three types of attributes:
◦ Nominal — values from an unordered set
◦ Ordinal — values from an ordered set
◦ Continuous — real numbers
— Discretization:
◦ divide the range of a continuous attribute into
intervals
◦ Some classification algorithms only accept categorical
attributes.
◦ Reduce data size by discretization
◦ Prepare for further analysis

21

10
Data Preparation Summary
— Data preparation is a big issue for both
warehousing and mining
— Data preparation includes
◦ Data cleaning and data integration
◦ Data reduction and feature selection
◦ Discretization
— A lot a methods have been developed but
still an active area of research

22

Association Rule Mining

— Finding frequent patterns, associations, correlations, or
causal structures among sets of items or objects in
transaction databases, relational databases, and other
information repositories.
◦ Frequent pattern: pattern (set of items, sequence, etc.) that
occurs frequently in a database [AIS93]
— Motivation: finding regularities in data
◦ What products were often purchased together? — Beer and
diapers?!
◦ What are the subsequent purchases after buying a PC?
◦ What kinds of DNA are sensitive to this new drug?
◦ Can we automatically classify web documents?

23

11
 Basic Concepts:
Association Rules
Transaction-id Items bought — Itemset X={x1, …, xk}
10 A, B, C — Find all the rules XàY with min
confidence and support
20 A, C
◦ support, s, probability that a
30 A, D transaction contains XÈY
40 B, E, F ◦ confidence, c, conditional
probability that a transaction
having X also contains Y.
Customer Customer
buys both buys diaper
Let min_support = 50%,
min_conf = 50%:
A à C (50%, 66.7%)
C à A (50%, 100%)
Customer
buys beer
24

The Apriori Algorithm—An Example

Itemset sup
Itemset sup
Database TDB {A} 2 L1 {A} 2
Tid Items C1 {B} 3
{B} 3
10 A, C, D {C} 3
20 B, C, E
1st scan {C} 3
{D} 1
{E} 3
30 A, B, C, E {E} 3
40 B, E
CFrequency
2 ≥ 50%, Confidence
Itemset sup C2 Itemset 100%:
{A, B} 1
L2 Itemset sup
{A, C}
nd à C {A, B}
2A scan
2
{A, C} 2
{B, C} 2
{A, E} 1 BàE {A, C}
{B, C} 2 BC à E {A, E}
{B, E} 3
{B, E} 3 {B, C}
{C, E} 2
{C, E} 2
CE à B {B, E}
BE à C {C, E}
C3 Itemset
3rd scan L3 Itemset sup
{B, C, E} {B, C, E} 2
26

12
 FP-Tree Algorithm
TID Items bought (ordered) frequent items
100 {f, a, c, d, g, i, m, p} {f, c, a, m, p}
200 {a, b, c, f, l, m, o} {f, c, a, b, m}
300 {b, f, h, j, o, w} {f, b} min_support = 3
400 {b, c, k, s, p} {c, b, p}
500 {a, f, c, e, l, p, m, n} {f, c, a, m, p}
{}
1. Scan DB once, find Header Table
frequent 1-itemset f:4 c:1
Item frequency head
(single item pattern) f 4
2. Sort frequent items in c 4 c:3 b:1 b:1
frequency descending a 3
order, f-list b 3 a:3 p:1
m 3
3. Scan DB again, p 3
m:2 b:1
construct FP-tree
F-list=f-c-a-b-m-p p:2 m:1 27

Constrained Frequent Pattern Mining: A

Mining Query Optimization Problem
— Given a frequent pattern mining query with a set of constraints C,
the algorithm should be
◦ sound: it only finds frequent sets that satisfy the given
constraints C
◦ complete: all frequent sets satisfying the given constraints C are
found
— A naïve solution
◦ First find all frequent sets, and then test them for constraint
satisfaction
— More efficient approaches:
◦ Analyze the properties of constraints comprehensively
◦ Push them as deeply as possible inside the frequent pattern
computation.

28

13
 Classification:
Model Construction
Classification
Algorithms
Training
Data

NAME RANK YEARS TENURED Classifier

Mike Assistant Prof 3 no (Model)
Mary Assistant Prof 7 yes
Bill Professor 2 yes
Jim Associate Prof 7 yes
IF rank = ‘professor’
Dave Assistant Prof 6 no
OR years > 6
Anne Associate Prof 3 no
THEN tenured = ‘yes’
29

Classification:
Use the Model in Prediction

Classifier

Testing
Data Unseen Data

(Jeff, Professor, 4)
NAME RANK YEARS TENURED
Tom Assistant Prof 2 no Tenured?
Merlisa Associate Prof 7 no
George Professor 5 yes
Joseph Assistant Prof 7 yes
30

14
 Naïve Bayes Classifier
— A simplified assumption: attributes are conditionally
independent:
n
P( X | C i) = Õ P( x k | C i)
k =1
— The product of occurrence of say 2 elements x1 and x2,
given the current class is C, is the product of the
probabilities of each element taken separately, given the
same class P([y1,y2],C) = P(y1,C) * P(y2,C)
— No dependence relation between attributes
— Greatly reduces the computation cost, only count the
class distribution.
— Once the probability P(X|Ci) is known, assign X to the
class with maximum P(X|Ci)*P(Ci)

31

Bayesian Belief Network

Family
Smoker
History
(FH, S) (FH, ~S) (~FH, S) (~FH, ~S)

LC 0.8 0.5 0.7 0.1

~LC 0.2 0.5 0.3 0.9
LungCancer Emphysema

The conditional probability table

for the variable LungCancer:
Shows the conditional probability
PositiveXRay Dyspnea for each possible combination of its
parents n
P ( z1,..., zn) = Õ P ( z i | Parents ( Z i ))
Bayesian Belief Networks i =1

32

15
 Decision Tree

age?

<=30 overcast
30..40 >40

student? yes credit rating?

no yes excellent fair

no yes no yes

33

Algorithm for Decision Tree Induction

— Basic algorithm (a greedy algorithm)
◦ Tree is constructed in a top-down recursive divide-and-conquer manner
◦ At start, all the training examples are at the root
◦ Attributes are categorical (if continuous-valued, they are discretized in
advance)
◦ Examples are partitioned recursively based on selected attributes
◦ Test attributes are selected on the basis of a heuristic or statistical
measure (e.g., information gain)
— Conditions for stopping partitioning
◦ All samples for a given node belong to the same class
◦ There are no remaining attributes for further partitioning – majority
voting is employed for classifying the leaf
◦ There are no samples left

34

16
 Attribute Selection Measure:
Information Gain (ID3/C4.5)
n Select the attribute with the highest information gain
n S contains si tuples of class Ci for i = {1, …, m}
n information measures info required to classify any
arbitrary tuple m
si si
I( s1,s2,...,sm ) = -å log 2
i =1 s s
n entropy of attribute A with values {a1,a2,…,av}
v
s1 j + ... + smj
E(A) = å I ( s1 j ,..., smj )
j =1 s

n information gained by branching on attribute A

Gain(A) = I(s 1, s 2 ,..., sm) - E(A)

35

Definition of Entropy
— Entropy H (X ) = å - P( x) log
xÎ AX
2 P( x)

— Example: Coin Flip

◦ AX = {heads, tails}
◦ P(heads) = P(tails) = ½
◦ ½ log2(½) = ½ * - 1
◦ H(X) = 1
— What about a two-headed coin?
— Conditional Entropy:
H (X |Y) = å P( y ) H ( X | y )
yÎ AY 36

17
 Attribute Selection by Information
Gain Computation
g Class P: buys_computer = “yes”
5 4
g Class N: buys_computer = “no” E (age) = I (2,3) + I (4,0)
g I(p, n) = I(9, 5) =0.940 14 14
g Compute the entropy for age: 5
+ I (3,2) = 0.694
14
age pi ni I(pi, ni) 5
I (2,3) means “age <=30” has 5 out
<=30 2 3 0.971 14
of 14 samples, with 2 yes’es and 3
30…40 4 0 0
no’s. Hence
>40 3 2 0.971
age
<=30
income student credit_rating
high no fair
buys_computer
no
Gain(age) = I ( p, n) - E (age) = 0.246
<=30 high no excellent no
31…40 high no fair yes
>40 medium no fair yes
Similarly,
>40 low yes fair yes
>40 low yes excellent no Gain(income) = 0.029
31…40 low yes excellent yes
<=30
<=30
medium
low
no
yes
fair
fair
no
yes
Gain( student ) = 0.151
>40
<=30
medium
medium
yes
yes
fair
excellent
yes
yes
Gain(credit _ rating ) = 0.048
31…40 medium no excellent yes
31…40 high yes fair yes
37
>40 medium no excellent no

Overfitting in Decision Trees

— Overfitting: An induced tree may overfit the training
data
◦ Too many branches, some may reflect anomalies due to noise or
outliers
◦ Poor accuracy for unseen samples
— Two approaches to avoid overfitting
◦ Prepruning: Halt tree construction early—do not split a node if
this would result in the goodness measure falling below a
threshold
– Difficult to choose an appropriate threshold
◦ Postpruning: Remove branches from a “fully grown” tree—get a
sequence of progressively pruned trees
– Use a set of data different from the training data to decide
which is the “best pruned tree”

38

18
 Artificial Neural Networks:
A Neuron
- mk
x0 w0
x1 w1
å f
output y
xn wn

Input weight weighted Activation

vector x vector w sum function
— The n-dimensional input vector x is mapped into variable y
by means of the scalar product and a nonlinear function
mapping
39

Artificial Neural Networks: Training

— The ultimate objective of training
◦ obtain a set of weights that makes almost all the tuples in the
training data classified correctly
— Steps
◦ Initialize weights with random values
◦ Feed the input tuples into the network one by one
◦ For each unit
– Compute the net input to the unit as a linear combination of all the
inputs to the unit
– Compute the output value using the activation function
– Compute the error
– Update the weights and the bias

40

19
 SVM – Support Vector Machines

Small Margin Large Margin

Support Vectors

41

Non-separable case
When the data set is
non-separable as
shown in the right xT b + b 0 = 0
figure, we will assign
X
weight to each x*
X
support vector which X
will be shown in the X
constraint. C

42

20
 Non-separable Cont.
1. Constraint changes to the following:
y ( xT b + b ), > C (1 - x ),Where
i i 0 i
N
"i, xi > 0, å xi < const.
i =1
2. Thus the optimization problem changes to:

Min|| b
ì
|| subject toï y ( xT b + b
i i 0 ) >1-xi ,i =1,..., N .
í N

ïî "i ,xi >0,å

i =1
xi < const .

43

General SVM

This classification problem

clearly do not have a good
optimal linear classifier.

Can we do better?
A non-linear boundary as
shown will do fine.

44

21
General SVM Cont.
— The idea is to map the feature space into
a much bigger space so that the boundary
is linear in the new space.
— Generally linear boundaries in the
enlarged space achieve better training-
class separation, and it translates to non-
linear boundaries in the original space.

45

Mapping
— Mapping F : � d a H
◦ Need distances in H: F ( xi ) × F ( x j )
— Kernel Function: K ( xi , x j ) = F ( xi ) × F ( x j )
-|| x - x ||2 / 2s 2
◦ Example: K ( xi , x j ) = e i j
— In this example, H is infinite-dimensional

46

22
 The k-Nearest Neighbor Algorithm
— All instances correspond to points in the n-D space.
— The nearest neighbor are defined in terms of Euclidean
distance.
— The target function could be discrete- or real- valued.
— For discrete-valued, the k-NN returns the most common
value among the k training examples nearest to xq.
— Voronoi diagram: the decision surface induced by 1-NN for a
typical set of training examples.
_
_
_ _ .
+
_ .
+
xq + . . .
_ + .
47

Case-Based Reasoning
— Also uses: lazy evaluation + analyze similar instances
— Difference: Instances are not “points in a Euclidean
space”
— Example:Water faucet problem in CADET (Sycara et
al’92)
— Methodology
◦ Instances represented by rich symbolic descriptions (e.g.,
function graphs)
◦ Multiple retrieved cases may be combined
◦ Tight coupling between case retrieval, knowledge-based
reasoning, and problem solving
— Research issues
◦ Indexing based on syntactic similarity measure, and when failure,
backtracking, and adapting to additional cases

48

23
Regress Analysis and Log-Linear
Models in Prediction
— Linear regression:Y = a + b X
◦ Two parameters , a and b specify the line and are to
be estimated by using the data at hand.
◦ using the least squares criterion to the known values
of Y1,Y2, …, X1, X2, ….
— Multiple regression:Y = b0 + b1 X1 + b2 X2.
◦ Many nonlinear functions can be transformed into the
above.
— Log-linear models:
◦ The multi-way table of joint probabilities is
approximated by a product of lower-order tables.
◦ Probability: p(a, b, c, d) = aab baccad dbcd

49

Bagging and Boosting

— General idea
Classification method (CM)
Training data Classifier C
CM

Altered Training data Classifier C1

CM

Altered Training data Classifier C2

……..
Aggregation …. Classifier C*

50

24
Test Taking Hints
— Open book/notes
◦ Pretty much any non-electronic aid allowed
— See old copies of my exams (and
solutions) at my web site
◦ CS 526
◦ CS 541
◦ CS 603
— Time will be tight
◦ Suggested “time on question” provided

51

Seminar Thursday:
Support Vector Machines
— Massive Data Mining via Support Vector
Machines
— Support Vector Machines for:
◦ classifying from large datasets
◦ single-class classification
◦ discriminant feature combination discovery

52

25