Introduction to

Data Mining

Intro to Data Mining 1

What to Cover

 Motivation: Why data mining?

 What is data mining?
 Data Mining: On what kind of data?
 Data mining functionality
 Are all the patterns interesting?
 Classification of data mining systems
 Major issues in data mining
Intro to Data Mining 2
 Necessity Is the Mother of
Invention
 Data explosion problem
 Automated data collection tools and mature database
technology lead to tremendous amounts of data
accumulated and/or to be analyzed in databases, data

warehouses, and other information repositories

 We are drowning in data, but starving for knowledge!

Intro to Data Mining 3

Solution
 Data warehousing and data mining
 Data warehousing and on-line analytical processing
 Mining interesting knowledge (rules, regularities, patterns,
constraints) from data in large databases

Intro to Data Mining 4

 Evolution of Database
Technology
 1960s:
 Data collection, database creation, IMS and network
DBMS
 1970s:
 Relational data model, relational DBMS implementation
 1980s:
 RDBMS, advanced data models (extended-relational,
OO, deductive, etc.)
 Application-oriented DBMS (spatial, scientific,
engineering, etc.)
Intro to Data Mining 5
Evolution..
 1990s:
 Data mining, data warehousing, multimedia
databases, and Web databases
 2000s
 Stream data management and mining
 Data mining with a variety of applications
 Web technology and global information systems

Intro to Data Mining 6

What Is Data Mining?
 Data Mining is the process of discovering new
correlations, patterns, and trends by digging into
(mining) large amounts of data stored in
warehouses, using artificial intelligence, statistical
and mathematical techniques.
 Data mining can also be defined as “The nontrivial
extraction of implicit, previously unknown, and
potentially useful information from data”.
 Alternative names
 Knowledge discovery (mining) in databases (KDD),
knowledge extraction, data/pattern analysis, etc.
Intro to Data Mining 7
Why Data Mining?
Potential Applications
 Data analysis and decision support
 Market analysis and management
 Target marketing, customer relationship management
(CRM), market basket analysis, cross selling, market
segmentation
 Risk analysis and management
 Forecasting, customer retention, improved underwriting,
quality control, competitive analysis
 Fraud detection and detection of unusual patterns
(outliers)
Intro to Data Mining 8
Why Data Mining?
 Other Applications
 Text mining (news group, email, documents) and Web
mining
 Stream data mining
 DNA and bio-data analysis

Intro to Data Mining 9

 Market Analysis and Management

 Where does the data come from?

 Credit card transactions, loyalty cards, discount coupons,
customer complaint calls, plus (public) lifestyle studies
 Target marketing
 Find clusters of “model” customers who share the same
characteristics: interest, income level, spending habits..
 Determine customer purchasing patterns over time
 Cross-market analysis
 Associations/co-relations between product sales, &
prediction based on such association

Intro to Data Mining 10

Market Analysis and Management
 Customer profiling
 What types of customers buy what products (clustering or
classification)
 Customer requirement analysis
 identifying the best products for different customers
 predict what factors will attract new customers
 Provision of summary information
 multidimensional summary reports
 statistical summary information (data central tendency and
variation)
Intro to Data Mining 11
Corporate Analysis & Risk
Management

 Finance planning and asset evaluation

 cash flow analysis and prediction
 contingent claim analysis to evaluate assets
 cross-sectional and time series analysis (financial-ratio, trend
analysis, etc.)
 Resource planning
 summarize and compare the resources and spending
 Competition
 monitor competitors and market directions
 group customers into classes and a class-based pricing
procedure
 set pricing strategy in a highly competitive market
Intro to Data Mining 12
Fraud Detection & Mining Unusual
Patterns
 Approaches: Clustering & model construction for frauds,
outlier analysis
 Applications: Health care, retail, credit card service,
telecomm.
 Auto insurance: ring of collisions
 Money laundering: suspicious monetary transactions
 Medical insurance
 Professional patients, ring of doctors, and ring of
references
 Unnecessary or correlated screening tests
Intro to Data Mining 13
Fraud Detection & Mining
Unusual Patterns
 Telecommunications: phone-call fraud
 Phone call model: destination of the call, duration,
time of day or week. Analyze patterns that deviate
from an expected norm
 Retail industry
 Analysts estimate that 38% of retail shrink is due to
dishonest employees
 Anti-terrorism

Intro to Data Mining 14

 Data Mining: A KDD Process

 Data mining—core of Pattern Evaluation

knowledge discovery process

Data Mining

Task-relevant Data

Data Selection
Warehouse
Data Cleaning

Data Integration

Intro to Data Mining 15

Databases
 Steps of a KDD Process
 Learning the application domain
 relevant prior knowledge and goals of application

 Creating a target data set: data selection

 Data cleaning and preprocessing: (may take 60% of
effort!)
 Data reduction and transformation
 Find useful features, dimensionality/variable reduction,
invariant representation.
 Choosing functions of data mining
 summarization, classification, regression, association,
clustering.
Intro to Data Mining 16
Steps of a KDD Process
 Choosing the mining algorithm(s)
 Data mining: search for patterns of interest
 Pattern evaluation and knowledge presentation
 visualization, transformation, removing redundant
patterns, etc.
 Use of discovered knowledge

Intro to Data Mining 17

Data Mining and Business
Intelligence
Increasing potential
to support
business decisions End User
Making
Decisions

Data Presentation Business

Analyst
Visualization Techniques
Data Mining Data
Information Discovery Analyst

Data Exploration
Statistical Analysis, Querying and Reporting
Data Warehouses / Data Marts
OLAP, MDA DBA
Data Sources
Paper, Files, InformationIntro
Providers,
to Data Mining Database Systems, OLTP 18
 Architecture: Typical Data
Mining System

Graphical user interface

Pattern evaluation

Data mining engine

Knowledge-base
Database or data
warehouse server
Data cleaning & data integration Filtering

Data
Databases Warehouse
Intro to Data Mining 19
 Data Mining: On What Kinds of
Data?
 Relational database
 Data warehouse
 Transactional database
 Advanced database and information repository
 Object-relational database

 Spatial and temporal data

 Time-series data

 Stream data

 Multimedia database

 Heterogeneous and legacy database

 Text databases & WWW

Intro to Data Mining 20
 Data Mining Functionalities
 Concept description: Characterization and discrimination
 Generalize, summarize, and contrast data characteristics
 Association (correlation and causality)
 Classification and Prediction
 Construct models (functions) that describe and distinguish
classes or concepts for future prediction
 Presentation: decision-tree, classification rule, neural
network
 Predict some unknown or missing numerical values

Intro to Data Mining 21

 Data Mining Functionalities
 Cluster analysis
 Class label is unknown: Group data to form new classes, e.g.,
cluster houses to find distribution patterns
 Maximizing intra-class similarity & minimizing interclass similarity
 Outlier analysis
 Outlier: a data object that does not comply with the general
behavior of the data
 Noise or exception? No! useful in fraud detection, rare events
analysis
 Trend and evolution analysis
 Trend and deviation: regression analysis
 Sequential pattern mining, periodicity analysis
 Similarity-based analysisIntro to Data Mining 22
Data Mining: Confluence of Multiple
Disciplines

Database
Statistics
Systems

Machine
Learning
Data Mining Visualization

Algorithm Other
Disciplines
Intro to Data Mining 23
 Major Issues in Data Mining
 Mining methodology
 Mining different kinds of knowledge from diverse data
types, e.g., bio, stream, Web
 Performance: efficiency, effectiveness, and scalability
 Pattern evaluation: the interestingness problem
 Incorporation of background knowledge
 Handling noise and incomplete data
 Parallel, distributed and incremental mining methods
 Integration of the discovered knowledge with existing
one: knowledge fusion
Intro to Data Mining 24
Major Issues in Data Mining
 User interaction
 Data mining query languages and ad-hoc mining
 Expression and visualization of data mining results
 Interactive mining of knowledge at multiple levels of
abstraction
 Applications and social impacts
 Domain-specific data mining & invisible data mining

 Protection of data security, integrity, and privacy

Intro to Data Mining 25

 Summary
 Data mining: discovering interesting patterns from large amounts
of data
 A natural evolution of database technology, in great demand, with
wide applications
 A KDD process includes data cleaning, data integration, data
selection, transformation, data mining, pattern evaluation, and
knowledge presentation
 Mining can be performed in a variety of information repositories
 Data mining functionalities: characterization, discrimination,
association, classification, clustering, outlier and trend analysis..
 Data mining systems and architectures
 Major issues in data miningIntro to Data Mining 26
 A Brief History of Data Mining
Society
 1989 IJCAI Workshop on Knowledge Discovery in Databases (Piatetsky-
Shapiro)
 Knowledge Discovery in Databases (G. Piatetsky-Shapiro and W. Frawley, 1991)

 1991-1994 Workshops on Knowledge Discovery in Databases

 Advances in Knowledge Discovery and Data Mining (U. Fayyad, G. Piatetsky-Shapiro, P.
Smyth, and R. Uthurusamy, 1996)

 1995-1998 International Conferences on Knowledge Discovery in

Databases and Data Mining (KDD’95-98)
 Journal of Data Mining and Knowledge Discovery (1997)

 1998 ACM SIGKDD, SIGKDD’1999-2001 conferences, and SIGKDD

Explorations
 More conferences on data mining
Intro to Data Mining 27
 PAKDD (1997), PKDD (1997), SIAM-Data Mining (2001), (IEEE) ICDM (2001), etc.
 Where to Find References?
 Data mining and KDD (SIGKDD: CDROM)
 Conferences: ACM-SIGKDD, IEEE-ICDM, SIAM-DM, PKDD, PAKDD, etc.
 Journal: Data Mining and Knowledge Discovery, KDD Explorations
 Database systems (SIGMOD: CD ROM)
 Conferences: ACM-SIGMOD, ACM-PODS, VLDB, IEEE-ICDE, EDBT, ICDT,
DASFAA
 Journals: ACM-TODS, IEEE-TKDE, JIIS, J. ACM, etc.
 AI & Machine Learning
 Conferences: Machine learning (ML), AAAI, IJCAI, COLT (Learning Theory), etc.
 Journals: Machine Learning, Artificial Intelligence, etc.
 Statistics
 Conferences: Joint Stat. Meeting, etc.
 Journals: Annals of statistics, etc.
 Visualization
 Conference proceedings: CHI, ACM-SIGGraph, etc.
 Journals: IEEE Trans. visualization
Intro toand
Data computer
Mining graphics, etc. 28
 Recommended Reference Books
 R. Agrawal, J. Han, and H. Mannila, Readings in Data Mining: A Database Perspective, Morgan
Kaufmann (in preparation)
 U. M. Fayyad, G. Piatetsky-Shapiro, P. Smyth, and R. Uthurusamy. Advances in Knowledge
Discovery and Data Mining. AAAI/MIT Press, 1996
 U. Fayyad, G. Grinstein, and A. Wierse, Information Visualization in Data Mining and Knowledge
Discovery, Morgan Kaufmann, 2001
 J. Han and M. Kamber. Data Mining: Concepts and Techniques. Morgan Kaufmann, 2001
 D. J. Hand, H. Mannila, and P. Smyth, Principles of Data Mining, MIT Press, 2001
 T. Hastie, R. Tibshirani, and J. Friedman, The Elements of Statistical Learning: Data Mining,
Inference, and Prediction, Springer-Verlag, 2001
 T. M. Mitchell, Machine Learning, McGraw Hill, 1997
 G. Piatetsky-Shapiro and W. J. Frawley. Knowledge Discovery in Databases. AAAI/MIT Press, 1991
 S. M. Weiss and N. Indurkhya, Predictive Data Mining, Morgan Kaufmann, 1998
 I. H. Witten and E. Frank, Data Mining: Practical Machine Learning Tools and Techniques with Java
Implementations, Morgan Kaufmann, 2001
Intro to Data Mining 29