1. Chapter 1 Introduction to Data Mining
Chen. Chun-Hsien
Department of Information Management
Chang Gung University
2018年8月23日星期四
Introduction to Data Mining
Introduction to DM & DW

2. Introduction to Data Mining
Outline
Motivation to data mining
What is data mining?
Applications of data mining
Data mining process
Main data mining techniques
Classification of data mining systems
2018年8月23日星期四
Introduction to Data Mining

3. Introduction to Data Mining
Motivation
Phenomenon : data explosion
(Automated data collection tools and mature database technology)
Tremendous amount of Web pages
40+ billion photos on Facebook
1 million new transactions/hour added in Walmart database
Data from wearable devices for healthcare
Big data in Clouds
Problem : we are drowning in data, but starving for knowledge
Solution : data Mining
One of the 10 emerging technologies that will change the world in the near future
2018年8月23日星期四
Introduction to Data Mining

4. Introduction to Data Mining
What Is Data Mining?
Formal Definition of Data mining
Automatic extraction of interesting (non-trivial, implicit, previously unknown and potentially useful) knowledge (rules, regularities, patterns, trends, affinities) from large amount of data
Alternative names
Business intelligence (BI), knowledge discovery in databases (KDD), data/pattern analysis, knowledge extraction, data dredging, information harvesting, data archeology, etc.
2018年8月23日星期四
Introduction to Data Mining

5. Example : Mining a Concept Hierarchy
all
all
Europe
...
North_America
region
Germany
...
Spain
Canada
...
Mexico
country
Vancouver
...
city
Frankfurt
...
Toronto
L. Chan
...
M. Wind
office
2018年8月23日星期四
Introduction to Data Mining

6. Part of International Sales, Shipping Data
(嘉義市是淘寶網2012年人均支付金額第一名: 32萬元)
2018年8月23日星期四
Introduction to Data Mining

7. Introduction to Data Mining
Confluence of Multiple Disciplines
Artificial
Intelligence
Statistics
Data Mining
Machine
Learning
Visualization
Information
Science
Database
Technology
KDD process
2018年8月23日星期四
Introduction to Data Mining

8. Introduction to Data Mining
Evolution of Database Technology
1960s:
Data collection, database creation, network DBMS
1970s:
Relational data model, relational DBMS
1980s:
Advanced data models (extended-relational, OO, spatial, temporal D/Bs, etc.)
1990s ~:
Data mining, data warehousing, multimedia D/B, and Web
2018年8月23日星期四
Introduction to Data Mining

9. Applications of Data Mining
Decision support
Business decision support
Consumer understanding and service improvement
Market trend analysis and management
Risk analysis and management
Fraud detection and management
Medical decision support
Other Applications
Text mining
Web analysis
Bioinformatics
2018年8月23日星期四
Introduction to Data Mining

10. Applications of Data Mining (Market Analysis and Management)
(1/2)
Data sources for analysis
Transactions of credit card, retail industry, etc.
Public lifestyle studies
Customer complaint calls
Market basket analysis and cross selling
Associations/co-relations between product sales
Prediction based on the association information
2018年8月23日星期四
Introduction to Data Mining

11. Applications of Data Mining (Market Analysis and Management)
(2/2)
Customer profiling
Find clusters of “model” customers who share the same characteristics: spending habits, income level, interest, etc.
Data mining can tell you what types of customers buy what products (by clustering or classification techniques)
Identifying customer requirements
Identifying potential product sales for eC customers
Use prediction to find what factors will attract new customers
2018年8月23日星期四
Introduction to Data Mining

12. Introduction to Data Mining
Applications of Data Mining (Risk Analysis and Management)
Finance planning and asset evaluation
Cash flow analysis and prediction
Asset evaluation
Time series analysis (trend analysis)
Competitive analysis and market segmentation
Monitoring competitors and market directions
Setting pricing strategy in a highly competitive market
Grouping customers/a class-based pricing procedure
2018年8月23日星期四
Introduction to Data Mining

13. Introduction to Data Mining
Applications of Data Mining (Fraud Detection and Management)
Applications
Health care, insurance, credit card services
Approach
use historical data to build models of fraudulent behavior, and use data mining techniques to help identify similar instances
Examples
Detection of money laundering:
Detect suspicious money transaction patterns in banks
Fraud detection of medical insurance:
Detect cheating ring of patients and doctors
2018年8月23日星期四
Introduction to Data Mining

14. Introduction to Data Mining
Applications of Data Mining (Other Applications)
Text Ming
News classification : find related articles
Detection of spam : analyze content
Medical informatics : automatic classification of cancer reports
Web Mining : mining web access logs
Discovering customer preference and behavior
Analyzing effectiveness of Web marketing
Improving Web site organization
Biomedical Informatics
Finding related genes of genetic diseases
Drug discovery
2018年8月23日星期四
Introduction to Data Mining

15. Steps in KDD Process (Technically)
Knowledge
Data mining
The core step of KDD process
Evaluation/Presentation
Pattern
Data Mining
Relevant Data
Data Preprocessing
Databases
2018年8月23日星期四
Introduction to Data Mining

16. Main Steps of a KDD Process (Fully)
Domain knowledge Acquisition
Learning relevant prior knowledge and goals of application
Data collection and preprocessing (may take 60% of effort!)
Data selection and integration : creating a target data set
Data cleaning, data transformation, and data reduction
Data mining
Choosing functions of data mining
association, classification, clustering, regression, summarization.
Choosing the mining algorithm(s)
Searching for patterns of interest
Pattern evaluation and knowledge presentation
removing redundant patterns, transformation, visualization, etc.
Use of discovered knowledge
2018年8月23日星期四
Introduction to Data Mining

17. Introduction to Data Mining
Mining On What Kind of Data?
Relational databases
Transactional databases
Data warehouses
Advanced D/B and information repositories
Web pages
Temporal data (Time-series data)
Spatial databases
Text databases and multimedia databases
Object-oriented databases
Heterogeneous and legacy databases
2018年8月23日星期四
Introduction to Data Mining

18. Introduction to Data Mining
Steps in KDD Process
Relevant Data
Data Preprocessing
Databases
2018年8月23日星期四
Introduction to Data Mining

19. Why Data Preprocessing?
Data in the real world is dirty (e.g., FaceBook)
incomplete
lacking attribute values, lacking certain attributes of interest, or containing only aggregate data
noisy
containing errors or outliers
inconsistent
containing discrepancies in codes or names
No quality data, no quality mining results!
Quality decisions must be based on quality data
2018年8月23日星期四
Introduction to Data Mining

20. What Major Tasks in Data Preprocessing
Data cleaning
Data integration
Data transformation
Data reduction
Data discretization
2018年8月23日星期四
Introduction to Data Mining

21. Introduction to Data Mining
Steps in KDD Process
Pattern
Data Mining
Relevant Data
Data Preprocessing
Databases
2018年8月23日星期四
Introduction to Data Mining

22. Main Data Mining Techniques
Association Rule Mining
(Descriptive Analysis)
Classification and Prediction
(Predictive Analysis)
Cluster Analysis
(Exploratory Analysis)
Regression Analysis
Outlier Analysis
Trend Analysis
2018年8月23日星期四
Introduction to Data Mining

23. Main Data Mining Techniques
(1/4)
Association Rule Mining
(association rule : correlation and causality)
Form of association rules
buy(T, “Beer”) à buy(T, “Diaper”)
[support = 2%, confidence = 70%] Walmart story
sales(T, “computer”) à sales(T, “software”)
[support = 1%, confidence = 75%] C retail stores
age(X, “21..25”) ^ income(X, “30..39K”) à buys(X, “PC”)
[support = 2%, confidence = 60%] IBM story
age(X, “31..35”) ^ income(X, “40..49K”) à buys(X, “iPad”)
[support = 1%, confidence = 70%] Acer story
2018年8月23日星期四
Introduction to Data Mining

24. Association Rule Mining (Support and Confidence)
transactions
buy both
Given a transaction D/B, find all the rules X  Y with minimum support and confidence
support, S, probability that a transaction contains {X & Y }
confidence, C, conditional probability that a transaction having {X} also contains Y
transactions
buy X
transactions
buy Y
all transactions
Association rules with sup. >= 50%
A  C (50%, 66.6%)
C  A (50%, 100%)
2018年8月23日星期四
Introduction to Data Mining

25. Main Data Mining Techniques Supervised Learning
(2/4)
Use a training set to construct a model for the outcome
forecast of future events. Two main types
Classification
Finding a model that distinguishes classes for future events
e.g., loan approval, customer classification, recognition of finger print
Model representation: decision-tree, artificial neural networks
Prediction
Finding a model that predicts numerical values for future events
e.g., stock price prediction
Model representation: regression, artificial neural networks
2018年8月23日星期四
Introduction to Data Mining

26. Classification vs. Prediction
Use a training set to construct a model for the outcome
forecast of future events
Classification
predicts categorical class labels
constructs a classification model to classify new data
Prediction
predicts numerical values
Constructs a continuous-valued function to predict unknown or missing values
Typical Applications
credit card approval
medical diagnosis & treatment
Pattern recognition
2018年8月23日星期四
Introduction to Data Mining
Introduction to DM & DW

27. Data Mining: Concepts and Techniques
Process of Classification & Prediction
(A Two-Step Process)
Model construction
Training
Data
(I, O)
Learning
Algorithms
Model f
y=f(x)
(xI, yO)
Model usage
Model f
input
features x’
output y’
class label
or value :
2018年8月23日星期四
Data Mining: Concepts and Techniques

28. An Example of Training Dataset (Data of Consumers' Buying Behavior)
Input features (I) : customer characteristics
class label (O)
This follows an example from Quinlan’s ID3
2018年8月23日星期四
Introduction to Data Mining

29. A Decision Tree Model for Predicting buy_PC
Model : buy_PC = f (age, student, credit rating)
<age, student, credit rating> : x
: test (input) attribute
: class label for Buy_PC
: attribute value ? f no
yes
fair
excellent
<= 30
> 40
30..40
student?
age?
credit rating?
buy_PC : y 29

30. A Decision Tree for CAD Screening (Constructed from ~500 Records)
2018年8月23日星期四
2018年8月23日星期四 30
Main Data Mining Techniques for Biomedical Informatics
Introduction to Data Mining

31. Main Data Mining Techniques Cluster analysis
(3/4)
Cluster analysis (unsupervised learning)
Class label is unknown: Group data to form new classes
Application example :
Customer profiling for product recommendation
(Online Bookstores)
Typical clustering principle
Maximizing the intra-class similarity and minimizing the interclass similarity
2018年8月23日星期四
Introduction to Data Mining

32. Introduction to Data Mining
Example of 2D Cluster Analysis X Y Z
3 clusters with points X, Y, and Z as outliers B C A
Difficulty : Data distribution of high dimension is not visually visible.
2018年8月23日星期四
Introduction to Data Mining

33. Introduction to Data Mining
Clustering Example in High Dimension (Gene Expression Analysis by Clustering)
Clustering dendrogram
Finding differentially regulated genes
Clustering
Many people have heard of microarrays, or genechips. Say it’s a way to look at the fingerprint of gene expression for hundreds or thousands of genes in a cell. Note that all proteins are involved in interconnected pathways, so you look for relationships. Groups of genes turned up or down, on or off. Unchanging ones not likely involved directly in the disease cause or course.
Data matrix for visualization
2018年8月23日星期四
Introduction to Data Mining
Introduction to DM & DW

34. Profile of Stroke Patients (Diagnosed by Indices of Chinese Medicine)
Many people have heard of microarrays, or genechips. Say it’s a way to look at the fingerprint of gene expression for hundreds or thousands of genes in a cell. Note that all proteins are involved in interconnected pathways, so you look for relationships. Groups of genes turned up or down, on or off. Unchanging ones not likely involved directly in the disease cause or course.
2018年8月23日星期四
2018年8月23日星期四 34
Main Data Mining Techniques for Biomedical Informatics
Introduction to Data Mining
Introduction to DM & DW
Introduction to DM & DW 34

35. Main Data Mining Techniques
Example of Linear Regression
(4/4) y
y = a x + b
Predict y’s value at X1
using linear regression
y = f (x), what is f ?
explore the meaning
of a and b
Y1 ? x X1
2018年8月23日星期四
2018年8月23日星期四 35
Main Data Mining Techniques for Biomedical Informatics
Introduction to Data Mining

36. Other Data Mining Techniques
(4/4)
Outlier analysis
Outlier: a data object that does not comply with the general behavior of the data
It can be considered as noise or exception but is quite useful in fraud detection, rare events analysis
Trend analysis
Trend and deviation: regression analysis
Sequential pattern mining, periodicity analysis
Other pattern-directed or statistical analyses
2018年8月23日星期四
Introduction to Data Mining

37. Are All the “Discovered” Patterns Interesting?
A data mining system/query may generate thousands of patterns, not all of them are interesting. Pattern screening becomes a problem.
Interestingness : a measure for automatic pattern screening
A pattern is interesting if it is easily understood, potentially useful, novel, valid on new or test data with some degree of certainty, or it validates some hypothesis that a user seeks to confirm
Objective vs. subjective interestingness measures for data screening
Objective: based on statistics and structures of data patterns,
e.g., support, confidence, etc.
Subjective: based on user’s belief in the data,
e.g., unexpectedness, novelty, actionability, etc.
2018年8月23日星期四
Introduction to Data Mining

38. Can We Find All and Only Interesting Patterns?
Completeness vs. Optimization
Completeness : Find all the interesting patterns
Can a data mining system find all the interesting patterns?
Optimization : Only find interesting patterns
Can a data mining system find only the interesting patterns?
Approaches
First generate all the patterns and then filter out the uninteresting ones.
Generate only the interesting patterns—mining query optimization
2018年8月23日星期四
Introduction to Data Mining

39. Classification Scheme of DM Techniques
General functionality
Descriptive/Exploratory data mining
Predictive data mining
Different views, different classifications
Kinds of databases to be mined
Kinds of knowledge to be discovered
Kinds of techniques utilized
Kinds of applications adapted
2018年8月23日星期四
Introduction to Data Mining

40. A Multi-Dimensional View of DM Technique Classification
Databases to be mined
Relational, transactional , Web, object-oriented, object-relational, active, spatial, time-series, text, multi-media, heterogeneous, legacy, etc.
Knowledge to be mined
Association, classification, clustering, trend, characterization, deviation and outlier analysis, etc.
Multiple/integrated functions and mining at multiple levels
Techniques utilized
Database-oriented, data warehouse (OLAP), machine learning, statistics, visualization, etc.
Applications adapted
Retail, telecommunication, banking, fraud analysis, stock market analysis, Web mining, Biomedical informatics, etc.
2018年8月23日星期四
Introduction to Data Mining

41. Summary for Data Mining
Data mining: automatic discovery of interesting knowledge from large amounts of data
A natural evolution of database technology, in great demand, with wide applications
A KDD process includes data pre-processing, data mining, pattern evaluation, and knowledge presentation
Main data mining functions: association, classification, clustering, outlier and trend analysis, characterization, etc.
2018年8月23日星期四
Introduction to Data Mining
Introduction to DM & DW

42. Introduction to Data Mining
Thanks !!!!
Have a Nice Day !
2018年8月23日星期四2018年8月23日星期四
Introduction to Data Mining