Your SlideShare is downloading. ×
0
isd314-01
isd314-01
isd314-01
isd314-01
isd314-01
isd314-01
isd314-01
isd314-01
isd314-01
isd314-01
isd314-01
isd314-01
isd314-01
isd314-01
isd314-01
isd314-01
isd314-01
isd314-01
isd314-01
isd314-01
isd314-01
isd314-01
isd314-01
isd314-01
Upcoming SlideShare
Loading in...5
×

Thanks for flagging this SlideShare!

Oops! An error has occurred.

×
Saving this for later? Get the SlideShare app to save on your phone or tablet. Read anywhere, anytime – even offline.
Text the download link to your phone
Standard text messaging rates apply

isd314-01

530

Published on

Published in: Education, Technology
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total Views
530
On Slideshare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
25
Comments
0
Likes
0
Embeds 0
No embeds

Report content
Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
No notes for slide

Transcript

  • 1. Teknik Data Mining Pendahuluan Anung B. Ariwibowo Teknik Data Mining - ISD314
  • 2. Administratif Perkuliahan 12 Perkuliahan dalam 12 pekan  Senin, 10.00 – 12.30 WIB Kehadiran  Paling lambat 15 menit (Mahasiswa, Dosen)  70% minimum (Mahasiswa) 2 Anung Ariwibowo 2011-2012-3
  • 3. Administratif Perkuliahan Komponen Penilaian  10% Kuis (9 + Pre-test)  20% Tugas (4)  30% UTS (Open book)  40% UAS (Open book)  5% Bonus (Aktifitas, pengumpulan dini, kreatifitas) 3 Anung Ariwibowo 2011-2012-3
  • 4. Administratif Perkuliahan WEKA  http://www.cs.waikato.ac.nz/ml/weka/  http://weka.wikispaces.com/Primer Toby Segaran, "Programming Collective Intelligence: Building Smart Web 2.0 Applications", OReilly, 1st Edition, 2007. Ian H. Witten, Eibe Frank, dan Mark A. Hall, "Data Mining: Practical Machine Learning Tools and Techniques", 3rd Edition, Morgan Kauffman, 2011. Slide buku Ian Witten, http://www.cs.waikato.ac.nz/ml/weka/book.html 4 Anung Ariwibowo 2011-2012-3
  • 5. Administratif Perkuliahan Jiawei Han, Micheline Kamber, dan Jian Pei, "Data Mining: Concepts and Techniques", 3rd Edition, Morgan Kauffman, 2011. Slide buku Jiawei Han, http://www.cs.uiuc.edu/~hanj/dmbook KDNuggets, http://www.kdnuggets.com/ Slide dari Andrew Moore, http://www.autonlab.org/tutorials/list.html Slides tjerdastangkas.blogspot.com/search/label/isd314 Slide ke-0.5 dari Anung B. Ariwibowo Teknik Data Mining - NumPages ISD314
  • 6. Disclaimer Following slides are derived from Jiawei Hans slideSlide ke-0.6 Anung B. Ariwibowo Teknik Data Mining -dari ISD314
  • 7. Chapter 1. Introduction 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 7 Data Mining: Concepts and Techniques March 12, 2012
  • 8. 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! Solution: Data warehousing and data mining  Data warehousing and on-line analytical processing  Miing interesting knowledge (rules, regularities, patterns, constraints) from data in large databases 8 Data Mining: Concepts and Techniques March 12, 2012
  • 9. 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.) 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 9 Data Mining: Concepts and Techniques March 12, 2012
  • 10. Trends leading to Data Flood More data is generated:  Bank, telecom, other business transactions ...  Scientific Data: astronomy, biology, etc  Web, text, and e-commerce 10
  • 11. Data Growth Large DB examples as of 2003:  France Telecom has largest decision-support DB, ~30TB; AT&T ~ 26 TB  Alexa internet archive: 7 years of data, 500 TB  Google searches 3.3 Billion pages, ? TB Twice as much information was created in 2002 as in 1999 (~30% growth rate) Knowledge Discovery is NEEDED to make sense and use of data. 11
  • 12. What Is Data Mining? Data mining (knowledge discovery from data)  Extraction of interesting (non-trivial, implicit, previously unknown and potentially useful) patterns or knowledge from huge amount of data  Data mining: a misnomer? Alternative names  Knowledge discovery (mining) in databases (KDD), knowledge extraction, data/pattern analysis, data archeology, data dredging, information harvesting, business intelligence, etc. Watch out: Is everything “data mining”?  (Deductive) query processing.  Expert systems or small ML/statistical programs 12 Data Mining: Concepts and Techniques March 12, 2012
  • 13. 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) Other Applications  Text mining (news group, email, documents) and Web mining  Stream data mining  DNA and bio-data analysis 13 Data Mining: Concepts and Techniques March 12, 2012
  • 14. 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, etc.  Determine customer purchasing patterns over time Cross-market analysis  Associations/co-relations between product sales, & prediction based on such association 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) 14 Data Mining: Concepts and Techniques March 12, 2012
  • 15. 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 15 Data Mining: Concepts and Techniques March 12, 2012
  • 16. 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  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 16 Data Mining: Concepts and Techniques March 12, 2012
  • 17. Other Applications Sports  IBM Advanced Scout analyzed NBA game statistics (shots blocked, assists, and fouls) to gain competitive advantage for New York Knicks and Miami Heat Astronomy  JPL and the Palomar Observatory discovered 22 quasars with the help of data mining Internet Web Surf-Aid  IBM Surf-Aid applies data mining algorithms to Web access logs for market-related pages to discover customer preference and behavior pages, analyzing effectiveness of Web marketing, improving Web site organization, etc. 17 Data Mining: Concepts and Techniques March 12, 2012
  • 18. Data Mining: A KDD Process  Data mining—core of Pattern Evaluation knowledge discovery process Data Mining Task-relevant Data Data Warehouse SelectionData Cleaning Data Integration 18 Databases Data Mining: Concepts and Techniques March 12, 2012
  • 19. 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. 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 19 Data Mining: Concepts and Techniques March 12, 2012
  • 20. Data Mining and Business IntelligenceIncreasing potentialto supportbusiness 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, Information Providers, Database Systems, OLTP20 Data Mining: Concepts and Techniques March 12, 2012
  • 21. Data Mining: Confluence of MultipleDisciplines Database Statistics Systems Machine Data Mining Visualization Learning Algorithm Other Disciplines 21 Data Mining: Concepts and Techniques March 12, 2012
  • 22. Weather Data: Play or not Play? Outlook Temperature Humidity Windy Play? sunny hot high false No Note: sunny hot high true No Outlook is the overcast hot high false Yes Forecast, rain mild high false Yes no relation to rain cool normal false Yes Microsoft rain cool normal true No email program overcast cool normal true Yes sunny mild high false No sunny cool normal false Yes rain mild normal false Yes sunny mild normal true Yes overcast mild high true Yes overcast hot normal false Yes rain mild high true No 22
  • 23. Example Tree for “Play?” For Outlook=sunny, Humidity=normal, Outlook Windy=true Play = ? sunny rain overcast Humidity Yes Windy high normal true false No Yes No Yes 23
  • 24. Pustaka Slides tjerdastangkas.blogspot.com/search/label/isd314 Slide ke-0.24 dari Anung B. Ariwibowo Teknik Data Mining - NumPages ISD314

×