PENGENALAN

DATA MINING
PTIK
Week 10

1
 Pokok Bahasan

Latar Belakang Data Mining

Apa dan Mengapa Data Mining
Task dalam Data mining
Fungsionalitas Data mining
Hubungan antara sistem data mining dengan
Sistem Basis Data, Sistem Data Warehouse,
dan Business Intelligence
Permasalahan dalam Data Mining

12/01/23
 Latar Belakang Data Mining (1)

Melimpahnya Data
– Terciptanya data dari tools otomatis dan teknologi basis data
sehingga jumlah yang tercatat dalam basis data atau media
penyimpanan lain semakin membesar

12/01/23
 Latar Belakang Data Mining (2)

Walaupun data teramat melimpah, namun yang diolah menjadi

knowledge sangat sedikit

Solusinya??  Data warehouse dan data mining

– Data warehouse dan OLAP (on-line analytical processing)
– Ekstraksi knowledge yang menarik dalam bentuk rule, regularities,
pola, konstrain dll dari data yang tersimpan dalam sejumlah besar
basis data

12/01/23
 Top 10 Database Terbesar 2012
No Badan/Organisasi Jumlah Data
1 World Data Centre for Climate • 20 terabytes of web data
• 6 petabytes of additional
data

2 National Energy Research • 2.8 petabytes of data

Scientific Computing Center • Operated by 2,000
computational scientists

3 AT&T • 23 terabytes of
information
• 1.9 trillion phone call
records
4 Google •1 million searches per day

Sumber: http://www.siliconindia.com/news/enterpriseit/Top-10-Largest-Databases-in-the-
World-nid-118891-cid-7.html
12/01/23
 Perkembangan Data di Dunia (1)

Source : Tan, 2004

12/01/23
 Perkembangan Data di Dunia (2)

The amount of data stored in various media has

doubled in three years, from 1999 to 2002. the
amount of data put into storage in 2002, five
exabytes (one quintillion bytes), was equal to the
contents pf ahalf a million new libraries, each
containing a digitised version of the print collection
of the entire US Library of Congress
(Lyman and varian, UC Berkeley, 2003)

12/01/23
 Perkembangan Data di Dunia (3)

" It is projected that just four years from now, the world’s
information base will be doubling in size every 11 hours. So rapid is
the growth in the global stock of digital data that the very
vocabulary used to indicate quantities has had to expand to keep
pace. A decade or two ago, professional computer users and
managers worked in kilobytes and megabytes. Now school children
have access to laptops with tens of gigabytes of storage, and
network managers have to think in terms of the terabyte (1,000
gigabytes) and the petabyte (1,000 terabytes). Beyond those lie the
exabyte, zettabyte and yottabyte, each a thousand times bigger
than the last.
(IBM Global Technical Services white paper published in July 2006, titled, "The toxic terabyte: How data-dumping threatens
business efficiency.)

12/01/23
 Pokok Bahasan

Latar Belakang Data Mining

Apa dan Mengapa Data Mining
Hubungan sistem data mining dengan Sistem Basis
Data, Sistem Data Warehouse , dan Business
Intelligence
Task dalam Data mining
Fungsionalitas Data mining
Permasalahan dalam Data Mining

12/01/23
12/01/23
 Just Joke..

12/01/23
 Definisi Data Mining

Data mining is an iterative process within which progress is

defined by discovery, through either automatic or manual
methods. [Kantardzic , 2003]
Data mining (DM) is the extraction of hidden predictive
information from large databases (DBs). With the automatic
discovery of knowledge implicit within DBs, DM uses
sophisticated statistical analysis and modeling techniques to
uncover patterns and relationships hidden in organizational
DBs [Wang, 2003]
Data mining refers to extracting or \mining" knowledge from
large amounts of data [Han, 2005]
Non-trivial extraction of implicit, previously unknown and
potentially useful information from data [Tan, 2003]

12/01/23
 Awal Data Mining

• Berawal dari beberapa

disiplin ilmu, bertujuan
untuk memperbaiki teknik
tradisional sehingga bisa
menangani:
– Jumlah data yang sangat
besar
– Dimensi data yang tinggi
– Data yang heterogen dan
berbeda bersifat

12/01/23
 Jenis Data pada Data Mining

database, data warehouse, database transaksional

Data streams dan sensor data
Time-series data, temporal data, sequence data
Struktur data, graf, social networks dan database link
Object-relational database
Spatial data
spatiotemporal data
Multimedia database
Text databases
The World-Wide Web

12/01/23
 Pokok Bahasan

Latar Belakang Data Mining

Apa dan Mengapa Data Mining
Hubungan sistem data mining dengan Sistem
Basis Data, Sistem Data Warehouse , dan
Business Intelligence
Fungsionalitas Data mining
Task dalam Data mining
Permasalahan dalam Data Mining

12/01/23
 Hubungan DM, DB dan DW

Untuk mengoptimalkan penggunaannya sistem Data Mining

seharusnya memiliki hubungan dengan sistem basis data dan data
warehouse.
Tidak adanya hubungan tidak direkomendasikan misalnya seperti flat
file processing
Hubungan Loose coupling misalkan mpengambilan data dari DB/DW
Hubungan Semi-tight coupling, yakni utnuk menambah performansi
DM dengan pengimplementasian primitif data mining dalam sistem
DB/DW misalkan sorting, indexing, aggregation, histogram analysis,
multiway join dll
Hubungan Tight coupling— merupakan enviroment pemrosesan yang
sama dimana DM terintegrasi dengan sistem DB/DW, mining query
dioptimasi berdasrkan mining query, indexing, metode pemrosesan
query processing methods, dll.

12/01/23
 Data Mining &
Business Intelligence

End User
Meningkatkan potensi untuk Making
mendukung keputusan bisnis Decisions

Data Presentation Business

Analyst
Visualization Techniques
Data Mining Data
Information Discovery Analyst

Data
Statistical Analysis, Querying and Reporting
Exploration
Data Warehouses / Data Marts
OLAP, MDA DBA
Data Sources
Paper, Files, Information Providers, Database Systems, OLTP

12/01/23
 Pokok Bahasan

Latar Belakang Data Mining

Apa dan Mengapa Data Mining
Integrasi sistem data mining dengan Sistem
Basis Data,Sistem Data Warehouse , dan
Business Intelligence
Task dalam Data mining
Fungsionalitas Data mining
Permasalahan dalam Data Mining

12/01/23
 Task dalam Data Mining

Metode Prediksi
– Dengan menggunakan beberapa variabel untuk memprediksi nilai
yang belum diketahui (unknown ) atau nilai selanjutnya (future)
dari variabel lain
Contoh:
Classification
Regression
Deviation Detection
Metode Deskripsi
– Menemukan pola pendeskripsian data yang dapat
diinterpretasikan oleh manusia
Contoh:
Clustering
Association Rule Discovery
Sequential Pattern Discovery

12/01/23
 Pokok Bahasan

Latar Belakang Data Mining

Apa dan Mengapa Data Mining
Integrasi sistem data mining dengan Sistem
Basis Data,Sistem Data Warehouse , dan
Business Intelligence
Task dalam Data mining
Fungsionalitas Data mining
Permasalahan dalam Data Mining

12/01/23
 Fungsionalitas Data Mining (1)

Klasifikasi dan Prediksi

Frequent patterns, asosiasi , korelasi dan kausalitas
Analisis klaster
Analisis Outlier
Analysis Trend dan evolution
Analisis statistik

12/01/23
 Aplikasi Data Mining (1)

 Analisis dan Manajemen Pasar

▪ target pemasaran, customer relation management (CRM),
market basket analysis, cross selling, segmentasi pasar
 Analisis dan Manajemen Resiko
▪ Forecasting, customer retention, quality control, analisis
kompetisi
 Deteksi dan manajemen fraud (kecurangan)
 Text mining (news group, email, dokumen)
dan Analisis Web.
 Intelligent query answering
12/01/23
 Aplikasi Data Mining (2)

Marketing and Sales Promotion

Supermarket shelf management.
Inventory Management
Diagnosis Medis
Collaborative Filtering
Business Intelligence
Network Intrusion detection
Deteksi spam
dll

12/01/23
 Pokok Bahasan

Latar Belakang Data Mining

Apa dan Mengapa Data Mining
Integrasi sistem data mining dengan Sistem
Basis Data,Sistem Data Warehouse , dan
Business Intelligence
Task dalam Data mining
Fungsionalitas Data mining
Permasalahan dalam Data Mining

12/01/23
 Permasalahan Utama

Bagaimana Menentukan metodologi mining? karena:

Tipe data berbeda
Performansi yang diharapkan dari segi keefektifan, efisiensi dan skalabilitas bisa
jadi berbeda tiap metodologi
Evaluasi pola yanki pengukuran “interestingness’ yang berbeda
Penanganan missing value dan noise
dll

Bagaimana Bentuk Interaksi dengan User? Apakah:

– Menggunakan Data mining query languages dan ad-hoc mining
– Hasil data mining berupa ekspresi dan visualisasi

Aplikasi dan Dampak Sosial

– Perlindungan terhadap keamanan , integrity dan privacy data

12/01/23