Data mining

1
Data mining

• Is an emerging field that consolidates a series
  of analytic technologies and links these to
  evolving computer technologies to extract meaning
  and relationships out of large databases

2
Data mining (II)

• while these databases are large they are not to
  be compared with the streams of data from the
  physical sciences
• examples
• customer purchases
• supreme court decisions
• patterns in health care expenses

3
Miner's toolkit

• query reporting tools
• multidimensional analysis tools
• intelligent agents

4
Miner's dynamite

• clean organize the data
• data warehousing
• distributed storage
• parallel computing
• axiomatic searches

5
Did you know that

• diapers and beer correlate?
• inbound calls can identify someone's carrier
• that social networks are the major determinant of
  telephone calling
• buying patterns vary over the year and correlate
  with events

6
Did you know that ..

• You can predict reasonable well what people will
  buy
• and maybe
• SURFING PATH IS A PREDICTOR OF PURCHASES
• TELEPHONE AND NET SERVICES TOGETHER CAN GENERATE
  ADDITIONAL BUNDLING OPPORTUNITIES

7
Profiles

• pay cash for a one-way plane trip to Syria, and a
  passenger-screening computer may tag you as a
  possible terrorist whole luggage must be
  thoroughly searched for a bomb

8
Profiles (II)

• visit a psychiatrist and an employment agency in
  the same month, and a credit scoring computer may
  mark you as someone heading for bankruptcy

9
Data mining

• Miklos A. Vasarhelyi