lecture6

lecture6 - Data Mining CS57300 Purdue University September...

Info iconThis preview shows pages 1–8. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Data Mining CS57300 Purdue University September 14, 2010 Decision making Heuristics and biases • Tversky & Kahneman, psychologists, propose that people often do not follow rules of probability • Instead, decision making may be based on heuristics • Lowers cognitive load but may lead to systematic errors and biases • Examples: • Availability heuristic • Representativeness heuristic • ConFrmation bias • Conjunction fallacy • Numerosity heuristic Estimating probabilities (Tversky & Kahneman ’73/’74) • Question: Is the letter R more likely to be the 1st or 3rd letter in English words? • Results: Most said R more probable as 1st letter • Reality: R appears much more often as the 3rd letter, but easier to think of words where R is the 1st letter Estimating probabilities (cont) • Question: Which causes more deaths in developed countries? (a) trafFc accidents or (b) stomach cancer • Typical guess: trafFc accident = 4X stomach cancer • Actual: 45,000 trafFc, 95,000 stomach cancer deaths in US • Ratio of newspaper reports on each subject: 137 (trafFc fatality) to 1 (stomach cancer death) • Availability heuristic : Tendency for people to make judgments of frequency on basis of how easily examples come to mind Base Rate Study (Kahneman & Tversky '73) • Participants told that for a set of 100 people are either: • 30% engineers/70% lawyers, or • 70% engineers/30% lawyers • Given: A description of a person Jack, which is representative of a prototypical engineer (e.g., likes carpentry and mathematical puzzles, careful, conservative) • Question: Is Jack more likely to be a lawyer or engineer? • Results: Participants in the 30% condition judged Jack just as likely to be an engineer as participants in the 70% condition. Base rate study (cont) • People use the representative heuristic to make inferences... • Inferences is based solely on similarity of target to category members • Base rates (70%-30%) are ignored • ...rather than using formal statistical rules to make inferences • Inferences should be based on similarity of target to category members AND base rates (70%-30%)...
View Full Document

This note was uploaded on 03/13/2012 for the course CS 573 taught by Professor Staff during the Fall '08 term at Purdue.

Page1 / 32

lecture6 - Data Mining CS57300 Purdue University September...

This preview shows document pages 1 - 8. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online