Data Mining!An Introduction
View Slide
Wufoo.com
What is data mining?
Collection? No!Extraction. Yup.
324 - 576 megapixelsStereo Audio 20-20,000hz10,000 Chemical Compounds5-6 FlavorsTemperature / Pressure / Texture2.5 PetabytesEyesEarsNoseMouthSkinMemory
The process of extracting patternsfrom large data sets.
What are some examples oflarge data sets?
AstronomyBiologyBusinessInternetGovernmentReligion
Online Surveys
Individuals, Developers, Designers,Non-Profits, Teachers, Students,Universities, Research, Real Estate,Marketing, Healthcare, Banks, SMBs
What do they do with all that data?
Positive / NegativeLikert ScaleRatingsMultiple ChoiceOpen Feedback
What are some potential problemswith data collected by asking?
Data collection is just the first part.
Association Rule LearningClusteringClassificationRegressionVisualization
StatisticsArtificial IntelligenceDatabase Management
Bayes Theorem (1700s)Regression Analysis (1800s)Neural Networks (1940s)Genetic Algorithms (1950s)Decision Tree Learning (1960s)Support Vector Machines (1990s)
Google Flu Trends
Hans Rosling
Recommendation Engines
Relationships!
Will my date have sex on the first date?Do you like the taste of beer?
Assuming you were in the position todo so, would you launch nuclearweapons under any circumstances?82%
In a certain light, wouldn'tnuclear war be exciting?83%
The Social Graph
Privacy & Confidentiality Issues
So that’s data mining!
Thanks!