MIS 302F Chapter Notes - Chapter 6: Affinity Analysis, Data Mining, Market Basket
17 Mar 2017
School
Department
Course
Professor
Document Summary
: how can we our data to predict patterns/forecast: classification (classification tree): take our understanding of the data to make predictions. Every customer can get a clubcard, which captures demographics, family size, income, race, etc. Every time clubcard is used, the customer is executing certain transactions (recorded: association rule: if i buy bread, i"ll also buy cheese , surprising findings from data mining. Dads who bought diapers would also buy beer because they couldn"t go to pubs with small children. Identified 300 items that a certain socioeconomic group typically bought and undercut the prices of those products. Used purchasing behavior to see who they should target with coupons (classification method: identified who was likely to use the coupon and targeted them) Typical coupon use <1. 5% (meaning tesco was very effective: key concepts: data mining methods, common types of data mining methods data. : find interpretable human patterns that describe the. Association: if antecedent, then some consequent (example:
