Unit 5 Flashcards
(14 cards)
what is data wrangling?
the process of transforming and structure raw data into a usable format for analysis
what are the 6 steps of data wrangling?
discover, structure, clean, enrich, verify/validate, publish
what is the 1st step of data wrangling?
discover - explore and understand your data
what is the 2nd step of data wrangling?
structure - organize and format raw data, handle missing values, convert data types, and standardize
what is the 3rd step of data wrangling?
clean - address inconsistencies, errors, and outliers
what is the 4th step of data wrangling?
enrich - provide additional information, context, depth, etc
what is the 5th step of data wrangling?
verify/validate - data integrity, quality, reliability
what is the 6th step in data wrangling?
publish - available and should be documented
what is index overlay?
combining criteria - identifies how much criteria is satisfied at each location
what is weighted linear combination?
combining criteria - simple additive weighting, involves linear tradeoffs
what is weighted product combination?
combining criteria - when criteria are interdependent
what are cobb-douglas functions?
combining criteria - an exponent causes diminishing returns for higher values (think like a curve), larger increase for smaller values
what are weighted exponential combining criteria used?
when extreme values need to be emphasized
what is the decision rule?
a means of selecting particular alternatives from a set of available options