Statistics > QUESTIONS & ANSWERS > Georgia Institute Of Technology - ISYE 6501Notes_Quiz1. (All)

Georgia Institute Of Technology - ISYE 6501Notes_Quiz1.

Document Content and Description Below

Lesson 1.3 - Models 1. 3 different ways to use model in analytics a. Regression i. e.g. regression to plan delivery dates = model b. Regression based on size, weight, distance i. e.g. use of regr... ession and s.w.d. = model c. Regression estimate = 37 + 81x size + 76xweight +4xdistance = model (the specific equation. Lesson 2.1 - Intro to classification 1. Classification = Putting things into categories a. e.g. yes/no category (those who pay loans and those who don't) b. Use classification models to put things into categories c. Can have more than two categories i. e.g. supportive voters, vs not supportive, vs undecided 2. Requires data to get answers a. e.g. income, credit history, age, etc. i. These are attributes, and based on these attributes for previous examples and how the outcome we can build a model b. More factors can be more dimensions Lesson 2.2 - Choosing a classifier 1. Separation lines - Separators or classifiers 2. Which line to choose? a. In general, want to choose line that's farther from making mistakes b. If classifier is too close to data point, it's very close to being misclassified i. This is important because data might be inexact. ii. Don't want small uncertainties in the data to cause classification errors c. Want to choose line that would take a much error in data to cause misclassification. [Show More]

Last updated: 2 years ago

Preview 1 out of 28 pages

Buy Now

Instant download

We Accept: