From the course: LinkedIn AI Academy AI-100: 1 Demystifying AI
Join today to access over 23,200 courses taught by industry experts.
Why are linear models not enough?
From the course: LinkedIn AI Academy AI-100: 1 Demystifying AI
Why are linear models not enough?
- [Instructor] So far, we have been working with linear models. For regression, we saw how to build a linear relationship between outcome variables and the features. And for classification, we saw how to find a straight line to separate the space into classes. We also discussed in regression that the relationship between the outcome variables and the features might not be linear. And we saw an example of models that are quadratic in the features but still linear in the parameters. This is one way in which practitioners try to model non-linear behavior but still using linear regression. And this also works in the classification problems. This is seen as the more traditional way to build models. You either identify what works for the data yourself or you might enlist some domain experts who know the specific transformations that might be important. Going back to our previous example, if you look at an article length,…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.