From the course: Data Preparation, Feature Engineering, and Augmentation for AI Models
Join today to access over 24,700 courses taught by industry experts.
Building models
From the course: Data Preparation, Feature Engineering, and Augmentation for AI Models
Building models
- [Instructor] AI Model Development spans both predictive and generative AI approaches. For example, with Predictive models, we're dealing with algorithms and models that do things like classify images, or calculate regressions or cluster data, whereas with Generative models, we're actually generating things like text, image or code. Now, each of these approaches requires specific development practices, so let's take a look at first, Predictive models. Now, we'll look at Classification first. Classification models categorize data into predefined classes or labels, so some retail examples where we may use classification models include things like fraud detection and customer segmentation. Now, some key considerations here as we're working with the data are, are there class imbalances within our training data? For example, in the case of fraud detection, oftentimes, we have to deal with the fact that there are relatively few numbers of fraudulent transactions and many legitimate…
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.