SDS 475: The 20% of Analytics Driving 80% of ROI — with David Langer

Transition

David Langer discusses the circumstances under which businesses might transition from traditional statistical methods to machine learning. He emphasizes the importance of the business questions being asked, noting that machine learning is particularly useful for classification tasks. For example, it can help determine whether to approve a loan or identify patterns associated with customer conversion 1.

Machine learning allows you to answer questions primarily. It's most often used, in my experience, in the classification space.

--- David Langer

Jon Krohn inquires about the specific scenarios that necessitate this shift, prompting David to elaborate on the practical applications of machine learning in business contexts 2.

   

Techniques

David explains that practical machine learning techniques like logistic regression and decision trees are accessible and effective for business professionals. He highlights the complexity of interpreting logistic regression models, which often requires understanding odds ratios and logarithms. Decision trees and random forests, on the other hand, are simpler to understand and communicate 3.

If you're using something like a decision tree or a decision tree based algorithm like the mighty random forest, not only is it easy for any professional to learn how they work, but the results are also very easy to interpret and to communicate to other people.

--- David Langer

These techniques are particularly useful for answering binary questions or predicting categorical labels, making them valuable tools in the business analytics toolkit.