Mathematizing a problem

Seems like from age old, from Galileo to Current date — one of things which has been constant is Mathematization the problem

Below are some problems and their representation in Mathematical forms

Examples for above have been mentioned in Book - Calculus of every day life

In the book, based on data author suggests/proposes which model fits which context better .

I used to wonder how is it possible to remember so many equations ? How to represent problem efficiently ? How to model the problem ? How to remember patterns - One solution is to have Giant Look Up table with process and equations/models associated with it

When I was exposed to Software Design patterns/ Gang Of Four, same kind of doubts came up. How to recognise and understand the patterns and fit it to problem at hand ?

Enter Machine Learning - Seems like exactly what it does ? It tries to mathematically come up with Model based on your data .

By providing ML Algorithm/technique with data you can either apply Regression or Classification or numerous other techniques and it comes up with equation/model which is indicative of underlying pattern