I'm trying to put together a self-directed math curriculum to prepare for learning data mining and machine learning. This is motivated by starting Andrew Ng's machine learning class on Coursera and feeling that before proceeding I needed to improve my math skills. I graduated from college a while ago so my algebra and statistics (specifically from political science/psychology classes) are rusty.
The answers in the thread Is a strong background in maths a total requisite for ML? only suggest books or classes directly related to machine learning; I have already looked into some of those classes and books and do not know exactly what math subject to study (for instance: what field[s] of math address deriving an equation to "minimize a cost function"?). The other thread suggested (Skills & coursework needed to be a data analyst) only mentions broad categories of skills needed for analyzing data. The thread Introduction to statistics for mathematicians does not apply because I do not already have a degree in math; a similar thread Mathematician wants the equivalent knowledge to a quality stats degree has an incredible list of stats books, but again, I'm looking at starting math from a rusty recollection of algebra and moving up from there.
So, for those that work in machine learning and data mining, what fields of math do you find essential to do your job? What math subjects would you suggest to prepare for data mining and machine learning, and in what order? Here is the list and order I have so far:
- Linear algebra
- Statistics (many different sub-fields here, but don't know how to break them out)
As for the data mining and machine learning, through my current job I have access to records on website/app activity, customer/subscription transactions, and real estate data (both static and time-series). I'm hoping to apply the data mining and machine learning to these datasets.