OBS: There is a version in portuguese.
Handwritten digit recognition task was one of first great successes of machine learning methods. Nowadays, the task can be carried out by multiple specialized libraries with very high accuracy (> 97% of correct answers), such that many times, despite of indirectly we use these features in tablets and smartphones, in general we do not know exactly how the method works.
Thinking about it, as I worked with this problem before, I will demonstrate in this post how the process works, the techniques used and how to implement it with R language. To begin, we will work with the problem of recognizing digits 0,1,2 , 3,4,5,6,7,8, or 9, i.e. a classification problem with 10 categories.
I’ll try to work here implementing all the modeling only with R base functions and a few extra packages with the required functions and algorithms; in the next post, I can try to use other packages to automate the various modeling tasks.