I am building a ML application to predict supervisory post grade in a group of organization .The supervisory post have subordinates with lower grade( G1 to G10). The goal to input the subordinates grade and job sears(occ) and then the output should predict supervisory(Head) grade and level.
example of a group:
OCC TITLE GRADE Level
0510 Head cost control E 2 <-----predict this
0510 Sr. Accountant P 6 <------input
0510 Accountant P 5 <-----input
0503 Financial specialist G 9 <-----input
OCC is a category not numerical , so what is the best way to form the matrix to prepare in the datasets to have a normalized data?