Concept Development for Challenging Machine Learning Project (text based data set)

We need a ML specialist to support us in the Concept Development (not software coding) of a ML project. the challenging part of this project lies in the data set preparation, which comes in from of various text files (each data set has an own text file). Labels are given, but the data must be searched in the text files by some algorithm according to certain criteria (there exist patterns in the text). Your result will be a clearlz defined technology solution, a specification and recommendation