First we read the image or pdf file and extract all the words from the file and stored in variables. This extracted words are matched with each words from dictionary file.

We have mentioned some important keywords in dictionary file i.e Words from PAN Card and Adhar Card 

if we want to add other file then we have to mention the keywords from that file into the dictionary file.

Initially we had only taken PAN Card and Adhar Card for the file classification but In the future we will include all the necessary files based on the requirement.