TextToCSVConvertor

I am facing problem with the data, the problem is adobe extracted different api structure for some pdf, that's why I am getting wrong data in 5 to 6 pdf files, not all the data is wrong only some columns like email, username, because sometime adobe only read .co .co m .c . c om, different structure like this should be hotmail.com not hotma and il.com under that, it affected the whole structure of the pdf, I recognised that the wrong data happen because of different api structure,t I tried in several ways but that hard to automate, there is no pattern to automate
Can you suggest some ways how to do that ?