To devise algorithm to clean up txt documents that have been converted from PDF

i need to write script(s) that will clean up a number of text files that have been converted from PDF to text docs and put them into summary using Python or any programming language and then finally applying a textrank(a script(s) that will be developed by me) that will do the final summary of the documents,compare and evaluate this algorithms with the textrank to determine which is the best summarizer. The purpose of this system is to enable a user when searching for academic articles online to read through a brief summary of any article before choosing which paper to use.
The script(s) that will be written is required to automatically carry out this cleaning up and summarization.