One System to Solve Them All

People are daily confronted with hundreds of situations in which they could use the knowledge of stylometry. In this paper, I propose a universal system to solve these situations using stylometry features, machine learning techniques and nature language processing tools. The proposed tool can help translation companies to recognize machine translation falsely submitted as a work of a human expert; identify school essays not written by the underwritten student; or cluster product reviews by authors and merge user reviews written by one author using multiple accounts. All examples above use same techniques and procedures to solve the problem, therefore it is preferred to merge algorithms and implementation of these tasks to a single framework.