Informační systém pro rozpoznávání plagiovaných dokumentů

Abstract

My thesis describes ways of detecting plagiarized or similar documents and displaying such similarity to the user. Main goal was to develop web application that would allow user to search for similar documents and based on various methods display this similarity in a graphical way. Application was developed using Microsoft's .NET platform specifically for the web part of the project I was using web application framework ASP.NET MVC that implements model-view-controller pattern. The theoretical part explains what plagiarism actually is and describes various ways how to approach computer based detection of plagiarized documents. The practical part focuses on description of used algorithms and techniques accompanied by basic software analysis. In the end of my thesis I am conducting benchmarks of the algorithms and from their performance I plot graphs from which I deduct my conclusions.