Plagiarism Detection Tool for Sinhala Language with Internet Resources Using Natural Language Processing

With the digitalization of text through the World Wide Web, plagiarism turned into a crucial problem and a way to detect plagiarism became an essential component. Even though there are many plagiarism detection systems, applying in the world by considering other languages, for Sinhala it’s only a few, and it senses a vacuum in the domain. With the significant improvement of availability in Sinhala text on WWW, still, there is no system to detect plagiarism for these Sinhala documents by comparing resources on the internet. The purpose of this research is to address and overcome the above-mentioned gap while introducing a plagiarism detection system for the Sinhala language, by using internet resources. In the process to obtain the outcome, text pre-processing, Google searching and similarity comparison using Jaccard coefficient were the steps followed. As the final product, a Sinhala plagiarism detection tool was developed with 88% accuracy. The outcome of the work will be a support for the lectures, teachers, authors, students who are using Sinhala as their literal language.