While investigating plagiarism is relevant in different fields, verification of original authorship has also attracted attention in academia and L2 learning and assessment contexts. Generally associated with academic misconduct and dishonesty, plagiarism in writing can take many shapes and be hard to detect. In addition to being a very time-consuming task, detecting plagiarism is not straightforward because committing plagiarism usually involves using techniques other than word-for-word copying and pasting the exact words, sentences, or other pieces of writing. That is why computer scientists are looking for more efficient and fast automatic detection tools. This study aims at developing an external plagiarism detection model, analyzing 500 suspicious documents from a corpus of artificially plagiarized documents while finding similarity levels between them and a corresponding 500 source documents.