Federated document search using hybrid peer-to-peer network for plagiarism detection

Detecting plagiarism between two papers located on different institutions cannot be done as long as there are institutions refusing to open their collection for public fearing them being plagiarized. This research developed a prototype of a cross-institutional federated document search using API-based hybrid peer-to-peer network. This approach enables searching for documents without institutions opening their collections for public. Secure link and PDF watermarking are implemented as access control methods to prevent document misuse. Emulation of text similarity detection are done using Elasticsearch. Based on the test result, we can conclude that the prototype fulfills the required business process.