Skip to main content

S.A.V.E

Link to project.

S.A.V.E is a plagiarism detection search engine that respects your privacy.

Simply upload a document, and we can search the document against our database and give you a report with similarity score that represent the overall percentage of similarity. All the document that user uploaded is encrypted so you never have to worry about privacy issue.

SAVE project
A Privacy-aware plagiarism dection search engine. Link to project.

My Contribution #

I was responsible for the optimization of document matching module. I successfully implemented several FM-index across the entirety of the database’s documents. The subsequent search operations yielded remarkable outcomes, with a 100,000x increase in speed and a substantial 60x reduction in memory consumption comparing to the BLAST implementation.

References #

Note #

This project later led to the publication of an IEEE paper and a startup.



Modi 石
Author
Modi 石
Modi is a software developer that talks about machine learning and software development. When he’s not banging on the keyboard, you’ll probably find him playing music or surfing.