![](/assets/img/products/rmse/rmse.jpg)
Scam sites identified by rMSE, including those reported through ScamShield by citizens, are forwarded to the Singapore Police Force and/or Google. Upon agreement, public access to these sites is disrupted, often through a red Google Safe Browsing warning displayed on supported browsers. Additionally, SPF may instruct internet service providers to block access to these sites within Singapore. SPF also continuously provides operational insights to rMSE to ensure it stays updated with evolving scams.
![Enhanced rMSE model](/assets/img/products/rmse/rmse-model.png)
During the initial stages of developing rMSE in Q1 2024, we evaluated a model crafted by external researchers using local scam data from SPF. We observed that the model, although sophisticated and carefully designed, exhibited low performance with approximately 70% precision and 10% recall. This outcome was expected because the model was primarily trained on globally known scam sites, many of which were already inactive or outdated, along with benign examples that were globally representative.
However, after fine-tuning the underlying model with (part of) the local scam data and further refining it based on insights into scam site operations, we achieved significantly improved results, attaining over 90% precision and over 90% recall.
Thanks for letting us know that this page is useful for you!
If you've got a moment, please tell us what we did right so that we can do more of it.
Did this page help you? - No
Thanks for letting us know that this page still needs work to be done.
If you've got a moment, please tell us how we can make this page better.
rMSE
A Classifier That Guards Citizens Against Websites Scams