Legal · ScamRadar+
TERMS OF USE
By using ScamRadar+, you agree to these terms. This is an academic research tool — read these before relying on any result.
Important: ScamRadar+ is a machine-learning tool designed to assist your judgement — not a guarantee. On its 25,306-message external benchmark the deployed E8-P9 pipeline achieves F1 = 0.913, but individual predictions can still be wrong. Never use it as your sole basis for financial or safety decisions. When in doubt, verify through official channels.
Use at Your Own Risk
- —ScamRadar+ is provided as-is, without any warranty of accuracy, completeness, or fitness for a particular purpose.
- —On a locked one-shot benchmark of 25,306 messages, the deployed E8-P9 pipeline (classifier + Rule Engine) achieves F1 = 0.913, meaning it still misclassifies roughly 3% of messages. Do not rely solely on this tool to make financial, legal, or safety decisions.
- —Always verify suspicious messages directly through official channels (not contact details provided in the message itself).
Not Legal or Financial Advice
- —Nothing produced by ScamRadar+ constitutes legal advice, financial advice, or professional security consultation.
- —Verdicts of SCAM or LEGIT are probabilistic model outputs, not legal determinations.
- —If you have been the victim of fraud, contact your local law enforcement and your bank directly.
Limitation of Liability
- —The creators of ScamRadar+ accept no liability for any loss, damage, or harm resulting from reliance on a scam verdict — whether the result was a false positive or a missed detection.
- —We are not responsible for any third-party services (VirusTotal, Google Safe Browsing) or their availability, accuracy, or uptime.
- —This project is a BSc final-year thesis (Emek Yezreel College, 2026) and carries no commercial support or service-level guarantees.
Acceptable Use
- —You may not use ScamRadar+ to submit content for purposes other than personal scam-detection (e.g. automated bulk scanning, stress-testing, or adversarial probing).
- —You may not attempt to reverse-engineer, circumvent rate limits, or extract training data from the model.
- —Submitting content that is illegal, offensive, or violates third-party rights is prohibited.
No Affiliation
- —ScamRadar+ is not affiliated with, endorsed by, or associated with any brand, company, or organisation mentioned in example messages or threat categories — including but not limited to Apple, Google, PayPal, or any financial institution.
- —All brand names referenced are trademarks of their respective owners and are used solely for illustrative purposes.
Contact & Changes
- —For any questions regarding these terms, contact: amerrhassouna@gmail.com
- —We reserve the right to update these terms at any time. Continued use of the service constitutes acceptance of any revised terms.
- —These terms were last updated: May 2026.
ScamRadar+ · BSc Final Year Project 2026 · Emek Yezreel College