Page 28 - 2024

XGBOOST, which is suitable for data streams, with PPXGBOOST for privacy preserva- tion.
Adaptive XGBOOST allows continuous learning from evolving data streams, which is
advantageous for scenarios like fraud detection where quick response times are crucial.
PPXGBOOST ensures data privacy while mining, which is vital in sensitive applications.
This integrated approach aims to strike a balance between accurate results and data
privacy, making it effective in handling data stream mining challenges.

Privacy Preserving Random Forest (PPRF) is also proposed in this paper and has
achieved remarkable results, with an accuracy rate of 92%. In an era marked by
heightened concerns about data privacy and security, PPRF addresses a critical need. By
harnessing the capabilities of Privacy-Preserving Random Forests, this approach ensures
that sensitive information remains shielded from prying eyes, even in the midst of
rigorous data analysis. PPRF accomplishes this feat through the use of cutting-edge
cryptographic methods, such as homomorphic encryption and secure multiparty
computation.

The profound significance of PPRF extends far beyond its impressive accuracy rate. It
opens new horizons for secure and efficient data analysis, particularly in domains where
data privacy is paramount. Industries such as healthcare, financial analysis, and
government sectors can now harness the power of data stream mining with the
confidence that sensitive information remains confidential.
Key words: Data mining, Data stream mining, Privacy preservation, Adaptive XG- BOOST,
Streaming data, Data privacy, Machine learning, EPPXGBOOST, PPRF

02