DATA ENVIRONMENT TO DETECT COLLABORATIVE SPAM WHILE MAINTAINING PRIVACY

ABSTRACT

In today's digital age, it is essential to create a safe big data environment to detect collaborative spam while retaining privacy. Spam emails present serious issues to both individuals and organizations, necessitating strong countermeasures. By utilizing a safe big data environment and utilizing machine learning methods for spam detection, this project seeks to overcome these issues. By using anonymization techniques to safeguard users' personally identifiable information (PII), the system ensures privacy preservation. The project makes use of a big data infrastructure built on Hadoop to manage the enormous amount of data needed for spam detection. The detection models were trained using the Multinomial Naive Bayes (MNB) version of the Naive Bayes algorithm. The algorithm distinguishes spam emails with an amazing accuracy rate of 97%. It encourages a cooperative strategy in the fight against spam by enabling users to actively engage in the detection process while respecting their anonymity. This study lays the groundwork for the creation of reliable and privacy-preserving spam detection systems in distributed settings. Future improvements might involve investigating sophisticated machine learning techniques and using real-time data streams for spam prevention. Overall, this secure big data environment offers a solid method for spam detection, protects user privacy, and improves email usage for both individuals and businesses.