Abstract

The paper presents a novel approach to addressing the challenges of multiuser operations in Twitter by focusing on the expense of queries and the existing solutions for data partitioning. It introduces a method for reducing server interaction by partitioning data based on user interactions and implementing selective replication for frequently requested user data. This approach significantly improves partition quality, especially with low replication ratios. Additionally, the paper delves into the issue of Twitter spammers and their evolving tactics, proposing new detection features to combat spam accounts. It also explores the detection of duplicate fake accounts on Twitter using feature based analysis and machine learning techniques, with SVM demonstrating the best performance at 93.3% accuracy

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call