NLPIR SEMINAR 15th ISSUE COMPLETED

Last week, Yvette gave a presentation about the paper, Credit Card Fraud Detection using Non-Overlapped Risk based Bagging Ensemble (NRBE), and shared some opinion on it.

After the presentation, several questions were asked. The Q&As are listed as following:

Q: can you tell me why Naïve Bayes is used in this paper?
A: To handle the noise and imbalance of the data

Q: What is the innovation of this article compared to related work?
A: He proposed the Non-Overlapped Risk based Bagging Ensemble(NRBE)

Q: In the data set how many are fraud?
A: The Brazilian Bank dataset consists of 0.3 million instances and exhibits an imbalance ratio of 25.7, but don’t have statistic the mount of the fraud data.