350-201 Question 52
Single answerA financial organization is experiencing an increasing number of fraudulent transactions and wants to identify anomalies in its payment systems in real-time. As a security engineer, you are tasked with recommending an AI-powered data analytic technique to address this issue. Which of the following techniques would be most appropriate for this scenario?
- A
Supervised machine learning using labeled data to classify legitimate and fraudulent transactions
- B
Unsupervised machine learning to detect anomalies without requiring labeled data
- C
Natural language processing (NLP) to analyze customer communication for potential fraud
- D
Sentiment analysis to gauge customer satisfaction trends related to fraudulent transactions
Show answer and explanation
Correct answer: B
Explanation
Unsupervised machine learning is the best option for real-time anomaly detection in payment systems because it can automatically identify unusual patterns or deviations in the transaction data without requiring pre-labeled datasets. This makes it highly effective in scenarios where new or evolving types of fraud need to be detected.
- A. Incorrect.
Supervised machine learning requires labeled datasets. While effective for classifying known patterns of fraud, it may struggle with new or unseen types of fraudulent activities since manual labeling is time-consuming and may not capture evolving threats.
- B. Correct.
Unsupervised machine learning is well-suited for detecting anomalies, as it does not require labeled data. It identifies patterns and deviations in the dataset, making it an ideal choice for real-time detection of unknown or emerging fraudulent behavior.
- C. Incorrect.
Natural language processing (NLP) is primarily used for analyzing text data, such as customer communication. While useful for fraud investigations, it does not directly address the detection of anomalies in payment systems.
- D. Incorrect.
Sentiment analysis focuses on gauging customer sentiment and is unrelated to detecting anomalies or preventing fraudulent transactions in payment systems.