350-201 Question 51
Select 3A midsize enterprise is experiencing frequent unauthorized login attempts on their internal systems. The cybersecurity team wants to implement an AI-powered data analytic technique to detect and mitigate these threats in real time. Which of the following techniques would be most appropriate to meet their needs?
- A
Anomaly detection using machine learning to identify unusual login patterns
- B
Natural Language Processing (NLP) to analyze emails for phishing attempts
- C
Time-series analysis to detect deviations in login behavior based on timestamps
- D
Supervised learning to classify login attempts as either legitimate or malicious using a labeled dataset
- E
Sentiment analysis to assess the emotional tone of login requests
Show answer and explanation
Correct answers: A, C, D
Explanation
Detecting unauthorized login attempts requires techniques that analyze patterns and classify behaviors. Anomaly detection and time-series analysis are effective for identifying unusual login activities, while supervised learning can classify login attempts if trained with labeled data. NLP and sentiment analysis, while valuable for other cybersecurity use cases, are not relevant to this specific scenario.
- A. Correct.
Anomaly detection using machine learning is highly effective for identifying irregular or unusual login patterns that could indicate unauthorized access. This technique is well-suited for real-time threat detection.
- B. Incorrect.
While NLP is useful for analyzing text-based threats like phishing emails, it is not relevant for detecting unauthorized login attempts in this scenario.
- C. Correct.
Time-series analysis is a powerful technique to detect patterns or deviations in login attempts over time (e.g., unusual activity during off-hours), making it applicable for this use case.
- D. Correct.
Supervised learning can be used to classify login attempts as legitimate or malicious if there is a labeled dataset available for training the model, making it a relevant choice here.
- E. Incorrect.
Sentiment analysis focuses on evaluating emotional tone, which is not applicable to login behavior or cybersecurity threat detection.