From the course: Artificial Intelligence for Cybersecurity
Join today to access over 25,500 courses taught by industry experts.
Security anomaly detection
From the course: Artificial Intelligence for Cybersecurity
Security anomaly detection
- [Instructor] So what if you don't have label data, or it's too cumbersome or expensive to apply labels to existing data? Well, it turns out that you can still draw useful insights from your data. For example, you want to find patterns in incoming internet traffic to your web server and hopefully, distinguish users from bots. Of course, you don't have label data to train the model, so you build an unsupervised learning model and let it analyze the logs and find interesting patterns of activities that can be useful to you. Such problems fall into the category of so-called clustering problems. Now, why clustering? Because the clustering algorithms group or cluster data with similar characteristics together. But wait, isn't a classification problem same as a clustering problem? Don't they both segregate data into groups? Well, no. In classification, categories are previously known, such as fraud versus genuine, and…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.