Comment on page
Detectors
Nightfall offers core machine learning-powered detectors that you can leverage out of the box. These detectors cover a range of sensitive information types, such as:
- Personally Identifiable Information (PII)
- Secrets and Credentials
- Protected Health Information
- Finance - PCI & Banking
- Image-based Documents - NEW
- Network & Hardware
For the latest list of Nightfall detectors, please visit the Detectors page within the Nightfall Dashboard.
Accuracy & testing: Our detectors use machine learning to yield 3x higher accuracy than traditional regex and rules-based approaches.
Our detectors are trained to recognize sensitive data in natural language text, source code, logs, HTML, JSON/XML, spreadsheets, and URLs for accuracy in real-world situations.
Improving Model-Fit: Even so, your business use case and data may introduce patterns outside our model's training dataset. If this is the case, you can tune detection by working with our data science team, by extending our detectors using context and token rules, or by creating your own detector.
Data Science Team Sessions: Often the best option to improve accuracy is to reach out to your Customer Success Manager to set up a time to work directly with Nightfall's data science team.
Context Rules: Most detectors can be extended to recognize specific contextual patterns in your data. For example, if your company's support ticketing ID numbers look like US social security numbers, you can extend our US SSN detector with a context rule to exclude them.
Example: Exclude US SSN findings that include any following strings [ticket#, support ID, support ticket:, ticket number] in the 15 characters preceding the finding.
Token Rules: You can create an exclude list based on the finding itself using regexes and dictionaries. For example, if there are fake SSNs in your data, you can use a token rule to recognize and exclude them from alerting.
Example: Exclude US SSN findings that include any following strings [234-56-7890, 234567890, 876-54-3210, 876543210] in the the finding.
See Custom Detectors for more information on using Context and Token Rules and options for creating your own detectors.
If you don’t see what you need in our core detector library, you can create custom detectors or regexes. While Nightfall's pre-built detectors listed above are trained via machine learning, Nightfall also supports RE2 regexes and word lists for any custom detectors that may be of interest to you. Over time, we've aggregated the following regex library, which you're welcome to select from to save you some time. Please note that a regular expression is an established yet limited method that searches for pre-defined patterns, so your mileage may vary. You can test regular expressions here. You can input custom detectors directly in the Nightfall console at app.nightfall.ai by navigating to Detectors → New Detector → Regular expression.
We can work with you to apply a range of other detectors. We continue to build a broad library of both machine learning and regex-based detectors spanning different industries, geographies, and compliance regimes, and our data science team continues to innovate upon new detectors on an ongoing basis. For additional beta detectors you can enable in your account, please e-mail us at [email protected].
Last modified 1mo ago