Skip to main content

Risks Page

Updated over 9 months ago

'

Navigating to the Risks page

You can navigate to the risks page by clicking on the Policy tab and then selecting Risks

Overview of Risks page

The risks page allows you to configure and set various risks to watch for on your domain.

There are three types of risks that you can configure:

  • Machine Learning

  • Keyword

  • Regex

Machine Learning (ML):

  • There are two ML models currently, Self Harm and Toxicity. Each of these are configured separately and can only have their thresholds changed. Self Harm is focused on looking for self harm. Toxicity is focused on toxic behavior such as racial slurs, identity attacks and violence.

  • Anything that the ML models flag will be scored between 1-100 and then based on the thresholds you configured it will assign them a Low, Medium or High value.

Keyword:

  • This allows you to set specific words or phrases that you want to scan for and be notified that someone has typed

Regex:

  • Similar to the keyword option, but significantly more powerful. Allows you to input a regex and scan for any input that matches with the regex.

  • NOTE This is an advanced option and you need to be careful before you implement this into your environment as it can cause performance issues on any machine with the extension installed.

  • You can contact ManagedMethods Support if you would like us to take a look at your Regex and help make it more performant for your environment.

Created Risks

On the far left you will see a list of all the risks you have currently configured with the selected domain/OU. Each one of these will have an icon to let you know what kind of risk it is. Below the list of risks is a legend for what each icon represents.

Risk Editor

After you've selected a risk, on the right side of the page a new form will appear and give you a bunch of options to update the risk:

  • Name - Sets the name of the risk

  • Type - Lets you pick between Keyword and Regex

  • Match Count - Number of risk matches required to be flagged as risky.

  • Keywords - List of words to flag

  • Context - List of words to be checked in proximity to a keyword or regular expressions

  • Context Match Length - The length which has to be checked for finding context around a risk.

  • Regex - The Regex pattern you've created

  • Allow - List of keywords to be removed before scanning against the regex pattern

Note: The Regex and Allow fields only appear when you set the type to Regex and the Keywords only appears when you've set the type to Keyword

Creating a New Risk

  1. Click the blue + icon on the left side of the screen.

  2. Set the name of the risk in the field that says New Risk

    • This is the displayed name when something is flagged.

  3. Set the type of risk between Keyword and Regex from the dropdown after clicking the field labeled Type.

  4. Set the match count in the field named Match Count. The default the value is 1.

If the type is Keyword

5a. Add in a word or comma separated list of words you'd like to be flagged into the Keywords field.

if the type is Regex

5a. In the Regex field, input the regex that you created
- If you need help writing your regex or testing it, check out https://regex101.com for a helpful online tool
5b. Add any words, comma separated, that you'd like to ignore for your regex scanning.

  1. In the Context field add in any words, comma separated, to check proximity to.

  2. In the Context Match Length field add in how many characters behind or in front of the context word to find a match

  3. Hit the blue Save button in the top right or hit Cancel to undo all pending changes.

Editing Existing Risks

  1. On the left side where the list of all existing risks are, click on the risk you want to edit

  2. On the right side you will see the selected risk details.

  3. Edit any of the fields as necessary

  4. Hit Save in the top right hand corner to save your changes or hit Cancel to undo all pending changes

Disabling/Enabling a Risk

  1. On the left side where the list of all existing risks are, you can click the toggle to enable or disable a risk.
    ​Note: The green check means it is enabled and the gray X means it is disabled.

Editing a ML Risk

  1. Select one of the ML risks by clicking on it.

  2. On the right hand side you will have 3 options to choose from

  3. Set the threshold of each (Low, Medium, High) to your preferred values.

  4. Hit Save in the top right hand corner or Cancel if you wish to revert all pending changes'

Related Articles
Customize - Editing a Risk
Customize - Creating Risks
Risk - Fields Allowed "Keywords" and "Regex"
Customize - Creating Risks(New)
Customize - Editing Patterns & Regex(New)
Did this answer your question?