You can reference common or technical terms used by HUMAN and Sightline Cyberfraud Defense here.
| Term | Definition |
|---|---|
| Activity | A semantic action representing one or more events that describe the intent of a visitor’s interactions. |
| Application ID | A HUMAN-assigned identifier for an application. An account may have multiple Application IDs for different environments, such as a development, staging, and production environment, or for different protected domains. Matches the format PX12AB34CD. |
| Console | The HUMAN dashboard, https://console.humansecurity.com/, where you can access HUMAN products and account settings. |
| Detector | The HUMAN Detector is HUMAN’s analysis engine that uses machine learning and behavioral analytics to determine an interaction’s risk score based on data from the HUMAN Sensor and Enforcer. It also continuously learns and improves its models based on reported results. |
| Direct mitigation | Mitigation that is meant to intervene and stop or prevent the user from completing their intended action. Some examples are hard blocking a request or freezing an account. |
| Enforcer | A HUMAN module installed and configured on an application, load balancer, or CDN. Responsible for blocking or allowing requests based off of the risk score sent by the HUMAN Sensor. |
| Indirect mitigation | A special type of mitigation often meant to obfuscate the decision from very sophisticated attackers or indirectly mitigate the risk. Some examples are canceling orders, suppressing ads, or rate limiting. |
| Mitigation | Any action taken to reduce or neutralize the risk or impact of a threat. |
| Organic traffic | Traffic that comes from non-paid interactions. This commonly means users navigating directly to your site through the browser or from a non-ad search engine result. |
| Paid traffic | Traffic from paid referrals or traffic generated through paid advertisements or promotional activities. These include paid advertising campaigns like display ads or search engine marketing. Bots in paid traffic are an indication of Invalid Traffic (IVT). |
| Ripple effect blocks | Traffic that was blocked because Precheck triggered a collection of additional signals. Precheck didn’t block the request immediately, but ensured it was caught later on by HUMAN’s detection mechanism. |
| Risk API | HUMAN’s API usually called by the HUMAN Enforcer to retrieve the risk score of a request from the HUMAN Detector. Typically, Enforcers call the Risk API when the risk cookie is missing, invalid, or expired. |
| Risk cookie | An encrypted cookie with a request’s calculated risk score sent by the HUMAN Sensor to the HUMAN Enforcer. |
| Risk score | A numeric risk value from 0-100 assigned to a request. The HUMAN Sensor calculates a risk score, then sends the risk cookie embedded with the score to the HUMAN Enforcer. Depending on the score, the Enforcer will block or allow the request. |
| Secondary detection mechanism | The use of advanced technology to analyze data and identify threat patterns after the initial block or allow decision is made. In bot management, this means reviewing attack data in aggregate to isolate attack profiles, correlate requests, and react to attackers’ specific adaptations over time. |
| Sensor | The HUMAN Sensor continuously collects client-side indicators and alerts the HUMAN detection system of suspicious activity for further analysis. The Sensor is injected into a website or application with embedded JavaScript. |
| Session | A series of requests and events within a continuous period of time from the same visitor. |
| Soft mitigation | Mitigation that is meant to de-risk a potential threat, usually by gathering more information. By de-risking threats, we reduce the overall friction on good users. Some examples are checking an ID, showing a CAPTCHA, or resetting a password. |
| User journey | A sequence of sessions by a single visitor. |
| Visitor | A unique visitor is defined by a stable ID (VID or device). Also referred to as a user. |