Anomaly detection

Learn about how EDAA™ detects anomalies.

EDAA™ detects mismatches between your Affective Computing by Virtue project's motivation and user responses or actions. These events are considered as anomalies, and EDAA™ can correct these situations.

Advantages

• Anomaly detection helps you identify and address deviations in user behavior. It enhances the accuracy and relevance of personalized actions, realigns your users' emotional states with the expected motivation, reduces anomalies, and maintains engagement. It is especially beneficial in situations that require immediate intervention when a user deviates from the expected behavioral pattern.

• As Anomaly detection is an integrated feature of EDAA™'s Motivation engine, it enables you to automate personalization.

Use case

In solutions in the Safety domain (for example, a solution that aims to minimize road accidents), Anomaly detection can identify behavioral anomalies and help realign the solution's users with the expected behavior, ensuring safety.

How it works

Anomaly detection identifies deviations in perception patterns (with respect to the expected motivations).

EDAA™, through its Sensations engine, performs rigorous data normalization and analysis to understand the consequences of behavioral changes before they cause deviation or anomalies. It uses complex psychological analysis including studying symptoms and pathologies to recognize when user actions or responses are not aligned with their typical motivation.

EDAA™ can also play the role of an observer (passive mode, when the Affective Computing by Virtue project runs in user calibration mode) and trigger actions only when it detects anomalies. This makes it effective in situations that require minimal interaction.