Anomaly detection general filtering with basic dimensions (e.g., Account Name) and setting a threshold value to control anomaly noise (e.g., anomalies above $100) in the UI.

See this idea on ideas.ibm.com

Enhance the anomaly detection system by introducing general filtering based on basic dimensions, such as Account Name, and a threshold-setting mechanism within the UI to control anomaly noise. Customer understand that a view can be created, but even within a view, having general filtering available would be helpful to see the relevant data. This feature will allow users to refine anomaly detection results by filtering data based on key attributes, reducing irrelevant noise and improving accuracy. Additionally, as we already have alerts where thresholds can be set, similarly, if the UI allows users to set a threshold to display only anomalies greater than a specific amount defined by the customer, it would improve usability. A configurable threshold will help users define what qualifies as a significant anomaly (e.g., anomalies above $100), ensuring that only meaningful deviations are flagged. By integrating these controls directly into the UI, users can dynamically adjust filters and thresholds, leading to a more efficient and intuitive anomaly monitoring experience.

Idea priority

Medium

Post comment

By clicking the "Post Comment" or "Submit Idea" button, you are agreeing to the IBM Ideas Portal Terms of Use.
Do not place IBM confidential, company confidential, or personal information into any field.

Shape the future of IBM!

Search existing ideas

Post your ideas

Specific links you will want to bookmark for future use

Anomaly detection general filtering with basic dimensions (e.g., Account Name) and setting a threshold value to control anomaly noise (e.g., anomalies above $100) in the UI.

Please enter your email address

RELATED IDEAS

Anomaly detection general filtering with basic dimensions (e.g., Account Name) and setting a threshold value to control anomaly noise (e.g., anomalies above $100) in the UI.