Comprehensive Scoring
Last updated
Last updated
Comprehensive scoring also works similar to definite scoring. Here also it considers all log clusters in v2. The difference in comprehensive scoring is it takes the ratio based scoring instead of the actual numbers. Comprehensive scoring works based on the following factors:
Normalizes the score by the total number of unexpected clusters.
Has a different ratio-based scoring mechanism for unexpected clusters present in both v1 and v2.
Comprehensive scoring is tolerant to unexpected events that are present in both v1 and v2 unless the ratio of those events is much higher in v2. Hence, Comprehensive Scoring is used in scenarios where there are unexpected clusters in v1 that you may want to ignore in v2.
Score Reduction is higher at higher sensitivity levels.
Score Reduction is higher for clusters with higher criticality.
By default, comprehensive scoring will give a score of 0 (fail) whenever there is a critical error.
To set the Comprehensive Scoring algorithm follow the steps below:
User can select the Comprehensive Scoring algorithm while creating a Log Template for the verification gate. Users can set the Comprehensive Scoring algorithm after creating a pipeline to the application.
To create an application, refer here.
To create a pipeline, refer here.
To know more about how to create a Log Template, refer here.
After creating a pipeline, go to Pipeline Builder page and select your pipeline and then click Add Stage.
Next, you get the option to select the type of stage. Select the Verification Gate from the drop down menu.
After selecting Verification gate, the following options appear and click "+" Symbol to create a log template:
The New Log Template form appears as shown below:
Select Comprehensive Scoring from the drop-down while creating a new Log Template as shown below:
