1. During an analysis configuration, you can select the Advanced Settings icon (gear shape) to tune Sisu output further.
2. Within the Advanced Settings modal, you can adjust Sisu’s sensitivity in finding facts:
- Min size: This is the minimum size (percentage of rows) that a subgroup must have for it to show up as a fact.
- For example, if there are 100,000 rows in the data set, a subgroup will appear as a fact only if it contains at least 100 rows (0.1% of 100,000 rows).
- For small data sets in particular, reducing this setting can help make more facts appear in the analysis result.
- Confidence level: Sisu tests for statistical significance for all the subgroups within the data set and only surfaces ones that pass the test. By default, the confidence level is set at 95% (or p-level at 0.05). For smaller data sets, you can decrease this level down (even to 50% or 0%) to see more facts that may not be statistically significant.
- Max order: For very large data sets, you may only want to look at Order 1 or Order 2 facts. Order 1 facts are single factor-value pairs, e.g. country = Canada. Order 2 facts are dual factor-value pairs, e.g. country = Canada & gender = F.
For each order, Sisu returns the top 200 facts with the greatest % difference from the overall metric (or overall metric change for Comparison Analysis) and the top 200 facts with the greatest impact in terms of absolute magnitude.