AI Root Causes - Revision history

TeeHiet: TK-63112

2026-02-23T13:16:22Z

TK-63112

← Older revision		Revision as of 13:16, 23 February 2026
Line 27:		Line 27:
	'''Presentation of the Results'''<br>		'''Presentation of the Results'''<br>
	The analysis results are presented in a hierarchical structure that reveals relationships between contributing factors:		The analysis results are presented in a hierarchical structure that reveals relationships between contributing factors:
	* Hierarchical Display: Contributors are organized in a hierarchy where items appearing in parent nodes are not repeated in child nodes. This prevents redundancy and makes it easier to understand how different factors combine to influence outcomes.		* Hierarchical Display: Contributors are organized in a hierarchy where items appearing in parent nodes are not repeated in child nodes. This prevents redundancy and makes it easier to understand how different factors combine to influence outcomes. However, when the table is sorted or column filtering is applied, all contributors are shown on each row.
	* Conditional Hierarchy View: The hierarchy layout is displayed only when the table is not sorted by another column. Sorting by a different column collapses the hierarchy to show a flat list view, allowing for alternative ways to explore the data.		* Conditional Hierarchy View: The hierarchy layout is displayed only when the table is not sorted by another column. Sorting by a different column collapses the hierarchy to show a flat list view, allowing for alternative ways to explore the data.
	This structured approach enables the identification of underlying issues with precision and depth.		This structured approach enables the identification of underlying issues with precision and depth.

TeeHiet: TK-62256

2025-12-11T09:43:14Z

TK-62256

← Older revision		Revision as of 09:43, 11 December 2025
Line 25:		Line 25:
	* Case duration		* Case duration

			'''Presentation of the Results'''<br>
			The analysis results are presented in a hierarchical structure that reveals relationships between contributing factors:
			* Hierarchical Display: Contributors are organized in a hierarchy where items appearing in parent nodes are not repeated in child nodes. This prevents redundancy and makes it easier to understand how different factors combine to influence outcomes.
			* Conditional Hierarchy View: The hierarchy layout is displayed only when the table is not sorted by another column. Sorting by a different column collapses the hierarchy to show a flat list view, allowing for alternative ways to explore the data.
	This structured approach enables the identification of underlying issues with precision and depth.		This structured approach enables the identification of underlying issues with precision and depth.

Ollvihe at 15:10, 25 November 2025

2025-11-25T15:10:13Z

← Older revision		Revision as of 15:10, 25 November 2025
Line 5:		Line 5:

	<div style="border:1px solid #dfdfdf;padding:0.5em 1em 0.5em 1em;background-color:#E7EAEC;margin:10px 0px 0px 10px;">		<div style="border:1px solid #dfdfdf;padding:0.5em 1em 0.5em 1em;background-color:#E7EAEC;margin:10px 0px 0px 10px;">
	The AI Root Cause Analysis is currently unavailable in the Snowflake Native App because the underlying ~~Snoflake~~ Top Insights feature doesn't work in Native apps. However, Snowflake is actively addressing this issue, and we anticipate improvements in the near future.		The AI Root Cause Analysis is currently unavailable in the Snowflake Native App because the underlying Snowflake Top Insights feature doesn't work in Native apps. However, Snowflake is actively addressing this issue, and we anticipate improvements in the near future.
	</div>		</div>

Ollvihe at 15:09, 25 November 2025

2025-11-25T15:09:37Z

← Older revision		Revision as of 15:09, 25 November 2025
Line 5:		Line 5:

	<div style="border:1px solid #dfdfdf;padding:0.5em 1em 0.5em 1em;background-color:#E7EAEC;margin:10px 0px 0px 10px;">		<div style="border:1px solid #dfdfdf;padding:0.5em 1em 0.5em 1em;background-color:#E7EAEC;margin:10px 0px 0px 10px;">
	AI Root ~~Causes analysis doesn't~~ currently ~~work~~ in the Snowflake Native App because the underlying Snoflake Top Insights ~~calculation is not available for~~ Native apps. Snowflake is ~~working on~~ this issue so we~~'ll expect improvement~~ in the future.		The AI Root Cause Analysis is currently unavailable in the Snowflake Native App because the underlying Snoflake Top Insights feature doesn't work in Native apps. However, Snowflake is actively addressing this issue, and we anticipate improvements in the near future.
	</div>		</div>

Ollvihe at 15:06, 25 November 2025

2025-11-25T15:06:10Z

← Older revision		Revision as of 15:06, 25 November 2025
Line 7:		Line 7:
	AI Root Causes analysis doesn't currently work in the Snowflake Native App because the underlying Snoflake Top Insights calculation is not available for Native apps. Snowflake is working on this issue so we'll expect improvement in the future.		AI Root Causes analysis doesn't currently work in the Snowflake Native App because the underlying Snoflake Top Insights calculation is not available for Native apps. Snowflake is working on this issue so we'll expect improvement in the future.
	</div>		</div>


	== How AI Root Cause Analysis Works ==		== How AI Root Cause Analysis Works ==

Ollvihe at 15:06, 25 November 2025

2025-11-25T15:06:02Z

← Older revision		Revision as of 15:06, 25 November 2025
Line 3:		Line 3:

	There are presets that simplify this analysis by providing pre-configured viewpoints. Instead of manually selecting and configuring attributes for analysis, you can use presets to quickly investigate common areas of interest.		There are presets that simplify this analysis by providing pre-configured viewpoints. Instead of manually selecting and configuring attributes for analysis, you can use presets to quickly investigate common areas of interest.

			<div style="border:1px solid #dfdfdf;padding:0.5em 1em 0.5em 1em;background-color:#E7EAEC;margin:10px 0px 0px 10px;">
			AI Root Causes analysis doesn't currently work in the Snowflake Native App because the underlying Snoflake Top Insights calculation is not available for Native apps. Snowflake is working on this issue so we'll expect improvement in the future.
			</div>

	== How AI Root Cause Analysis Works ==		== How AI Root Cause Analysis Works ==

Ollvihe: /* How AI Root Cause Analysis Works */

2025-11-13T12:21:10Z

How AI Root Cause Analysis Works

← Older revision		Revision as of 12:21, 13 November 2025
Line 6:		Line 6:
	== How AI Root Cause Analysis Works ==		== How AI Root Cause Analysis Works ==
	AI Root Cause Analysis begins by constructing features based on user-configured settings. These features are then analyzed by the Snowflake Top Insights tool to identify the most impactful combinations of correlating features. Leveraging a decision tree model, the analysis can uncover complex correlations among features. For instance, it can detect the following types of combinations:		AI Root Cause Analysis begins by constructing features based on user-configured settings. These features are then analyzed by the Snowflake Top Insights tool to identify the most impactful combinations of correlating features. Leveraging a decision tree model, the analysis can uncover complex correlations among features. For instance, it can detect the following types of combinations:
	* Combination of ~~Different Features~~: Such as when attribute A value is B and attribute C value is D.		* Combination of different features: Such as when attribute A value is B and attribute C value is D.
	* Combination of ~~Feature Values~~: Such as when attribute value is either A or B.		* Combination of feature values: Such as when attribute value is either A or B.
	* ~~Numeric~~ and ~~Date Value Ranges~~: ~~This includes identifying conditions where~~ an attribute is greater than or less than a ~~specified~~ value.		* Date and numeric value ranges: For example, when an attribute is greater than or less than a specific value.
	* Inverse of ~~Feature Values~~: For example, cases where the attribute value is not A.		* Inverse of feature values: For example, cases where the attribute value is not A.

	The analysis can evaluate various features as potential root causes, including:		The analysis can evaluate various features in the case and its events as potential root causes, including:
	* Values of case attributes		* Values of case attributes
	* Values of event attributes		* Values of event attributes (whether certain event attribute values exist in the events of the case)
	* Specific event types occurring within a case		* Specific event types occurring within a case
	* Whether the case has a specific flow (transition between two ~~consecutive~~ events)		* Whether the case has a specific flow (transition between two successive events)
	* Starting and ending events		* Starting and ending events in a case
	* Variation within ~~cases~~		* Variation within a case (sequence of events in a case)
	* ~~Duration of cases~~		* Case duration

	This structured approach enables the identification of underlying issues with precision and depth.		This structured approach enables the identification of underlying issues with precision and depth.

Ollvihe: /* How AI Root Cause Analysis Works */

2025-11-13T12:15:53Z

How AI Root Cause Analysis Works

← Older revision		Revision as of 12:15, 13 November 2025
Line 6:		Line 6:
	== How AI Root Cause Analysis Works ==		== How AI Root Cause Analysis Works ==
	AI Root Cause Analysis begins by constructing features based on user-configured settings. These features are then analyzed by the Snowflake Top Insights tool to identify the most impactful combinations of correlating features. Leveraging a decision tree model, the analysis can uncover complex correlations among features. For instance, it can detect the following types of combinations:		AI Root Cause Analysis begins by constructing features based on user-configured settings. These features are then analyzed by the Snowflake Top Insights tool to identify the most impactful combinations of correlating features. Leveraging a decision tree model, the analysis can uncover complex correlations among features. For instance, it can detect the following types of combinations:
	~~* ~~Combination of Different Features**: Such as when attribute A value is B and attribute C value is D.		* Combination of Different Features: Such as when attribute A value is B and attribute C value is D.
	~~* ~~Combination of Feature Values**: Such as when attribute value is either A or B.		* Combination of Feature Values: Such as when attribute value is either A or B.
	~~* ~~Numeric and Date Value Ranges**: This includes identifying conditions where an attribute is greater than or less than a specified value.		* Numeric and Date Value Ranges: This includes identifying conditions where an attribute is greater than or less than a specified value.
	~~* ~~Inverse of Feature Values**: For example, cases where the attribute value is not A.		* Inverse of Feature Values: For example, cases where the attribute value is not A.

	The analysis can evaluate various features as potential root causes, including:		The analysis can evaluate various features as potential root causes, including:

Ollvihe: /* How AI Root Causes Analysis Works */

2025-11-13T12:15:35Z

How AI Root Causes Analysis Works

← Older revision		Revision as of 12:15, 13 November 2025
Line 4:		Line 4:
	There are presets that simplify this analysis by providing pre-configured viewpoints. Instead of manually selecting and configuring attributes for analysis, you can use presets to quickly investigate common areas of interest.		There are presets that simplify this analysis by providing pre-configured viewpoints. Instead of manually selecting and configuring attributes for analysis, you can use presets to quickly investigate common areas of interest.

	== How AI Root ~~Causes~~ Analysis Works ==		== How AI Root Cause Analysis Works ==
	~~The~~ Root ~~Causes~~ Analysis ~~first constructs so called~~ features based on ~~the configurations~~ user ~~has made~~. ~~The~~ features are then analyzed by the Snowflake Top Insights to ~~find~~ the ~~best~~ combinations of correlating features. ~~The analysis is based on the~~ decision tree model ~~which is able to find very~~ complex ~~correlating combinations of the~~ features. For ~~example~~, the following ~~kinds~~ of combinations ~~can be found~~:		AI Root Cause Analysis begins by constructing features based on user-configured settings. These features are then analyzed by the Snowflake Top Insights tool to identify the most impactful combinations of correlating features. Leveraging a decision tree model, the analysis can uncover complex correlations among features. For instance, it can detect the following types of combinations:
	* Combination of ~~the different features (e.g.,~~ attribute A is B and attribute C is D)		* Combination of Different Features: Such as when attribute A value is B and attribute C value is D.
	* Combination of ~~feature values (e.g.,~~ attribute value is A or B)		* Combination of Feature Values: Such as when attribute value is either A or B.
	* Numeric and ~~date value ranges (e.g.,~~ attribute is greater/less than)		* Numeric and Date Value Ranges: This includes identifying conditions where an attribute is greater than or less than a specified value.
	* Inverse of the ~~features values (e.g.,~~ attribute value is not A)		* Inverse of Feature Values: For example, cases where the attribute value is not A.

	~~Following~~ features ~~can be analyzed~~ as potential root causes:		The analysis can evaluate various features as potential root causes, including:
	* ~~Case attribute values~~		* Values of case attributes
	* ~~Event attribute values~~		* Values of event attributes
	* ~~Case has specific~~ event ~~type~~		* Specific event types occurring within a case
	* ~~Case~~ has specific flow		* Whether the case has a specific flow (transition between two consecutive events)
	* Starting and ending ~~event~~		* Starting and ending events
	* Variation		* Variation within cases
	* ~~Case duration~~		* Duration of cases

			This structured approach enables the identification of underlying issues with precision and depth.

	== AI Root Causes (case attributes) Preset ==		== AI Root Causes (case attributes) Preset ==

Ollvihe at 12:11, 13 November 2025

2025-11-13T12:11:43Z

← Older revision		Revision as of 12:11, 13 November 2025
Line 3:		Line 3:

	There are presets that simplify this analysis by providing pre-configured viewpoints. Instead of manually selecting and configuring attributes for analysis, you can use presets to quickly investigate common areas of interest.		There are presets that simplify this analysis by providing pre-configured viewpoints. Instead of manually selecting and configuring attributes for analysis, you can use presets to quickly investigate common areas of interest.

			== How AI Root Causes Analysis Works ==
			The Root Causes Analysis first constructs so called features based on the configurations user has made. The features are then analyzed by the Snowflake Top Insights to find the best combinations of correlating features. The analysis is based on the decision tree model which is able to find very complex correlating combinations of the features. For example, the following kinds of combinations can be found:
			* Combination of the different features (e.g., attribute A is B and attribute C is D)
			* Combination of feature values (e.g., attribute value is A or B)
			* Numeric and date value ranges (e.g., attribute is greater/less than)
			* Inverse of the features values (e.g., attribute value is not A)

			Following features can be analyzed as potential root causes:
			* Case attribute values
			* Event attribute values
			* Case has specific event type
			* Case has specific flow
			* Starting and ending event
			* Variation
			* Case duration

	== AI Root Causes (case attributes) Preset ==		== AI Root Causes (case attributes) Preset ==