Prerequisites for AI modelling
For each different AI algorithm, you must integrate with the right tools and gather specific types of data so the AI can begin learning.
This topic describes the prerequisites for each of the following AI algorithms.
- Change risk and Similar tickets
- Log anomaly detection
- Alert grouping, Alert seasonality detection, and Probable cause
Change risk and Similar tickets
Both the Change risk algorithm and the Similar ticket algorithm require ticket data from ServiceNow. Before attempting to configure these algorithms, ensure that you have created a functioning data integration to ServiceNow. For more information, see Creating a ServiceNow integration.
Log anomaly detection
The Log anomaly detection algorithms require log data from one or more of your log systems. Before attempting to configure these algorithms, ensure that you have created a functioning data integration to at least one of the log systems listed in the following table.
Log system | For more details, see: |
---|---|
Falcon LogScale | Creating Falcon LogScale integrations |
Mezmo | Creating Mezmo integrations |
Splunk | Creating Splunk integrations |
You can also load log data into the system using one of these generic data loading methods.
Data loading method | For more details, see: |
---|---|
Custom | Creating custom integrations |
ELK | Creating ELK integrations |
Kafka | Creating Kafka integrations |
Event grouping, Alert seasonality detection, and Probable cause
The three alert grouping algorithms, the Alert seasonality detection algorithm, and the Probable cause algorithm all require event data. In the case of the Topological grouping algorithm and the Probable cause algorithm, topological data is also required. This topological data must be correlated with the event data.
Event data
In some cases, your event data is loaded into IBM Cloud Pak® for AIOps from a IBM® Netcool® Operations Insight®-based data source by using the following Netcool Operations Insight integration.
Data loading method | For more details, see: |
---|---|
Connecting an external ObjectServer | Creating IBM® Tivoli® Netcool®/OMNIbus integrations |
You can also load event data into the system by using other integration types. For more information, see Incoming integrations.
Topological data
Load topology data into IBM Cloud Pak® for AIOps by configuring observers. For more information, see Defining observer jobs.