CPC G06F 11/079 (2013.01) [G06F 11/0772 (2013.01); G06F 11/2263 (2013.01); G06F 11/3006 (2013.01); G06F 11/3075 (2013.01); G06F 11/327 (2013.01); G06F 11/3409 (2013.01); G06N 20/00 (2019.01)] | 20 Claims |
1. A computer program product, the computer program product being tangibly embodied on a non-transitory computer-readable storage medium and comprising instructions that, when executed by at least one computing device, are configured to cause the at least one computing device to:
determine an event graph schema for a technology landscape, the technology landscape being characterized by scores assigned to performance metrics for the technology landscape, wherein the event graph schema includes a plurality of nodes corresponding to the performance metrics and the scores, and includes directional edges connecting node pairs of the plurality of nodes, each directional edge having a score-dependent validity criterion defined by the scores of a corresponding node pair;
determine anomalous scores of the scores associated with an event within the technology landscape;
determine, from the anomalous scores, anomalous nodes;
generate an event graph instance of the event graph schema including instantiating valid edges from the directional edges, each valid edge connecting two of the anomalous nodes and satisfying the score-dependent validity criterion of the directional edges;
determine at least one path within the event graph instance that includes the valid edges and connected anomalous nodes;
traverse the at least one path to identify at least one of the connected anomalous nodes as a root cause node of the event;
store the scores in association with the event to obtain labelled training data;
train a machine learning model using the labelled training data and a supervised machine learning algorithm; and
predict a future event, based on the trained machine learning model and current values of the scores.
|