Determining dependent causes of a computer system event
Disclosed are methods and systems for determining combinations of system parameters that indicate a root cause of a system level experience deterioration (SLED). Some of the disclosed embodiments generate a decision tree from a first class of operational parameter datasets. Rules are derived from the decision tree. Filtered rule sets for feature parameters included in the system parameters are then determined. Pairs of features within a particular dataset that each satisfy their respective filtered rule sets are indicative of a root cause of the degradation, at least in some embodiments.
1. A system comprising:
hardware processing circuitry;
one or more memories storing instructions that when executed cause the hardware processing circuitry to:
obtain a plurality of datasets, each dataset of the plurality of datasets comprising one or more operational parameters that each include a feature parameter and a target parameter;
classify each dataset of the plurality of datasets based on a corresponding target parameter of a corresponding dataset;
generate, based on the classification of each dataset of the plurality of datasets, a rule set including one or more sub-rules;
determine, based on identifying one or more datasets of the plurality of datasets that satisfy at least one of the one or more sub-rules, one or more feature parameters of the one or more datasets that satisfy the one or more sub-rules; and
perform a remedial action determined from the one or more feature parameters of the one or more datasets that satisfy at least one of the one or more sub-rules.
2. The system of claim 1 , wherein to classify each dataset of the one or more datasets, the instructions cause the processing circuitry to:
evaluate the corresponding target parameter of the corresponding dataset with a criterion; and
classify, based on evaluating the corresponding target parameter with the criterion, the corresponding dataset as a first class indicating a degradation in service level experience (SLE) or a second class indicating a nominal SLE.
3. The system of claim 2 , wherein to classify the corresponding dataset as the first class or the second class, the instructions cause the processing circuitry to:
determine a probability that the corresponding target parameter is associated with the first class or the second class.
4. The system of claim 1 , wherein to generate the rule set, the instructions cause the processing circuitry to:
generate, based on the classification of each dataset of the one or more datasets, a decision tree.
5. The system of claim 4 , wherein the decision tree comprises one or more nodes that each indicate a classification of the corresponding dataset and is associated with one or more rules of the rule set.
6. The system of claim 1 , wherein to determine the one or more feature parameters of the one or more datasets that satisfy the one or more sub-rules, the instructions cause the processing circuitry to:
determine a frequency of the one or more feature parameters that satisfy the one or more sub-rules.
7. The system of claim 6 , wherein to determine the one or more feature parameters of the one or more datasets that satisfy the one or more sub-rules, the instructions cause the processing circuitry to:
identify, based on the frequency of the one or more feature parameters that satisfy the one or more sub-rules, a subset of one or more sub-rules.
8. The system of claim 1 , wherein to determine the one or more feature parameters of the one or more datasets that satisfy the one or more sub-rules, the instructions cause the processing circuitry to:
determine a probability that the one or more feature parameters of the one or more datasets that satisfy the one or more sub-rules is a root cause of a degradation in service level experience (SLE).
9. The system of claim 1 , wherein to determine the one or more feature parameters of the one or more datasets that satisfy the one or more sub-rules, the instructions cause the processing circuitry to:
determine whether each of the one or more feature parameters that satisfy the one or more sub-rules is an outlier value.
10. A method comprising:
obtaining, by processing circuitry, one or more datasets of operational parameters associated with one or more devices of a network, each operational parameter comprising a feature parameter and a target parameter;
classifying, by the processing circuitry, each dataset of the one or more datasets based on a corresponding target parameter of a corresponding dataset;
generating, based on classifying each dataset of the one or more datasets, a rule set including one or more sub-rules;
determining one or more feature parameters of the one or more datasets that satisfy at least one of the one or more sub-rules; and
performing a remedial action determined based on the one or more feature parameters of the one or more datasets that satisfy at least one of the one or more sub-rules.
11. The method of claim 10 , wherein classifying each dataset of the one or more datasets comprises:
evaluating the corresponding target parameter of the corresponding dataset with a criterion; and
classifying, based on evaluating the corresponding target parameter with the criterion, the corresponding dataset as a first class indicating a degradation in service level experience (SLE) or a second class indicating a nominal SLE.
12. The method of claim 11 , wherein classifying the corresponding dataset as the first class or the second class comprises:
determining a probability that the corresponding target parameter is associated with the first class or the second class.
13. The method of claim 10 , wherein generating the rule set comprises:
generating, based on the classifying of each dataset of the one or more datasets, a decision tree.
14. The method of claim 13 , wherein the decision tree comprises one or more nodes that each indicate a classification of the corresponding dataset and is associated with one or more rules of the rule set.
15. The method of claim 10 , wherein determining the one or more feature parameters of the one or more datasets that satisfy the one or more sub-rules comprises:
determining a frequency of the one or more feature parameters that satisfy the one or more sub-rules.
16. The method of claim 15 , wherein determining the one or more feature parameters of the one or more datasets that satisfy the one or more sub-rules comprises:
identifying, based on the frequency of the one or more feature parameters that satisfy the one or more sub-rules, a subset of one or more sub-rules.
17. The method of claim 10 , wherein determining the one or more feature parameters of the one or more datasets that satisfy the one or more sub-rules comprises:
determining a probability that the one or more feature parameters of the one or more datasets that satisfy the one or more sub-rules is a root cause of a degradation in service level experience (SLE).
18. The method of claim 10 , wherein determining the one or more feature parameters of the one or more datasets that satisfy the one or more sub-rules comprises:
determining whether each of the one or more feature parameters that satisfy the one or more sub-rules is an outlier value.
19. Non-transitory computer-readable storage media comprising instructions that when executed by processing circuitry, cause the processing circuitry to:
obtain one or more datasets of operational parameters associated with one or more devices of a network, each operational parameter comprising a feature parameter and a target parameter;
classify each dataset of the one or more datasets based on a corresponding target parameter of a corresponding dataset;
generate, based on the classification of each dataset of the one or more datasets, a rule set including one or more sub-rules;
determine one or more feature parameters of the one or more datasets that satisfy at least one of the one or more sub-rules; and
perform a remedial action determined based on the one or more feature parameters of the one or more datasets that satisfy at least one of the one or more sub-rules.