System, method, and computer program for identification of common root causes with sequential patterns

1. A computer program product embodied on a non-transitory computer readable medium, comprising computer code for:

identifying a plurality of customer event sequences, each of the plurality of customer event sequences including a plurality of events associated with a predetermined customer that occur in a particular order;

grouping the plurality of customer event sequences into a hierarchical framework of patterns, including:

organizing the plurality of customer event sequences into a first set of groups based on a type of each of the plurality of events,

for each of the first set of groups, organizing the plurality of customer event sequences within the group into a second set of groups based on the particular order in which the plurality of events occur for each of the plurality of customer event sequences within the group, and

for each of the second set of groups, organizing the plurality of customer event sequences within the group into a third set of groups based on a time duration of each of the plurality of events for each of the plurality of customer event sequences within the group; and

performing a root-cause analysis for the customer event sequences utilizing the hierarchical framework of patterns.

2. The computer program product of claim 1 , wherein grouping the plurality of customer event sequences into a hierarchical framework of patterns includes transforming the plurality of customer event sequences into an N-dimensional space, where N is a positive integer equal to a total number of possible event types.

3. The computer program product of claim 2 , wherein grouping the plurality of customer event sequences into a hierarchical framework of patterns includes mapping the plurality of customer event sequences in the N-dimensional space.

4. The computer program product of claim 1 , wherein grouping the plurality of customer event sequences into a hierarchical framework of patterns includes determining a list of all possible sub-sequences of all lengths for all of the plurality of customer event sequences.

5. The computer program product of claim 4 , wherein grouping the plurality of customer event sequences into a hierarchical framework of patterns includes generating a set of representative sub-sequences from the possible sub-sequences.

6. The computer program product of claim 4 , wherein generating the set of representative sub-sequences from the possible sub-sequences includes iteratively assembling the set of representative sub-sequences using sub-sequences with a highest score, the score being based on a product of a length of a particular sub-sequence and a number of occurrences of the particular sub-sequence in the plurality of customer event sequences.

7. The computer program product of claim 6 , further comprising computer code for utilizing a threshold to exclude rare patterns.

8. The computer program product of claim 7 , wherein the computer program product is operable such that the threshold is based on the number of occurrences of the particular sub-sequence in the plurality of customer event sequences divided by a number of subsequences in the plurality of customer event sequences.

9. The computer program product of claim 1 , wherein grouping the plurality of customer event sequences into a hierarchical framework of patterns includes grouping the plurality of customer event sequences based on a time gap between successive events of the plurality of customer event sequences.

10. The computer program product of claim 9 , wherein the computer program product is operable such that an existence of different time spans between events of the plurality of customer event sequences indicates different patterns.

11. The computer program product of claim 1 , wherein grouping the plurality of customer event sequences into a hierarchical framework of patterns functions to identify dominant patterns in the plurality of customer event sequences.

12. The computer program product of claim 1 , wherein grouping the plurality of customer event sequences into a hierarchical framework of patterns functions to identify standard or expected sub-sequences associated with the plurality of customer event sequences.

13. The computer program product of claim 12 , wherein the computer program product is operable such that sub-sequences not identified as standard or expected sub-sequences are identified as significant to root-cause analysis and are utilized to perform the root-cause analysis for the customer event sequences.

14. The computer program product of claim 1 , further comprising ranking the plurality of customer event sequences in the hierarchical framework of patterns according to one or more business policies.

15. A method, comprising:

identifying, by a system, a plurality of customer event sequences, each of the plurality of customer event sequences including a plurality of events associated with a predetermined customer that occur in a particular order;

grouping, by the system, the plurality of customer event sequences into a hierarchical framework of patterns, including:

organizing the plurality of customer event sequences into a first set of groups based on a type of each of the plurality of events,

for each of the first set of groups, organizing the plurality of customer event sequences within the group into a second set of groups based on the particular order in which the plurality of events occur for each of the plurality of customer event sequences within the group, and

for each of the second set of groups, organizing the plurality of customer event sequences within the group into a third set of groups based on a time duration of each of the plurality of events for each of the plurality of customer event sequences within the group; and

performing, by the system, a root-cause analysis for the customer event sequences utilizing the hierarchical framework of patterns.

16. A system comprising:

a memory system; and

one or more processing cores coupled to the memory system and that are each configured for:

identifying a plurality of customer event sequences, each of the plurality of customer event sequences including a plurality of events associated with a predetermined customer that occur in a particular order;

grouping the plurality of customer event sequences into a hierarchical framework of patterns, including:

organizing the plurality of customer event sequences into a first set of groups based on a type of each of the plurality of events,

for each of the first set of groups, organizing the plurality of customer event sequences within the group into a second set of groups based on the particular order in which the plurality of events occur for each of the plurality of customer event sequences within the group, and

for each of the second set of groups, organizing the plurality of customer event sequences within the group into a third set of groups based on a time duration of each of the plurality of events for each of the plurality of customer event sequences within the group; and

performing a root-cause analysis for the customer event sequences utilizing the hierarchical framework of patterns.