Program identifier response to unstructured input
A computer-implemented method for identifying a problem from unstructured input includes executing on a computer processor the step of identifying context of a problem description from a service support k ticket which adds one or more tags to the service support ticket, each tag corresponding to an end-user symptom within the problem domain. Intent is mapped according to a machine learning model and the one or more tags which identifies a problem and a confidence measure.
1. A computer-implemented method, comprising executing on a computer processor:
parsing a service support ticket into structured and unstructured content, wherein the unstructured content comprises a problem description;
identifying an entity and a problem domain within the structured content; adding one or more tags to the service support ticket in response to identifying a context of the problem description, wherein each of the added one or more tags corresponds to a different end-user symptom within the problem domain;
tracking problems longitudinally by entity to thereby tally occurrences of problems that are mapped to the different end-user symptoms and associated with the identified entity within a plurality of generated service support tickets; and
identifying the mapped problem having a highest tally of occurrences associated with the identified entity as a likely root cause of the service support ticket.
2. The method of claim 1 , further comprising:
integrating computer-readable program code into a computer system comprising a processor, a computer readable memory in circuit communication with the processor, and a computer readable storage medium in circuit communication with the processor; and
wherein the processor executes program code instructions stored on the computer-readable storage medium via the computer readable memory and thereby performs the steps of parsing the service support ticket into structured and unstructured content, identifying the entity and the problem domain within the structured content, adding the one or more tags to the service support ticket in response to identifying the context of the problem description, tracking the problems longitudinally by entity to thereby tally occurrences of problems that are mapped to the different end-user symptoms and associated with the identified entity within the plurality of generated service support tickets, and identifying the mapped problem having the highest tally of occurrences associated with the identified entity as the likely root cause of the service support ticket.
3. The method of claim 2 , wherein the computer-readable program code is provided as a service in a cloud environment.
4. The computer-implemented method of claim 1 , further comprising:
identifying the context of the problem description as a function of cleaning the problem description and applying semantic analysis to the cleaned problem description.
5. The computer-implemented method of claim 1 , further comprising:
determining a confidence measure for identifying the mapped problem having the highest tally of occurrences associated with the identified entity as the likely root cause according to a machine learning model.
6. The computer-implemented method of claim 5 , further comprising:
training the machine learning model on a training corpus that includes the one or more tags added to the service support ticket and a corresponding problem of previously resolved service support tickets.
7. A system for identifying a problem from unstructured input, comprising:
a processor;
a computer readable memory in circuit communication with the processor; and
a computer readable storage medium in circuit communication with the processor;
wherein the processor executes program instructions stored on the computer-readable storage medium via the computer readable memory and thereby:
parses a service support ticket into structured and unstructured content, wherein the unstructured content comprises a problem description;
identifies an entity and a problem domain within the structured content;
adds one or more tags to the service support ticket in response to identifying a context of the problem description, wherein each of the added one or more tags corresponds to a different end-user symptom within the problem domain;
tracks problems longitudinally by entity to thereby tally occurrences of problems that are mapped to the different end-user symptoms and associated with the identified entity within a plurality of generated service support tickets; and
identifies the mapped problem having a highest tally of occurrences associated with the identified entity as a likely root cause of the service support ticket.
8. The system of claim 7 , wherein the processor executes program instructions stored on the computer-readable storage medium via the computer readable memory and thereby:
identifies the context of the problem description as a function of cleaning the problem description and applying semantic analysis to the cleaned problem description.
9. The system of claim 7 , wherein the processor executes program instructions stored on the computer-readable storage medium via the computer readable memory and thereby:
determines a confidence measure for identifying the mapped problem having the highest tally of occurrences associated with the identified entity as the likely root cause according to a machine learning model.
10. The system of claim 9 , wherein the processor executes program instructions stored on the computer-readable storage medium via the computer readable memory and thereby:
trains the machine learning model on a training corpus that includes the one or more tags added to the service support ticket and a corresponding problem of previously resolved service support tickets.
11. A computer program product, comprising:
a computer readable storage medium having computer readable program code embodied therewith, wherein the computer readable storage medium is not a transitory signal per se, the computer readable program code comprising instructions for execution by a processor that cause the processor to:
parse a service support ticket into structured and unstructured content, wherein the unstructured content comprises a problem description;
identify an entity and a problem domain within the structured content;
add one or more tags to the service support ticket in response to identifying a context of the problem description, wherein each of the added one or more tags corresponds to a different end-user symptom within the problem domain;
track problems longitudinally by entity to thereby tally occurrences of problems that are mapped to the different end-user symptoms and associated with the identified entity within a plurality of generated service support tickets; and
identify the mapped problem having a highest tally of occurrences associated with the identified entity as a likely root cause of the service support ticket.
12. The computer program product of claim 11 , wherein the instructions for execution cause the processor to:
identify the context of the problem description as a function of cleaning the problem description and applying semantic analysis to the cleaned problem description.
13. The computer program product of claim 11 , wherein the instructions for execution cause the processor to:
determine a confidence measure for identifying the mapped problem having the highest tally of occurrences associated with the identified entity as the likely root cause according to a machine learning model.
14. The computer program product of claim 13 , wherein the instructions for execution cause the processor to:
train the machine learning model on a training corpus that includes the one or more tags added to the service support ticket and a corresponding problem of previously resolved service support tickets.