Recommendation of relevant information to support problem diagnosis
The disclosure generally relates to knowledge retrieval using a knowledgebase storing general and/or expert knowledge. In particular, the disclosure relates to using complex retrieval techniques to implement an enhanced troubleshooting system that provides relevant information. Aspects of the disclosure retrieve relevant documents to assist in troubleshooting by reasoning with stored representations about the faulty system and its operating environment. After receiving a description of the symptoms and/or other information related to the problem, the system may recommend textual and/or non-textual documents with relevant information for identifying the cause of the problem and restoring operation.
1. A method of an enhanced troubleshooting system, the method comprising:
a. receiving input data, where the input data comprises a description of symptoms troubling a system;
b. extracting concepts from the input data;
c. providing the concepts to a backend system, where the backend system comprises an optimized knowledgebase combining a company-specific knowledge representation and a change knowledge representation;
d. receiving a results record from the backend system in response to the providing of the concepts; and
e. generating, using a reporting module, a formatted output using the results record, where the formatted output provides information to assist in troubleshooting the system,
where the optimized knowledgebase includes a general knowledge representation, and
where the company-specific knowledge representation, the change knowledge representation, and the general knowledge representation each are stated as natural language assertions that specify conceptual relationships holding among entities.
2. The method of claim 1 , where the step of extracting concepts from the input data includes populating a data structure with the concepts, and where the step of providing the concepts to the backend system includes sending the data structure to the backend system.
3. The method of claim 1 , further comprising: sending change documents to the backend system to update the change knowledge representation at a periodic interval.
4. The method of claim 3 , where the periodic interval is daily.
5. The method of claim 1 , further comprising: sending network architecture diagrams to the backend system to update the change knowledge representation.
6. The method of claim 1 , where the input data is automatically retrieved from a document using a data extraction module.
7. The method of claim 6 , where the document is an incident ticket.
8. The method of claim 1 , where the input data is entered by a user attempting to troubleshoot the system.
9. The method of claim 1 , where the formatted output is integrated into an existing graphical user interface to further assist in troubleshooting the system.
10. The method of claim 1 , where the formatted output is in a form of an ordered list.
11. An apparatus comprising:
a. an electronic processor;
b. a computer memory storing modules configured to be executed by the processor, the modules comprising:
i. a data extraction module configured to extract concepts from input data and provide the concepts to a backend system; and
ii. a report module configured to receive a results record from the backend system in response to the data extraction module providing the concepts to the backend system and formatting output using the results record; and
c. a communications interface configured to receive the input data comprising a description of symptoms troubling a system, and send formatted output,
where the backend system comprises an optimized knowledgebase combining a company-specific knowledge representation and a change knowledge representation, and
where the optimized knowledgebase further combines a general knowledge representation, and
where the company-specific knowledge representation, the change knowledge representation, and the general knowledge representation each are stated as natural language assertions that specify conceptual relationships holding among entities.
12. The method of claim 1 , wherein the change knowledge representation comprises at least one document listing previous changes to the system, at least one previous incident ticket listing prior bugs in the system, software revision history, software patch installations, and hardware updates.
13. The apparatus of claim 11 , where the change knowledge representation is updated on a periodic basis.
14. The apparatus of claim 11 , where the optimized knowledgebase further combines an expert knowledge representation.
15. The apparatus of claim 11 , where the data extraction module provides the concepts to the backend system using a concept vector.
16. A non-transitory computer-readable medium storing computer-executable instructions that, when executed by a processor, cause a system to perform steps of:
a. receiving input data, where the input data comprises a description of symptoms troubling the system;
b. extracting concepts from the input data;
c. populating a data structure with the concepts;
d. sending the data structure to a backend system, where the backend system comprises an optimized knowledgebase combining a general knowledge representation, a company-specific knowledge representation, and a change knowledge representation;
e. receiving a results record from the backend system in response to the providing of the concepts; and
f. generating, using a reporting module, a formatted output using the results record, where the formatted output provides information to assist in troubleshooting the system,
where the input data is automatically retrieved from a document using a data extraction module, and
where the input data is automatically retrieved from the document responsive to the document being selected for display, and wherein a generated display of the document provides an indication that the information is available to assist in troubleshooting the system.
17. The non-transitory computer-readable medium of claim 16 , where the document is an incident ticket.
18. The non-transitory computer-readable medium of claim 16 , where the formatted output is preconfigured to be integrated into an external troubleshooting application.
19. The non-transitory computer-readable medium of claim 16 , where the data structure is a concept vector.