Conversational Problem Determination based on Bipartite Graph

• US 20190188067A1
• Filed: 12/15/2017
• Published: 06/20/2019
• Est. Priority Date: 12/15/2017
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Chen, Hao, Dang, Ya Bin, Li, Qi Cheng, Li, Shao Chun, Mei, Li Jun, Nahamoo, David, Wang, Jian, Yu, Yi Peng
• Granted Patent: US 10,635,521 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 11/0709 in a distributed system consisting of a plurality of standalone computer nodes, e.g. clusters, client-server systems
• CPC Class Codes: G06F 11/0751 Error or fault detection not based on redundancy power supply failures G06F1/30; network fault management H04L41/06
• CPC Class Codes: G06F 11/0775 Content or structure details of the error report, e.g. specific table structure, specific error fields
• CPC Class Codes: G06F 11/079 Root cause analysis, i.e. error or fault diagnosis in a hardware test environment G06F11/22; in a software test environment G06F11/36

Abstract

A cognitive conversation system that generates effective diagnostic questions is provided. The cognitive conversation system receives a set of currently known symptoms (or currently available answers to diagnostic questions) of a reported problem or fault. The system identifies (i) a set of possible root causes of the reported problem based on the currently known symptoms and (ii) probabilities for the set of possible root causes by using a bipartite graph data structure that links possible symptoms with possible root causes. Upon determining that at least one possible root cause has a probability that is higher than a threshold, the system presents an explanation or solution associated with the at least one possible root cause. Upon determining that none of the possible root causes in the set of possible root causes has a probability higher than the threshold, the system presents a question based on information entropy that is computed based on probabilities of the identified possible root causes.

Claims

1. A computer-implemented method comprising:

harvesting a set of symptoms from a conversation between the computer and a user, where the set of symptoms is related to a reported problem;

retrieving a bipartite graph data structure that links possible root causes with possible symptoms from a memory of the computer;

identifying, by the computer, (i) a set of possible root causes of the reported problem based on the set of symptoms and (ii) a probability for each possible root cause in the set of possible root causes, by using the bipartite graph data structure;

upon determining, by the computer, that at least one possible root cause has a probability that is higher than a threshold, presenting, as part of the conversation, an explanation or solution associated with the at least one possible root cause having a probability that is higher than the threshold; and

upon determining, by the computer that none of the possible root causes in the set of possible root causes has a probability higher than the threshold, presenting, as part of the conversation, a question based on an information entropy that is computed based on each of the probabilities of the identified possible root causes of the set of root causes.

2. The computer-implemented method of claim 1, further comprising identifying one or more possible symptoms in addition to the set of symptoms for the set of possible root causes, wherein the presented question is selected from a set of candidate questions that are associated with the identified possible symptoms.

3. The computer-implemented method of claim 2, wherein selecting the question comprises identifying a candidate question among the set of candidate questions that achieves a greatest reduction in the information entropy.

4. The computer-implemented method of claim 3:

wherein the information entropy is computed based on each of the probabilities of the different possible root causes, and

wherein the reduction in the information entropy is computed based on probabilities of (i) different possible symptoms associated with the candidate question and (ii) links between the different possible symptoms associated with the candidate question and the possible root causes.

5. The computer-implemented method of claim 1, further comprising:

receiving an answer for the presented question;

identifying a symptom based on the answer; and

adding the identified symptom to the set of symptoms.

6. The computer-implemented method of claim 1, further comprising eliminating the possible root cause associated with the presented explanation or solution from the set of possible root causes upon determining that a feedback for the presented explanation or solution indicates that the explanation or solution is inappropriate for the reported problem.

7. The computer-implemented method of claim 1, further comprising updating the bipartite data structure based on a response to the presented question or the presented explanation received from the conversation.

8. The computer-implemented method of claim 7, further comprising creating the bipartite data structure by extracting the possible symptoms, the possible root causes, and relationships between the possible symptoms and the possible root causes from a set of historical records.

9. A computing device comprising:

a processor;

a storage device storing a set of instructions, wherein an execution of the set of instructions by the processor configures the computing device to perform acts, comprising;

conducting a conversation over a user interface device;

receiving a set of symptoms of a reported problem from the conversation;

identifying (i) a set of possible root causes of the reported problem based on the set of symptoms and (ii) a probability for each possible root cause in the set of possible root causes by using a bipartite graph data structure that links possible root causes with possible symptoms;

upon determining that at least one possible root cause has a probability that is higher than a threshold, presenting, as part of the conversation, an explanation or solution associated with the at least one possible root cause having a probability that is higher than the threshold; and

upon determining that none of the possible root causes in the set of possible root causes has a probability higher than the threshold, presenting, as part of the conversation, a question based on an information entropy that is computed based on each of the probabilities of the identified possible root causes of the set of root causes.

10. The computing device of claim 9, wherein execution of the set of instructions further configures the computing device to perform acts comprising:

identifying one or more possible symptoms in addition to the set of symptoms for the set of possible root causes, wherein the presented question is selected from a set of candidate questions that are associated with the identified possible symptoms.

11. The computing device of claim 10:

wherein selecting the question comprises identifying a candidate question among the set of candidate questions that achieves a greatest reduction in the information entropy,

wherein the information entropy is computed based on the probabilities of the different possible root causes, and

wherein the reduction in the information entropy is computed based on probabilities of (i) different possible symptoms associated with the candidate question and (ii) links between the different possible symptoms associated with the candidate question and the possible root causes.

12. The computing device of claim 9, wherein execution of the set of instructions further configures the computing device to perform acts comprising:

receiving an answer for the presented question;

identifying a symptom based on the answer; and

adding the identified symptom to the set of symptoms.

13. The computing device of claim 9, wherein execution of the set of instructions further configures the computing device to perform acts comprising:

eliminating the possible root cause associated with the presented explanation or solution from the set of possible root causes upon determining that a feedback for the presented explanation or solution indicates that the explanation or solution is inappropriate for the reported problem.

14. The computing device of claim 9, wherein execution of the set of instructions further configures the computing device to perform acts comprising updating the bipartite graph data structure based on a response to the presented question or the presented explanation received from the conversation.

15. A computer program product comprising:

one or more non-transitory computer-readable storage device and program instructions stored on at least one of the one or more non-transitory storage devices, the program instructions executable by a processor, the program instructions comprising sets of instructions for;

conducting a conversation over a user interface device;

receiving a set of symptoms of a reported problem from the conversation;

identifying (i) a set of possible root causes of the reported problem based on the set of symptoms and (ii) a probability for each possible root cause in the set of possible root causes by using a bipartite graph data structure that links possible root causes with possible symptoms;

upon determining that at least one possible root cause has a probability that is higher than a threshold, presenting, as part of the conversation, an explanation or solution associated with the at least one possible root cause having a probability that is higher than the threshold; and

upon determining that none of the possible root causes in the set of possible root causes has a probability higher than the threshold, presenting, as part of the conversation, a question based on an information entropy that is computed based on each of the probabilities of the identified possible root causes of the set of root causes.

16. The computer program product of claim 15, wherein the program instructions further comprising:

identifying one or more possible symptoms in addition to the set of symptoms for the set of possible root causes, wherein the presented question is selected from a set of candidate questions that are associated with the identified possible symptoms.

17. The computer program product of claim 16, wherein selecting the question comprises identifying a candidate question among the set of candidate questions that achieves a greatest reduction in the information entropy,

wherein the information entropy is computed based on the probabilities of the different possible root causes,

wherein the reduction in the information entropy is computed based on probabilities of (i) different possible symptoms associated with the candidate question and (ii) links between the different possible symptoms associated with the candidate question and the possible root causes.

18. The computer program product of claim 15, wherein the program instructions further comprising:

receiving an answer for the presented question;

identifying a symptom based on the answer; and

adding the identified symptom to the set of symptoms.

19. The computer program product of claim 15, wherein the program instructions further comprising eliminating the possible root cause associated with the presented explanation or solution from the set of possible root causes upon determining that a feedback for the presented explanation or solution indicates that the explanation or solution is inappropriate for the reported problem.

20. The computer program product of claim 15, wherein the program instructions further comprising updating the bipartite graph data structure based on a response to the presented question or the presented explanation received from the conversation.