Machine learned decision guidance for alerts originating from monitoring systems

1. A method for automatically determining whether a detected issue in a computing system is a bug, comprising:

receiving information about a detected issue;

storing the information about the detected issue in a storage device;

generating a feature vector for the detected issue;

analyzing the feature vector to determine a probability that the detected issue is a bug caused by an alteration of at least a portion of source code used by the computing system, wherein the analysis is based, at least in part, on the feature vector;

when the probability is above a threshold, reporting the detected issue to a client device;

receiving input corresponding to the detected issue, the input indicating whether the detected issue is a bug;

using the input to update the information about the detected issue in the storage device; and

providing the updated information to a system that identified the detected issue to automatically train the system.

2. The method of claim 1 , wherein the input comprises a solution to the detected issue.

3. The method of claim 2 , further comprising updating the analysis based, at least in part, on the received solution to the detected issue.

4. The method of claim 2 , further comprising:

associating the solution with the detected issue; and

storing the solution in the storage device.

5. The method of claim 1 , further comprising prioritizing the detected issue with respect to one or more additional detected issues.

6. The method of claim 1 , wherein generating a feature vector comprises performing one or more transformations on the information about the detected issue.

7. The method of claim 6 , wherein the one or more transformations are selected from a group comprising: normalization, counting occurrences of key words from a stack trace, and binning.

8. A system, comprising:

at least one processing unit; and

a memory storing computer executable instructions that, when executed by the at least one processing unit, cause the system to perform a method for detecting occurrences of issues in a software system, comprising:

receiving information about a detected issue;

comparing a fingerprint of the detected issue to one or more fingerprints of previously detected issues to determine whether the detected issue is new;

when it is determined that the detected issue is new:

storing the information about the detected issue in a storage device;

generating a feature vector for the detected issue;

analyzing the feature vector to determine a probability that the detected issue is a bug caused by an alteration of at least a portion of source code used by the system;

providing an ordered list of detected issues to a client device;

receiving input corresponding one or more of the detected issues in the ordered list of detected issues, the input indicating whether the detected issue is a bug or a false positive; and

using the input to update the information about the detected issue.

9. The system of claim 8 , further comprising instructions for using one or more models to evaluate a received feature vector based, at least in part, on information contained in the feature vector.

10. The system of claim 8 , further comprising instructions for receiving a solution to the detected issue.

11. The system of claim 10 , further comprising instructions for associating the solution with the detected issue.

12. The system of claim 10 , further comprising instructions for updating one or more models based, at least in part, on the solution.

13. The system of claim 8 , wherein an order of the ordered list is based on the determined probability.

14. The system of claim 8 , further comprising instructions for assigning a priority to the detected issue in view of previously detected issues.

15. The system of claim 8 , further comprising instructions for storing information about the detected issue in the storage device when it is determined that the detected issue is not a bug.

16. A computer-readable storage medium storing computer executable instructions that, when executed by a processing unit, causes the processing unit to perform a method, comprising:

generating a feature vector for a detected issue using information associated with the detected issue;

analyzing the feature vector using one or more models to determine a probability that the detected issue is a bug caused by an alteration of at least a portion of source code;

reporting the occurrence of the detected issue;

receiving input that indicates whether the detected issue is a false positive or a bug;

receiving a solution to the detected issue when the input indicates the detected issue is a bug; and

updating at least one of the one or more models using the solution to the detected issue.

17. The computer-readable storage medium of claim 16 , further comprising instructions for storing the solution in a storage device.

18. The computer-readable storage medium of claim 16 , further comprising instructions for associating the solution with the detected issue.

19. The computer-readable storage medium of claim 16 , further comprising instructions for storing the information about the detected issue.

20. The computer-readable storage medium of claim 16 , further comprising instructions for determining which model of the one or more models is configured to analyze the feature vector.