Similarity-based retrieval of software investigation log sets for accelerated software deployment
A method in one embodiment comprises extracting features from each of a plurality of software investigation log sets, generating representations for respective ones of the software investigation log sets based at least in part on the corresponding extracted features, and storing the representations in a knowledge base. In conjunction with obtaining at least one additional software investigation log set, the method generates a representation of the additional software investigation log set, identifies one or more of the representations previously stored in the knowledge base that exhibit at least a specified similarity to the representation of the additional software investigation log set in accordance with one or more statistical models, and presents information characterizing the one or more software investigation log sets corresponding to respective ones of the identified one or more representations in a user interface. The method is illustratively implemented in a machine learning system of a processing platform.
1. An apparatus comprising:
a processing platform configured to implement a machine learning system for automated similarity-based retrieval of software investigation log sets;
wherein the machine learning system comprises:
a log set preprocessor configured to extract features from each of the software investigation log sets and to generate representations for respective ones of the software investigation log sets based at least in part on the corresponding extracted features;
a knowledge base configured to store the representations;
a similarity-based log set retrieval module comprising one or more statistical models each characterizing at least a portion of the knowledge base; and
a visualization module comprising at least one view generator;
a given one of the software investigation log sets characterizing multiple events relating to a particular piece of software and being generated by a software tracking system that is configured to track activities associated with deployment within at least one enterprise of multiple distinct pieces of software including the particular piece of software;
the software tracking system being further configured to provide metadata that is incorporated into the representations of the respective software investigation log sets;
wherein the log set preprocessor generates the representation of the given one of the software investigation log sets as a vector representation having entries corresponding to respective ones of the extracted features;
wherein the log set preprocessor in generating the representation of the given one of the software investigation log sets is further configured to augment the representation utilizing at least a portion of the metadata obtained from the software tracking system;
said at least a portion of the metadata obtained from the software tracking system and utilized to augment the given representation comprising root cause information characterizing the given software investigation log set;
wherein in conjunction with obtaining at least one additional software investigation log set, the machine learning system is configured to generate a representation of the additional software investigation log set using the log set preprocessor, and to identify one or more of the representations previously stored in the knowledge base that are determined by the similarity-based log set retrieval module to exhibit at least a specified similarity to the representation of the additional software investigation log set in accordance with at least one of the one or more statistical models;
wherein information characterizing the one or more software investigation log sets corresponding to respective ones of the identified one or more representations is presented in a user interface under control of the view generator of the visualization module; and
wherein the processing platform comprises one or more processing devices each comprising a processor coupled to a memory.
2. The apparatus of claim 1 wherein each of the software investigation log sets is generated by the software tracking system.
3. The apparatus of claim 1 wherein at least a portion of the machine learning system is implemented within the software tracking system.
4. The apparatus of claim 1 wherein the given one of the software investigation log sets comprises serial log instances relating to at least one test performed on the particular piece of software.
5. The apparatus of claim 1 wherein the given one of the software investigation log sets comprises a set of log files relating to a plurality of different events involving the particular piece of software.
6. The apparatus of claim 1 wherein the similarity-based log set retrieval module utilizes a k-nearest neighbors algorithm in order to determine the one or more of the representations previously stored in the knowledge base that exhibit at least the specified similarity to the representation of the additional software investigation log set.
7. The apparatus of claim 1 wherein the machine learning system further comprises a root cause prediction module that is configured to determine a root cause for the at least one additional software investigation log set based at least in part on root cause information associated with respective ones of the one or more of the representations previously stored in the knowledge base that exhibit at least the specified similarity to the representation of the additional software investigation log set.
8. The apparatus of claim 1 wherein the machine learning system is further configured to store the representation of the additional software investigation log set in the knowledge base for use in processing other software investigation log sets subsequently obtained by the machine learning system.
9. The apparatus of claim 1 wherein the one or more statistical models are updated responsive to the determination of the one or more of the representations previously stored in the knowledge base that exhibit at least the specified similarity to the representation of the additional software investigation log set.
10. The apparatus of claim 1 wherein the machine learning system is further configured to receive user feedback regarding at least one of the identified one or more representations via the user interface and to adjust at least one of the one or more statistical models responsive to the received user feedback.
11. An information processing system comprising the apparatus of claim 1 .
12. A method comprising:
extracting features from each of a plurality of software investigation log sets;
generating representations for respective ones of the software investigation log sets based at least in part on the corresponding extracted features;
storing the representations in a knowledge base;
a given one of the software investigation log sets characterizing multiple events relating to a particular piece of software and being generated by a software tracking system that is configured to track activities associated with deployment within at least one enterprise of multiple distinct pieces of software including the particular piece of software;
the software tracking system being further configured to provide metadata that is incorporated into the representations of the respective software investigation log sets;
wherein generating a given one of the representations further comprises generating the given representation as a vector representation having entries corresponding to respective ones of the extracted features;
wherein generating the given one of the representations further comprises augmenting the given representation utilizing at least a portion of the metadata obtained from the software tracking system;
said at least a portion of the metadata obtained from the software tracking system and utilized to augment the given representation comprising root cause information characterizing the given software investigation log set;
in conjunction with obtaining at least one additional software investigation log set, generating a representation of the additional software investigation log set, and identifying one or more of the representations previously stored in the knowledge base that exhibit at least a specified similarity to the representation of the additional software investigation log set in accordance with one or more statistical models; and
presenting information characterizing the one or more software investigation log sets corresponding to respective ones of the identified one or more representations in a user interface;
wherein the method is implemented by a processing platform comprising one or more processing devices.
13. The method of claim 12 further comprising determining a root cause for the at least one additional software investigation log set based at least in part on root cause information associated with respective ones of the one or more of the representations previously stored in the knowledge base that exhibit at least the specified similarity to the representation of the additional software investigation log set.
14. A computer program product comprising a non-transitory processor-readable storage medium having program code of one or more software programs embodied therein, wherein the program code when executed by at least one processing device of a processing platform causes the processing device:
to extract features from each of a plurality of software investigation log sets;
to generate representations for respective ones of the software investigation log sets based at least in part on the corresponding extracted features;
to store the representations in a knowledge base;
a given one of the software investigation log sets characterizing multiple events relating to a particular piece of software and being generated by a software tracking system that is configured to track activities associated with deployment within at least one enterprise of multiple distinct pieces of software including the particular piece of software;
the software tracking system being further configured to provide metadata that is incorporated into the representations of the respective software investigation log sets;
wherein generating a given one of the representations further comprises generating the given representation as a vector representation having entries corresponding to respective ones of the extracted features;
wherein generating the given one of the representations further comprises augmenting the given representation utilizing at least a portion of the metadata obtained from the software tracking system;
said at least a portion of the metadata obtained from the software tracking system and utilized to augment the given representation comprising root cause information characterizing the given software investigation log set;
in conjunction with obtaining at least one additional software investigation log set, to generate a representation of the additional software investigation log set, and to identify one or more of the representations previously stored in the knowledge base that exhibit at least a specified similarity to the representation of the additional software investigation log set in accordance with one or more statistical models; and
to present information characterizing the one or more software investigation log sets corresponding to respective ones of the identified one or more representations in a user interface.
15. The computer program product of claim 14 wherein the program code when executed by said at least one processing device further causes the processing device to determine a root cause for the at least one additional software investigation log set based at least in part on root cause information associated with respective ones of the one or more of the representations previously stored in the knowledge base that exhibit at least the specified similarity to the representation of the additional software investigation log set.
16. The computer program product of claim 14 wherein the program code when executed by said at least one processing device further causes the processing device to store the representation of the additional software investigation log set in the knowledge base for use in processing other subsequently obtained software investigation log sets.
17. The computer program product of claim 14 wherein the program code when executed by said at least one processing device further causes the processing device to update the one or more statistical models responsive to the determination of the one or more of the representations previously stored in the knowledge base that exhibit at least the specified similarity to the representation of the additional software investigation log set.
18. The computer program product of claim 14 wherein the program code when executed by said at least one processing device further causes the processing device to receive user feedback regarding at least one of the identified one or more representations and to adjust at least one of the one or more statistical models responsive to the received user feedback.
19. The computer program product of claim 14 wherein the program code when executed by said at least one processing device further causes the processing device to utilize a k-nearest neighbors algorithm in order to determine the one or more of the representations previously stored in the knowledge base that exhibit at least the specified similarity to the representation of the additional software investigation log set.
20. The computer program product of claim 14 wherein the given one of the software investigation log sets comprises a set of log files relating to a plurality of different events involving the particular piece of software.