Method, apparatus and computer storage medium for error diagnostics of an application

1. A method of application error diagnosis, comprising:

obtaining information related to an error in an application;

generating a code of the information related to the error in the application, wherein generating the code comprises vectorizing the information to obtain the code;

determining a similarity between the code and at least one predetermined code, wherein the at least one predetermined code is associated with a known software bug; and

determining at least one software bug related to the error in the application by comparing the similarity with a first threshold.

2. The method according to claim 1 , wherein vectorizing the information comprises:

vectorizing words in the information to obtain word vectors; and

obtaining a vector of a sentence in the information based on the word vectors.

3. The method according to claim 1 , wherein generating the code further comprises:

deleting a pause word from the information; and

generating the code based on the information with the pause word deleted.

4. The method according to claim 1 , wherein vectorizing the information is based on at least one selected from a group consisting of a term frequency-inverse document frequency (TF-IDF) and a Word2vec model.

5. The method according to claim 1 , wherein the information related to the error in the application comprises environmental information of the application and at least one of the following:

a critical error log segment from a customer environment,

bug information from a bug tracking system, and

solution information from at least one selected from a group consisting of support page and a support server.

6. The method according to claim 1 , wherein the at least one predetermined code is from a pre-stored corpus.

7. The method according to claim 1 , wherein determining the similarity between the code and the at least one predetermined code comprises:

determining the similarity based on at least one selected from a group consisting of an Euclidean distance and a cosine function between the code and the at least one predetermined code.

8. The method according to claim 1 , further comprising:

obtaining information related to the known software bug; and

updating the at least one predetermined code based on the information related to the known software bug for subsequent error diagnosis.

9. The method according to claim 1 , further comprising:

obtaining a description of the at least one software bug;

determining a description similarity between the description and a description of at least one predetermined solution; and

determining, based on a comparison of the description similarity with a second threshold, at least one solution for solving the error in the application.

10. The method according to claim 9 , wherein the at least one predetermined solution is obtained from a support webpage or a support server.

11. The method according to claim 9 , wherein determining the description similarity comprises:

obtaining a vectorized representation of the description of the at least one software bug; and

determining the description similarity by comparing the vectorized representation with a description vector of the at least one predetermined solution.

12. The method according to claim 11 , wherein determining the description similarity by comparing the vectorized representation and the description vector of the at least one predetermined solution is based on one selected from a group consisting of an Euclidean distance and a cosine function between the vectorized representation and the description vector of the at least one predetermined solution.

13. The method according to claim 9 , wherein the description comprises a title of the at least one software bug.

14. The method according to claim 9 , further comprising:

obtaining an available solution related to the application; and

updating, based on the available solution, the at least one predetermined solution for subsequent error diagnosis.

15. An apparatus for application error diagnosis, comprising:

a data collector configured to obtain information related to an error in an application;

an information encoder configured to generate a code of the information related to the error in the application, wherein generating the code of the information comprises vectorizing the information to obtain the code;

an analyzer configured to determine a similarity between the code and at least one predetermined code, wherein the at least one predetermined code is associated with a known software bug; and

an error diagnostor configured to determine at least one software bug related to the error in the application by comparing the similarity and a first threshold.

16. The apparatus according to claim 15 , wherein the information encoder is further configured to:

vectorize words in the information to obtain word vectors; and

obtain a vector of a sentence in the information based on the word vectors.

17. The apparatus according to claim 15 , wherein the information encoder is configured to:

delete a pause word from the information; and

generate the code based on the information with the pause word deleted.

18. A computer program product tangibly stored on a non-transient computer readable medium and including machine executable instructions which, when executed, cause a machine to perform a method, the method comprising:

obtaining information related to an error in an application;

generating a code of the information related to the error in the application, wherein generating the code comprises vectorizing the information to obtain the code;

determining a similarity between the code and at least one predetermined code, wherein the at least one predetermined code is associated with a known software bug; and

determining at least one software bug related to the error in the application by comparing the similarity with a first threshold.