Microservice deployment in multi-tenant environments
A method includes providing a server executing computer code embodying the chat-bot, receiving, via the server, a requirement for customization of a main computer code, the main computer code including modules, and determining, by the chat-bot, top-k modules of the main computer code potentially affected given the requirement for customization. The determination of the top-k modules includes representing each of the modules of the main computer code in an embedding vector space, representing the requirement for customization as a vector in the embedding space, and fetching the top-k modules nearest the requirement for customization in the embedding space. The method further includes determining a question, communicating the question to an entity providing the requirement for customization, receiving a response to the question, and filtering the top-k modules using the response to determine a set of the modules associated with the requirement for customization.
1. A method of operating an artificial intelligence chat-bot, the method comprising:
providing a server executing computer code embodying the chat-bot;
receiving, via the server, a requirement for customization of a main computer code, wherein the main computer code comprises a plurality of modules;
determining, by the chat-bot, top-k modules of the plurality of modules of the main computer code potentially affected given the requirement for customization, wherein the determination of the top-k modules comprises:
representing each module of the plurality of modules of the main computer code in an embedding vector space;
representing the requirement for customization as a vector in the embedding space; and
fetching the top-k modules nearest the requirement for customization in the embedding space;
applying an entropy analysis across the top-k modules to determine a distinguishing next question;
communicating the distinguishing next question to an entity providing the requirement for customization;
receiving a response to the distinguishing next question; and
filtering the top-k modules using the response to determine a set of the plurality of modules associated with the requirement for customization.
2. The method of claim 1 , wherein the embedding space is formed using code comments contained in the plurality of modules.
3. The method of claim 2 , wherein the distinguishing next question is configured to request an information item resolving a conflict in the code comments.
4. The method of claim 1 , further comprising:
performing, given the requirement for customization and a log comprising the distinguishing next question and the response, an entropy analysis of code documentation associated with the main computer code to determine a plurality of attributes; and
enriching the code documentation using the attributes.
5. The method of claim 1 , further comprising:
performing, given the requirement for customization and a log comprising the distinguishing next question and the response, an entropy analysis of code documentation associated with the main computer code to determine a plurality of attributes;
determining at least one ambiguity in the code documentation; and
outputting a communication including an indication of the at least ambiguity.
6. The method of claim 1 , further comprising:
extracting first topics from code comments contained in the module using natural language processing;
extracting second topics from the requirement for customization using natural language processing; and
generating a set of features by combing the first topics and the second comments, which describes a potential modification of the main code enabling requirement for customization.
7. The method of claim 1 , given the requirement for customization and a log comprising the distinguishing next question and the response, further comprising:
mapping a conversation block of the log to a portion of the requirement for customization;
identifying a code snippet of at least one module of the plurality of modules of the main code, the code snippet corresponding to the portion of requirement for customization; and
generating a set of hyperlinks mapping the conversation block with the code snippet.
8. A non-transitory computer readable storage medium comprising computer executable instructions which when executed by a computer cause the computer to perform a method of operating an artificial intelligence chat-bot, the method comprising:
providing a server executing the computer executable instructions embodying the chat-bot;
receiving, via the server, a requirement for customization of a main computer code, wherein the main computer code comprises a plurality of modules;
determining, by the chat-bot, top-k modules of the plurality of modules of the main computer code potentially affected given the requirement for customization, wherein the determination of the top-k modules comprises:
representing each module of the plurality of modules of the main computer code in an embedding vector space;
representing the requirement for customization as a vector in the embedding space; and
fetching the top-k modules nearest the requirement for customization in the embedding space;
applying an entropy analysis across the top-k modules to determine a distinguishing next question;
communicating the distinguishing next question to an entity providing the requirement for customization;
receiving a response to the distinguishing next question; and
filtering the top-k modules using the response to determine a set of the plurality of modules associated with the requirement for customization.
9. The non-transitory computer readable storage medium of claim 8 , wherein the embedding space is formed using code comments contained in the plurality of modules.
10. The non-transitory computer readable storage medium of claim 9 , wherein the distinguishing next question is configured to request an information item resolving a conflict in the code comments.
11. The non-transitory computer readable storage medium of claim 8 , further comprising:
performing, given the requirement for customization and a log comprising the distinguishing next question and the response, an entropy analysis of code documentation associated with the main computer code to determine a plurality of attributes; and
enriching the code documentation using the attributes.
12. The non-transitory computer readable storage medium of claim 8 , further comprising:
performing, given the requirement for customization and a log comprising the distinguishing next question and the response, an entropy analysis of code documentation associated with the main computer code to determine a plurality of attributes;
determining at least one ambiguity in the code documentation; and
outputting a communication including an indication of the at least ambiguity.
13. The non-transitory computer readable storage medium of claim 8 , further comprising:
extracting first topics from code comments contained in the module using natural language processing;
extracting second topics from the requirement for customization using natural language processing; and
generating a set of features by combing the first topics and the second comments, which describes a potential modification of the main code enabling the requirement for customization.
14. The non-transitory computer readable storage medium of claim 13 , given the requirement for customization and a log comprising the distinguishing next question and the response, further comprising:
mapping a conversation block of the log to a portion of the requirement for customization;
identifying a code snippet of at least one module of the plurality of modules of the main code, the code snippet corresponding to the portion of the requirement for customization; and
generating a set of hyperlinks mapping the conversation block with the code snippet.