My company has been using Jira to track production over the past 6-8 years, and as a result , our Jira has a wealth of data on performance issues.
Usually every Jira ticket for any production support problems consists of some useful information, such as:
- Error message
- Adopted system
- Root cause
- Resolution
- Time taken
- etc.
My company has its own chat service that supports the Chatbot API in Java / Python / etc. I would like to create a smart chatbot (if not AI) that is smart enough to exchange chat conversations:
DevOps) Hi Jirabot, what do you know about this error message? [xyzvxc exception occurred on line 82 ....]
Jirabot) Hi, on what systems did this happen? You can choose one of the following options:
DevOps) 1
Jirabot) Correctly, it seems that the following Jira tickets have experienced similar problems. Please check the following tickets.
- Jira-12zx
- Jira-52123zz
- Jira-vvvbbb
I would like to ask people with experience in implementing something similar or have any relevant experience in ML / Neural Network / Natural Language Processing the following questions.
1. Is this possible for a non-NPL expert? Do you think it is possible to build something like this for a software engineer with 8 years of experience working with advanced beginner skills in cars? (Advanced Beginner in the Dreyfus Skill Model )
, . 8 , Andrew Ng . ( ), Deep Learning Udacity , Tensorflow.
2. / ?
, , / ? - - , ?
3. , ?
, , , ?
, ,
J