Semantic Search On Stack Overflow
Semantic Search On Stack Overflow
Stack Overflow is a question and answer site for professional and enthusiast programmers.
It's built and run by the community of developer, as part of the
Stack Exchange network of Q&A sites. A lot of content is present in form of stack overflow
questions and answers, various studies point that developers face problems while
development life cycles and they ask questions on stack overflow which gets answered by
fellow developers across the globe.
In order for a new developer to understand a concept or solve an issue, it is very difficult to
identify the problems. It involves domain experts in form of experienced software
developers. The information present is overwhelming and at times can be too much to
handle for a budding developer.
Problem Description
1. To identify most relevant questions to a query [text similarity]
2. Identify the matching tags and pick top relevant questions from stack overflow.
3. To identify top k solutions of the problem.
(sentiment analysis of review content)
Expectation
The solution should have a mechanism to identify relevant questions on stack overflow with
tags and top k relevant answers that should be provided to user. So that he can understand
the concept clearly and try to get his queries resolved more effectively and efficiently.
Examples:
Query on stack overflow search:
1. "Node - how to run node app.js?" - No results.
2. "Node - how to run app.js?" – fetches a result.
https://stackoverflow.com/search?q=%22Node++how+to+run+++app.js%3F%22
Evaluation Criteria
Accuracy of the answers obtained by the solution will be the mainstay of evaluation. The
other evaluation parameters are listed on the hackathon landing page.