Quora, a well known social media platform, allows people from all around the world to ask questions and enables them to connect with various domain experts who provide quality answers and several new insights. With over 100 million people visiting Quora every month, the user base is humungous. It is not difficult to inference that several people may end us asking similarly worded questions. When there are multiple questions being asked with the same intent, a reader would have to spend more time in order to find the most suitable answer to their question. At the same time a writer may feel that they are answering multiple versions of the same question which may seem like a waste of time. This in turn leads us to our problem. Given any two questions, identify whether they have the same intent i.e. is one question a duplicate of the other or not. We will be using the dataset provided by Quora for the same. However, we aim to generalise our approach to be able to tackle the problem for any Question - Answer Forum, such as Quora itself or Stack Overflow.
Collaborators - Rahul Ragesh, N.S. Kumar
Please find all the relevant resources below: