Detecting Semantically Similar Questions on Quora Dataset

Details

This project was done as part of the course CS671A: Introduction to Natural Language Processing, in the Spring ‘18 term at IIT Kanpur under Prof. Harish Karnick, Department of Computer Science and Engineering, IIT Kanpur.

Brief summary of the work done in the project

Websites like quora provide a platform for Question and Answering across various subjects and fields. However, the websites also contain loads of similar questions which have different sets of replies although they seek same answer. We aim to develop an application which can detect whether two questions on quora are semantically similar. More formally we want a model which when given two questions, gives

• 0 if both questions cannot be answered by the same answer
• 1 if both question can be answered by the same answer

depending on whether the questions are similar in terms of meaning or not. The criteria for two questions being semantically similar is that they seek same answer.