An enhanced deep learning model for duplicate question pairs recognition

Abstract

Quora is a question-and-answer site, where questions are asked, answered, edited, and organized by its community of users. It is a platform to ask questions and connect with people who contribute unique insights and quality answers. But, huge number of questions ends up being duplicates of other questions asked by different user with various words or expressions. Preferably, these duplicate inquiries would be consolidated into a solitary accepted query, as doing as such would give various advantages. Quora uses a random forest model for duplicate question pair’s identification. This method is not efficient for a large number of questions. The proposed model uses GloVe embedding which consists of 100d Wikipedia vectors and improves the prediction accuracy.