Question Similarity Detection Using Bidirectional LSTM through Natural Language Processing

Identifying the duplicate questions is challenging; the sentence composition and word selection vary among persons. There is a way to find the document similarity based on the degree of overlapping in classic natural language processing. However, duplicate question detection does not perform well, because most of the questions are too small and the degree of overlapping is insufficient. We have implemented an LSTM based architecture that detects the duplicate question if they have the same intent. [Poster]