International Journal of Advanced Research in Computer and Communication Engineering

A monthly peer-reviewed online and print journal

ISSN Online 2278-1021
ISSN Print 2319-5940

Since 2012

Abstract: In social media platforms like Twitter, You Tube etc. generates huge amount of user contents daily. In order to detect the user behaviour and interests keyphrases plays a crucial role. Keyphrases are short text pieces that can quickly express the key idea of source post. In case of extracting the main points from articles or documents the keyphrase generation is also important. Here proposes a methodology by generating keyphrases from the users post with the help of neural network representations and also generates the missing keyphrases which is the drawback of the previous systems. That is key phrase generation aims at predicting both present and absent keyphrases for user’s posts. The proposed method is a sequence to-sequence (seq2seq) based neural keyphrase generation frame work. Also, this model is topic- aware for avoiding sparsity in social media languages. Here also discussing about key phrase generation using BERT which is a latest technology in today’s world.

Keywords: Keyphrase, Seq2seq, Bert, Topic-Aware.

PDF | DOI: 10.17148/IJARCCE.2020.9643

Open chat
Chat with IJARCCE