Marathi Text Summarization for News Articles using Sequence To Sequence with Attention Mechanism

  • Kavya Nair, Kotian Snita, Aarati Lomte, Praharsha Gottapu, Rushali Deshmukh


         Text summarization is a common problem in machine learning as it involves developing algorithms and statistical models to create a coherent summary to convey the intended message in fewer words, keeping in mind the user's time-constrained environment. We have built an automatic text summarization model to produce compact and concise summaries while retaining the vital information from the news article using RNN’s encoder-decoder using the Sequence-to-Sequence model with an attention mechanism. This model ensures to summarize news articles in the Marathi language on which there have been fewer previous contributions in the field. We have created our dataset, which has proved to be extremely beneficial to validate our model.