Abstractive Text Summarization for Bangla Language Using NLP and Machine Learning Approaches

TL;DR

The study tackles Bengali abstractive text summarization in the face of scarce language-specific resources. It proposes a seq2seq LSTM encoder-decoder with attention and builds a large Bengali news dataset (19,096 articles with summaries) to enable evaluation. Results indicate the model generates coherent, abstractive summaries that are more natural than baselines, but struggle with long inputs, leading to repetition and potential factual drift. The work provides a public dataset and a neural attention-based framework, outlining future directions such as hierarchical encoders and pointer-generator approaches to improve robustness and multi-document summarization.

Abstract

Text summarization involves reducing extensive documents to short sentences that encapsulate the essential ideas. The goal is to create a summary that effectively conveys the main points of the original text. We spend a significant amount of time each day reading the newspaper to stay informed about current events both domestically and internationally. While reading newspapers enriches our knowledge, we sometimes come across unnecessary content that isn't particularly relevant to our lives. In this paper, we introduce a neural network model designed to summarize Bangla text into concise and straightforward paragraphs, aiming for greater stability and efficiency.

Figures (3)

• Figure 1: Dataset Overview
• Figure 2: Design of the LSTM-based encoder-decoder model for summarizing Bengali text.
• Figure 3: Visual comparison of the generated summary with the predicted summary.