Neural Abstractive Text Summarization and Fake News Detection

TL;DR

This paper compares various neural models for abstractive text summarization and explores their use as feature extractors for fake news detection, demonstrating the effectiveness of summarization in improving classification accuracy.

Contribution

It introduces a comprehensive comparison of neural summarization models and extends their application to fake news detection as a novel feature extraction method.

Findings

Transformer-based models outperform LSTM models in summarization quality.

Summarization as a feature extractor improves fake news classification accuracy.

Extensive hyperparameter tuning enhances model performance.

Abstract

In this work, we study abstractive text summarization by exploring different models such as LSTM-encoder-decoder with attention, pointer-generator networks, coverage mechanisms, and transformers. Upon extensive and careful hyperparameter tuning we compare the proposed architectures against each other for the abstractive text summarization task. Finally, as an extension of our work, we apply our text summarization model as a feature extractor for a fake news detection task where the news articles prior to classification will be summarized and the results are compared against the classification using only the original news text. keywords: LSTM, encoder-deconder, abstractive text summarization, pointer-generator, coverage mechanism, transformers, fake news detection