NORMA eResearch @NCI Library

Online News Popularity Prediction using LSTM and Bi-LSTM

Rudrappa Shivu, Prasad (2021) Online News Popularity Prediction using LSTM and Bi-LSTM. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
Preview
PDF (Master of Science)
Download (1MB) | Preview
[thumbnail of Configuration manual]
Preview
PDF (Configuration manual)
Download (1MB) | Preview

Abstract

Online news has become an apparently essential source of information, capturing the attention of a considerable audience. People are reading and posting news more often online, on websites, blogs, and social media sites such as Facebook and Twitter. The rapid advancement of smart phones, as well as the introduction of other electronic devices, has made access to news far simpler than ever before. Every day, millions of news stories are published online. The writings are shortlived; some fade away quickly, some acquire traction as their reach grows, and others reclaim lost attention thanks to follow-up news or related articles. Predicting their popularity would benefit not just news stakeholders, but also the general public, who would be kept up to speed with the newest information. Several studies have been conducted over the last decade to forecast the popularity of news using various machine learning algorithms. While some researchers utilized factors such as the number of times the news was shared on social media, the amount of visitor comments, and the overall number of likes the article received to forecast popularity, others relied on word count, title, and abstract length. To forecast popularity, a variety of regression and classification techniques were used. The models’ accuracy, on the other hand, appeared to be approximately 70%. This research uses LSTM and Bi-LSTM, two powerful unsupervised machine learning models, to forecast the popularity of news items before they are published, utilizing Feature Selection to increase the model’s accuracy. The accuracy of the model improved to 79% for BiLSTM with normally distributed features with high Fisher Score. Aforementioned model, showing improved accuracy, can be deployed to predict the popularity of news published across online news media and social media.

Item Type: Thesis (Masters)
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
Q Science > QA Mathematics > Computer software
T Technology > T Technology (General) > Information Technology > Computer software
Z Bibliography. Library Science. Information Resources > ZA Information resources > ZA4150 Computer Network Resources > The Internet > World Wide Web > Websites
T Technology > TK Electrical engineering. Electronics. Nuclear engineering > Telecommunications > The Internet > World Wide Web > Websites
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Clara Chan
Date Deposited: 14 Dec 2021 12:49
Last Modified: 14 Dec 2021 12:49
URI: https://norma.ncirl.ie/id/eprint/5218

Actions (login required)

View Item View Item