Fake News Detection Using Long Short-Term Memory Networks and Natural Language Processing: A Deep Learning Approach with GloVe Word Embeddings
Interdisciplinary Journal of AI, Machine Learning & Data Science · 08.06.2026 · 1 (2) · e015
DOI: 10.66261/sgh6wd89 ↗Kurz gefasst
Fachartikel in der Zeitschrift Interdisciplinary Journal of AI, Machine Learning & Data Science, online erschienen am 08.06.2026. Verfasst von Gourav Yadav und 1 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
The exponential growth of fake news across digital media platforms poses a significant threat to public trust, democratic processes, and social stability. This paper presents the design, implementation, and experimental evaluation of an automated fake news detection system using Long Short-Term Memory (LSTM) networks combined with Natural Language Processing (NLP) pre-processing and GloVe pre-trained word embeddings. The proposed system is trained and evaluated on the Kaggle Fake News Dataset comprising 44,898 labeled English-language news articles. A comprehensive NLP pre-processing pipeline — including text cleaning, tokenization, stop-word removal, and lemmatization — transforms raw article text into structured input sequences. GloVe embeddings (100-dimensional) initialize the model's embedding layer, providing rich semantic representations with 86.4% vocabulary coverage. The LSTM-based architecture, incorporating spatial dropout and recurrent dropout regularization, achieves a test accuracy of 96.34% and a macro F1-score of 96.34%, significantly outperforming four traditional machine learning baselines: Naive Bayes (84.67%), Random Forest (89.76%), Logistic Regression (91.23%), and Support Vector Machine (93.48%). An ablation study quantifies the individual contribution of each system component. Error analysis identifies systematic misclassification patterns. The paper discusses ethical considerations, practical deployment implications, and directions for future research including transformer-based architectures, multilingual detection, and explainable AI integration.
Zitieren
APA
Yadav, G., & Moghe, D. P. (2026). Fake News Detection Using Long Short-Term Memory Networks and Natural Language Processing: A Deep Learning Approach with GloVe Word Embeddings. Interdisciplinary Journal of AI, Machine Learning & Data Science, 1(2), e015. https://doi.org/10.66261/sgh6wd89
Vancouver
Yadav G, Moghe DP. Fake News Detection Using Long Short-Term Memory Networks and Natural Language Processing: A Deep Learning Approach with GloVe Word Embeddings. Interdisciplinary Journal of AI, Machine Learning & Data Science. 2026;1(2):e015. doi:10.66261/sgh6wd89
BibTeX
@article{Yadav2026,
author = {Yadav, Gourav and Moghe, Dr. Piyush},
title = {Fake News Detection Using Long Short-Term Memory Networks and Natural Language Processing: A Deep Learning Approach with GloVe Word Embeddings},
journal = {Interdisciplinary Journal of AI, Machine Learning & Data Science},
year = {2026},
volume = {1},
number = {2},
pages = {e015},
doi = {10.66261/sgh6wd89}
}
EUVIMED