Optimisation-Enabled Deep Learning for Fake Word Identification and Conversion of Actual Reviews from Fake Reviews

Steni Mol T S and S. Gokila
Additional contact information
Steni Mol T S: Department of Computer Applications, Hindustan Institute of Technology & Science, Kelambakkam, Chennai 603103, Tamil Nadu, India
S. Gokila: Department of Computer Applications, Hindustan Institute of Technology & Science, Kelambakkam, Chennai 603103, Tamil Nadu, India

Journal of Information & Knowledge Management (JIKM), 2025, vol. 24, issue 05, 1-33

Abstract: In social media, fake news refers to false information or gross falsification of an actual incident. Nevertheless, it is challenging for average people to distinguish fake news from the large amount of information available online, because of the limits on their knowledge, time or space. As a result, it is essential to create automated and auxiliary techniques for early fake news detection. Hence, Fractional Golden Search Dingo Optimisation_Transfer Learning (FGSDO_TL) is proposed. First, the input data is obtained from the review database, and then review vectorisation is carried out by Word2Vec and Bidirectional Encoder Representations from Transformers (BERT) for converting the text to numerical vectors. Thereafter, fake news detection is done by applying TL using the Convolutional Neural Network (CNN)-based Long Short-Term Memory (LSTM), which is trained by Fractional Golden Search Dingo Optimisation (FGSDO). Here, FGSDO is developed by integrating the Golden Search Dingo Optimisation (GSDO) with Fractional Calculus (FC). Moreover, the fake word is recognised by identifying the word impact first and then using Deep LSTM (DLSTM). Later, the regeneration of the actual review from the fake review is achieved by using Simple Natural Language Generation (SimpleNLG). The proposed method has achieved 94.36% of sensitivity, 93.94% of specificity and 93.86% of accuracy.

Keywords: Fake word; Word2Vec; transfer learning; Dingo optimiser; Fractional Calculus (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.worldscientific.com/doi/abs/10.1142/S0219649225500418
Access to full text is restricted to subscribers

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:wsi:jikmxx:v:24:y:2025:i:05:n:s0219649225500418

Ordering information: This journal article can be ordered from

DOI: 10.1142/S0219649225500418

Access Statistics for this article

Journal of Information & Knowledge Management (JIKM) is currently edited by Professor Suliman Hawamdeh

More articles in Journal of Information & Knowledge Management (JIKM) from World Scientific Publishing Co. Pte. Ltd.
Bibliographic data for series maintained by Tai Tone Lim ().