An Attention-Based Neural Network Using Human Semantic Knowledge and Its Application to Clickbait Detection

Abstract

Clickbait is a commonly used social engineering technique to carry out phishing attacks, illegitimate marketing, and dissemination of disinformation. As a result, clickbait detection has become a popular research topic in recent years due to the prevalence of clickbait on the web and social media. In this article, we propose a novel attention-based neural network for the task of clickbait detection. To the best of our knowledge, our work is the first that incorporates human semantic knowledge into an artificial neural network, and uses linguistic knowledge graphs to guide attention mechanisms for the clickbait detection task. Extensive experimental results show that the proposed model outperforms existing state-of-the-art clickbait classifiers, even when training data is limited. The proposed model also performs better or comparably to powerful pre-trained models, namely, BERT, RoBERTa, and XLNet, while being much more lightweight. Furthermore, we conducted experiments to demonstrate that the use of human semantic knowledge can significantly enhance the performance of pre-trained models in the semi-supervised domain such as BERT, RoBERTa, and XLNet.

Multiview Clickbait Detection via Jointly Modeling Subjective and Objective Preference

“…KED (Wei and Nguyen, 2022): A method of integrating human semantic knowledge into a neural network model, by building a language knowledge graph to guide attention mechanisms for inference. Implementation Details: In the experiments, we used Bert to initialize the word embedding layer, each with a dimension of 768 dimensions and Longformer to initialize the article embedding.…”

“…The primary objective of clickbait detection is to extract crucial linguistic features from textual content. Most of the current research is focused on modeling article content and detecting the correlation between posts and articles (Cao et al, 2017;Wei and Wan, 2017;Zheng et al, 2017), with the use of deep learning algorithms (Dong et al, 2018;Anand et al, 2017;Wei and Nguyen, 2022) increasingly employed to enhance the accuracy and efficiency of clickbait detection. Chen et al proposed the use of text and stylistic patterns for detection.…”

“…considered the interaction between the context of posts and articles as well as the style patterns of posts. Wei et al (Wei and Nguyen, 2022) proposed for the first time to use human semantic knowledge in an attention model and used language knowledge graphs to guide the attention mechanism. However, these methods only focus on considering the content correlation and neglect emotional and stylistic features that are crucial for clickbait detection.…”

Цифрові Представлення Telegram-Каналів

“…Було отримано високу точність в оцінках на наявному датасеті. Подібним чином в роботі [8] Результатом є цифрове представлення, яке придатне для подальшої обробки та вирішення завдань із застосуванням машинного навчання для класифікації, фільтрації, агрегації, категоризації, рекомендації.…”

Fake News in Marketing

“…Clickbait is the deliberate use of deceiving headlines to encourage visitors to click on a webpage or video. These sensational headlines do not regard the truth and primarily focus on attracting clicks and views [39,40]. On the other hand, conspiracy theories are used to explain complex events or phenomena with little or no evidence [41].…”