Automatic Detection of Pornographic and Gambling Websites Based on Visual and Textual Content Using a Decision Mechanism

Abstract

Pornographic and gambling websites become increasingly stubborn via disguising, misleading, blocking, and bypassing, which hinder the construction of a safe and healthy network environment. However, most traditional approaches conduct the detection process through a single aspect of these sites, which would fail to handle the more intricate and challenging situations. To alleviate this problem, this study proposed an automatic detection system for porn and gambling websites based on visual and textual content using a decision mechanism (PG-VTDM). This system can be applied to the intelligent wireless router at home or school to realize the identification, blocking, and warning of ill-suited websites. First, Doc2Vec was employed to learn the textual features that can be used to represent the textual content in the hypertext markup language (HTML) source code of the websites. In addition, the traditional bag-of-visual-words (BoVW) was improved by introducing local spatial relationships of feature points for better representing the visual features of the website screenshot. Then, based on these two types of features, a text classifier and an image classifier were both trained. In the decision mechanism, a data fusion algorithm based on logistic regression (LR) was designed to obtain the final prediction result by measuring the contribution of the two classification results to the final category prediction. The efficiency of this proposed approach was substantiated via comparison experiments using gambling and porn website datasets crawled from the Internet. The proposed approach outperformed the approach based on a single feature and some state-of-the-art approaches, with accuracy, precision, and F-measure all over 99%.

Identifying Gambling Websites with Co-training

“…Zhang et al [8] propose a two-stage extreme learning machine for phishing website detection based on the mixed features of URL, web, and text content. Chen et al [9] extract features from images and text in the webpages to detect gambling and porn websites.…”

“…The existing malicious website identification methods could be classified into black-list based, URL based [1,2,3], webpage content based [4,5,6,7] and mixed-features based [8,9]. Black-list based methods establish a black list by collecting the malicious URLs or domain names.…”

Birds of a Feather Flock Together: Generating Pornographic and Gambling Domain Names Based on Character Composition Similarity

“…The majority of traditional approaches focus on a particular feature of these pornographic and gambling websites, which leaves out more nuanced and problematic scenarios. Chen et al [16] developed an automatic detection system for pornographic and gambling websites based on visual and textual content using a decision process to address this issue. Similarly, Zhao et al [17] proposed Porn2Vec, a robust end-to-end framework for detecting pornographic websites using contrastive learning.…”

Transfer Learning on Deep Neural Networks to Detect Pornography

“…In order to design their model, Li and colleagues combined frame sampling, which resulted in a higher accuracy level when compared to the naïve form of deep learning model. Chen et al [22] proposed one of the most successful pornographic detection systems to date. Designed to detect pornographic related content on both porn and gambling websites from common wireless routers often used in schools and homes, their decision mechanism yielded measures over 0.99 for accuracy, precision, and F-measure.…”