LookAhead: Augmenting Crowdsourced Website Reputation Systems With Predictive Modeling

TL;DR

This paper introduces LookAhead, a predictive modeling system that uses website structural and content features to efficiently estimate safety ratings, reducing reliance on manual assessments and addressing scalability issues.

Contribution

It presents a novel ensemble classification approach that accurately predicts trustworthiness and child safety ratings for websites using structural and content data.

Findings

Achieved average F1-score of 74-90% across datasets.

Effectively predicts subjective safety ratings and maliciousness.

Reduces time lag and increases scalability of website reputation systems.

Abstract

Unsafe websites consist of malicious as well as inappropriate sites, such as those hosting questionable or offensive content. Website reputation systems are intended to help ordinary users steer away from these unsafe sites. However, the process of assigning safety ratings for websites typically involves humans. Consequently it is time consuming, costly and not scalable. This has resulted in two major problems: (i) a significant proportion of the web space remains unrated and (ii) there is an unacceptable time lag before new websites are rated. In this paper, we show that by leveraging structural and content-based properties of websites, it is possible to reliably and efficiently predict their safety ratings, thereby mitigating both problems. We demonstrate the effectiveness of our approach using four datasets of up to 90,000 websites. We use ratings from Web of Trust (WOT), a popular crowdsourced web reputation system, as ground truth. We propose a novel ensemble classification technique that makes opportunistic use of available structural and content properties of webpages to predict their eventual ratings in two dimensions used by WOT: trustworthiness and child safety. Ours is the first classification system to predict such subjective ratings and the same approach works equally well in identifying malicious websites. Across all datasets, our classification performs well with average F$_1$-score in the 74--90\% range.