On Obstructing Obscenity Obfuscation

Abstract

Obscenity (the use of rude words or offensive expressions) has spread from informal verbal conversations to digital media, becoming increasingly common on user-generated comments found in Web forums, newspaper user boards, social networks, blogs, and media-sharing sites. The basic obscenity-blocking mechanism is based on verbatim comparisons against a blacklist of banned vocabulary; however, creative users circumvent these filters by obfuscating obscenity with symbol substitutions or bogus segmentations that still visually preserve the original semantics, such as writing shit as $h¡;t or s.h.i.t or even worse mixing them as $.h….¡.t . The number of potential obfuscated variants is combinatorial, yielding the verbatim filter impractical. Here we describe a method intended to obstruct this anomaly inspired by sequence alignment algorithms used in genomics, coupled with a tailor-made edit penalty function. The method only requires to set up the vocabulary of plain obscenities; no further training is needed. Its complexity on screening a single obscenity is linear, both in runtime and memory, on the length of the user-generated text. We validated the method on three different experiments. The first one involves a new dataset that is also introduced in this article; it consists of a set of manually annotated real-life comments in Spanish, gathered from the news user boards of an online newspaper, containing this type of obfuscation. The second one is a publicly available dataset of comments in Portuguese from a sports Web site. In these experiments, at the obscenity level, we observed recall rates greater than 90%, whereas precision rates varied between 75% and 95%, depending on their sequence length (shorter lengths yielded a higher number of false alarms). On the other hand, at the comment level, we report recall of 86%, precision of 91%, and specificity of 98%. The last experiment revealed that the method is more effective in matching this type of obfuscation compared to the classical Levenshtein edit distance. We conclude discussing the prospects of the method to help enforcing moderation rules of obscenity expressions or as a preprocessing mechanism for sequence cleaning and/or feature extraction in more sophisticated text categorization techniques.

Automatic Identification of Harmful, Aggressive, Abusive, and Offensive Language on the Web: A Survey of Technical Biases Informed by Psychology Literature

“…The design of rules is often done in two steps. Dictionaries expressing OCL are prepared [11,224,296] (e.g., subjectivity lexicon, hate lexicon, and list of haterepresentative grammatical relations [83], lists of profane words augmented with genomicsinspired techniques [225]). Then, lists of rules are defined to attribute a score to samples based on their use of the dictionary vocabulary, often using pattern or word matching.…”

Conversational Networks for Automatic Online Moderation

“…Such an approach yields good results for de-obfuscation, but it is computationally expensive and requires a dataset of obfuscated words for training. More recently, Rojas et al [15] describe a more compact approach based on a dynamic programming sequence alignment algorithm. It has a different set of limitations, the main one being that it does not allow for one character to be used as an obfuscated version of several distinct original characters (it uses a one-to-one character mapping).…”

Below the Radar: Private Groups, Locked Platforms, and Ephemeral Content—Introduction to the Special Issue

“…This is due to the increasing plurality of online contexts and the fact that various obfuscation tactics have become common among users' repertoire of actions, next to the most studied practices of visibility. The assimilation-by protest movements, extremist groups but also by average users-of methods to escape censorship and being banned (Rojas-Galeano, 2017), along with the widespread use of VPNs, the algorithmic opacity of companies and institutions, and the use of memetic idiolects as forms of cultural cryptography, indicate that online social action takes increasingly place below the radar of research.…”