Covering Cracks in Content Moderation: Delexicalized Distant Supervision for Illicit Drug Jargon Detection

2025-03-19

AI Summary

Social media drug-related slang is becoming increasingly obfuscated and dynamic, rendering traditional lexicon-based detection methods vulnerable to evasion and ambiguous interpretations (e.g., “pot” may denote cannabis or a cooking utensil). To address this, we propose JEDIS—a lexicon-free, lemmatization-agnostic distant supervision framework that integrates stemming, contextualized embeddings, and BERT-based sequence classification. By modeling semantic intent rather than surface-form lexical patterns, JEDIS achieves robust generalization to out-of-vocabulary terms, homophones, and abbreviations. Evaluated on two manually annotated datasets, JEDIS outperforms state-of-the-art lexicon-based methods in F1-score and improves detection coverage by 37.2%. Qualitative analysis further demonstrates its strong robustness against semantic ambiguity and lexical evolution.

Technology Category

• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Data Mining & Knowledge Management: Semantic Web
• Search and Optimization: Distributed Search

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

In light of rising drug-related concerns and the increasing role of social media, sales and discussions of illicit drugs have become commonplace online. Social media platforms hosting user-generated content must therefore perform content moderation, which is a difficult task due to the vast amount of jargon used in drug discussions. Previous works on drug jargon detection were limited to extracting a list of terms, but these approaches have fundamental problems in practical application. First, they are trivially evaded using word substitutions. Second, they cannot distinguish whether euphemistic terms such as"pot"or"crack"are being used as drugs or in their benign meanings. We argue that drug content moderation should be done using contexts rather than relying on a banlist. However, manually annotated datasets for training such a task are not only expensive but also prone to becoming obsolete. We present JEDIS, a framework for detecting illicit drug jargon terms by analyzing their contexts. JEDIS utilizes a novel approach that combines distant supervision and delexicalization, which allows JEDIS to be trained without human-labeled data while being robust to new terms and euphemisms. Experiments on two manually annotated datasets show JEDIS significantly outperforms state-of-the-art word-based baselines in terms of F1-score and detection coverage in drug jargon detection. We also conduct qualitative analysis that demonstrates JEDIS is robust against pitfalls faced by existing approaches.

Problem

• Detecting illicit drug jargon in social media content
• Overcoming limitations of word-based drug detection methods
• Training context-aware models without human-labeled data

Innovation

• Uses distant supervision for training without human-labeled data
• Employs delexicalization to enhance robustness against new terms
• Analyzes context for detecting illicit drug jargon effectively