Abstract
The expanding accessibility and appeal of investing have attracted millions of new retail investors. As such, investment discussion boards became the de facto communities where traders create, disseminate, and discuss investing ideas. These communities, which can provide useful information to support investors, have anecdotally also attracted a wide range of misbehavior – toxicity, spam/fraud, and reputation manipulation. This paper is the first comprehensive analysis of online misbehavior in the context of investment communities. We study TradingView, the largest online communication platform for financial trading. We collect 2.76M user profiles with their corresponding social graphs, 4.2M historical article posts, and 5.3M comments, including information on nearly 4,000 suspended accounts and 17,000 removed comments. Price fluctuations seem to drive abuse across the platform and certain types of assets, such as meme stocks, attract disproportionate misbehavior. Suspended user accounts tend to form more closely-knit communities than those formed by non-suspended accounts; and paying accounts are less likely to be suspended than free accounts even when posting similar levels of content violating platform policies. We conclude by offering guidelines on how to adapt content moderation efforts to fit the particularities of online investment communities.
Tom Magelinski, PhD
Senior Data Scientist - Information Extraction and Generative AI
I build AI systems that help domain experts understand vast amounts of data through state-of-the-art techniques from natural language processing, generative modeling, graph ML, and network science. I’m particularly interested researching and developing techniques to combine NLP and graph-based approaches to capture complex relationships in unstructured data.