Pseudo Personalised Outreach
Pseudo-personalised outreach Real unsolicited emails sent to one person, labelled genuine or pseudo_personalised, with every sender identity replaced by a plausible fake. Code, spec and labelling rules: https://github.com/danielrosehill/Pseudo-Personalised-Spam-Classifier Labels and subtypes: docs/spec.md#labels How identities were replaced: docs/pseudonymisation.md Intended use Training and evaluating a small, cheap classifier (SetFit / ModernBERT class, well… See the full description on the dataset page: https://huggingface.co/datasets/danielrosehill/pseudo-personalised-outreach.