{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/like-trainer-like-bot-inheritance-of-bias-in","title":"Like trainer, like bot? Inheritance of bias in algorithmic content moderation","arxiv_id":"1707.01477","date":"2017-07-05","proceeding":null,"authors":["Reuben Binns","Michael Veale","Max Van Kleek","Nigel Shadbolt"],"abstract":"The internet has become a central medium through which `networked publics'\nexpress their opinions and engage in debate. Offensive comments and personal\nattacks can inhibit participation in these spaces. Automated content moderation\naims to overcome this problem using machine learning classifiers trained on\nlarge corpora of texts manually annotated for offence. While such systems could\nhelp encourage more civil debate, they must navigate inherently normatively\ncontestable boundaries, and are subject to the idiosyncratic norms of the human\nraters who provide the training data. An important objective for platforms\nimplementing such measures might be to ensure that they are not unduly biased\ntowards or against particular norms of offence. This paper provides some\nexploratory methods by which the normative biases of algorithmic content\nmoderation systems can be measured, by way of a case study using an existing\ndataset of comments labelled for offence. We train classifiers on comments\nlabelled by different demographic subsets (men and women) to understand how\ndifferences in conceptions of offence between these groups might affect the\nperformance of the resulting models on various test sets. We conclude by\ndiscussing some of the ethical choices facing the implementers of algorithmic\nmoderation systems, given various desired levels of diversity of viewpoints\namongst discussion participants.","url_abs":"http://arxiv.org/abs/1707.01477v1","url_pdf":"http://arxiv.org/pdf/1707.01477v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"like-trainer-like-bot-inheritance-of-bias-in","repo_url":"https://github.com/sociam/liketrainer","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"navigate","task_name":"Navigate"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1707.01477","atlas_url":"https://app.syntology.ai/?focus=1707.01477","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}