{"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/safe-a-neural-survival-analysis-model-for","title":"SAFE: A Neural Survival Analysis Model for Fraud Early Detection","arxiv_id":"1809.04683","date":"2018-09-12","proceeding":null,"authors":["Panpan Zheng","Shuhan Yuan","Xintao Wu"],"abstract":"Many online platforms have deployed anti-fraud systems to detect and prevent\nfraudulent activities. However, there is usually a gap between the time that a\nuser commits a fraudulent action and the time that the user is suspended by the\nplatform. How to detect fraudsters in time is a challenging problem. Most of\nthe existing approaches adopt classifiers to predict fraudsters given their\nactivity sequences along time. The main drawback of classification models is\nthat the prediction results between consecutive timestamps are often\ninconsistent. In this paper, we propose a survival analysis based fraud early\ndetection model, SAFE, which maps dynamic user activities to survival\nprobabilities that are guaranteed to be monotonically decreasing along time.\nSAFE adopts recurrent neural network (RNN) to handle user activity sequences\nand directly outputs hazard values at each timestamp, and then, survival\nprobability derived from hazard values is deployed to achieve consistent\npredictions. Because we only observe the user suspended time instead of the\nfraudulent activity time in the training data, we revise the loss function of\nthe regular survival model to achieve fraud early detection. Experimental\nresults on two real world datasets demonstrate that SAFE outperforms both the\nsurvival analysis model and recurrent neural network model alone as well as\nstate-of-the-art fraud early detection approaches.","url_abs":"http://arxiv.org/abs/1809.04683v2","url_pdf":"http://arxiv.org/pdf/1809.04683v2.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":"safe-a-neural-survival-analysis-model-for","repo_url":"https://github.com/PanpanZheng/SAFE","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"safe-a-neural-survival-analysis-model-for","repo_url":"https://github.com/jimmyhu4/fraud-detection","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"safe-a-neural-survival-analysis-model-for","repo_url":"https://github.com/leondepf/SAFE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"survival-analysis","task_name":"Survival Analysis"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1809.04683","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}