{"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/solving-the-false-positives-problem-in-fraud","title":"Solving the \"false positives\" problem in fraud prediction","arxiv_id":"1710.07709","date":"2017-10-20","proceeding":null,"authors":["Roy Wedge","James Max Kanter","Santiago Moral Rubio","Sergio Iglesias Perez","Kalyan Veeramachaneni"],"abstract":"In this paper, we present an automated feature engineering based approach to\ndramatically reduce false positives in fraud prediction. False positives plague\nthe fraud prediction industry. It is estimated that only 1 in 5 declared as\nfraud are actually fraud and roughly 1 in every 6 customers have had a valid\ntransaction declined in the past year. To address this problem, we use the Deep\nFeature Synthesis algorithm to automatically derive behavioral features based\non the historical data of the card associated with a transaction. We generate\n237 features (>100 behavioral patterns) for each transaction, and use a random\nforest to learn a classifier. We tested our machine learning model on data from\na large multinational bank and compared it to their existing solution. On an\nunseen data of 1.852 million transactions, we were able to reduce the false\npositives by 54% and provide a savings of 190K euros. We also assess how to\ndeploy this solution, and whether it necessitates streaming computation for\nreal time scoring. We found that our solution can maintain similar benefits\neven when historical features are computed once every 7 days.","url_abs":"http://arxiv.org/abs/1710.07709v1","url_pdf":"http://arxiv.org/pdf/1710.07709v1.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":"solving-the-false-positives-problem-in-fraud","repo_url":"https://github.com/An0wn/machinelearning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"automated-feature-engineering","task_name":"Automated Feature Engineering"},{"task_slug":"feature-engineering","task_name":"Feature Engineering"},{"task_slug":"prediction","task_name":"Prediction"},{"task_slug":null,"task_name":"valid"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}