{"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/realistic-risk-mitigating-recommendations-via","title":"Realistic risk-mitigating recommendations via inverse classification","arxiv_id":"1611.04199","date":"2016-11-13","proceeding":null,"authors":["Michael T. Lash","W. Nick Street"],"abstract":"Inverse classification, the process of making meaningful perturbations to a\ntest point such that it is more likely to have a desired classification, has\npreviously been addressed using data from a single static point in time. Such\nan approach yields inflated probability estimates, stemming from an implicitly\nmade assumption that recommendations are implemented instantaneously. We\npropose using longitudinal data to alleviate such issues in two ways. First, we\nuse past outcome probabilities as features in the present. Use of such past\nprobabilities ties historical behavior to the present, allowing for more\ninformation to be taken into account when making initial probability estimates\nand subsequently performing inverse classification. Secondly, following inverse\nclassification application, optimized instances' unchangeable features\n(e.g.,~age) are updated using values from the next longitudinal time period.\nOptimized test instance probabilities are then reassessed. Updating the\nunchangeable features in this manner reflects the notion that improvements in\noutcome likelihood, which result from following the inverse classification\nrecommendations, do not materialize instantaneously. As our experiments\ndemonstrate, more realistic estimates of probability can be obtained by\nfactoring in such considerations.","url_abs":"http://arxiv.org/abs/1611.04199v1","url_pdf":"http://arxiv.org/pdf/1611.04199v1.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":"realistic-risk-mitigating-recommendations-via","repo_url":"https://github.com/michael-lash/LongARIC","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"classification","task_name":"General Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}