2018/08/30 by Leander Löw, Löw, Leander, Martin Spindler +3
Computer Science · Engineering · #Econometrics (econ.EM) #FOS: Computer and information sciences #FOS: Economics and business #Imbalanced Data Classification Techniques #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Machine Learning in Healthcare #Medical Imaging and Analysis #Topic Modeling
paper · pdf · doi:10.48550/arxiv.1808.10543
openalex publication_date 2018/08/30 · openalex created_date 2021/02/01 · openalex updated_date 2026/07/28
Insurance companies must manage millions of claims per year. While most of\nthese claims are non-fraudulent, fraud detection is core for insurance\ncompanies. The ultimate goal is a predictive model to single out the fraudulent\nclaims and pay out the non-fraudulent ones immediately. Modern machine learning\nmethods are well suited for this kind of problem. Health care claims often have\na data structure that is hierarchical and of variable length. We propose one\nmodel based on piecewise feed forward neural networks (deep learning) and\nanother model based on self-attention neural networks for the task of claim\nmanagement. We show that the proposed methods outperform bag-of-words based\nmodels, hand designed features, and models based on convolutional neural\nnetworks, on a data set of two million health care claims. The proposed\nself-attention method performs the best.\n