This paper aims to better predict highly skewed auto insurance claims by combining candidate predictions. We analyze a version of the Kangaroo Auto Insurance company data and study effects different methods using five measures prediction accuracy. The results show following. First, when there is an outstanding (in terms Gini Index) among candidates, “forecast combination puzzle” phenomenon disa...

With the developments in the information technology, fraud is spreading all over the world, resulting in huge financial losses. Though fraud prevention mechanisms such as CHIP&PIN are developed for credit card systems, these mechanisms do not prevent the most common fraud types such as fraudulent credit card usages over virtual POS (Point Of Sale) terminals or mail orders so called online credi...

In this paper a fraud detection method is proposed which user behaviors are modeled using two main components namely the un-normal trend analysis component and scenario based component. The extent of deviation of a transaction from his/her normal behavior is estimated using fuzzy membership functions. The results of applying all membership functions on a transaction will then be infused and a f...

One method for detecting fraud is to check for suspicious changes in user behavior over time. This paper describes the automatic design of user profiling methods for the purpose of fraud detection, using a series of data mining and machine learning techniques. It uses a rule-learning program to uncover indicators of fraudulent behavior from a large database customer transactions. Then the indic...

Title of dissertation: Securities Fraud: An Economic Analysis Yue Wang, Doctor of Philosophy, 2005 Dissertation directed by: Professor Lemma Senbet, Professor Nagpurnanand Prabhala Department of Finance This thesis develops an economic analysis of securities fraud. The thesis consists of a theory essay and an empirical essay. In the theory essay, I analyze a firm’s propensity to commit securiti...

As online commerce becomes more common, fraud is an increasingly important concern. Current anti-fraud techniques include blacklists, hand-crafted rules, and review of individual transactions by human experts. These methods have achieved good results, but a significant number of costly fraudulent transactions still occur, and the fraud detection process is expensive due to its heavy reliance on...

This paper reports an experiment aimed at generating synthetic test data for fraud detection in an IP based videoon-demand service. The data generation verifies a methodology previously developed by the present authors [7] that ensures that important statistical properties of the authentic data are preserved by using authentic normal data and fraud as a seed for generating synthetic data. This ...

<span lang="EN-US">In a streaming environment, data is continuously generated and processed in an ongoing manner, it necessary to detect fraudulent transactions quickly prevent significant financial losses. Hence, this paper proposes machine learning-based approach for detecting with focus on addressing concept drift. The utilizes the extreme gradient boosting (XGBoost) algorithm. Additio...