| 1. |
- Bahnsen, Alejandro Correa, et al.
(författare)
-
Feature engineering strategies for credit card fraud detection
- 2016
-
Ingår i: Expert systems with applications. - : Elsevier BV. - 0957-4174 .- 1873-6793. ; 51, s. 134-142
-
Tidskriftsartikel (refereegranskat)abstract
- Every year billions of Euros are lost worldwide due to credit card fraud. Thus, forcing financial institutions to continuously improve their fraud detection systems. In recent years, several studies have proposed the use of machine learning and data mining techniques to address this problem. However, most studies used some sort of misclassification measure to evaluate the different solutions, and do not take into account the actual financial costs associated with the fraud detection process. Moreover, when constructing a credit card fraud detection model, it is very important how to extract the right features from the transactional data. This is usually done by aggregating the transactions in order to observe the spending behavioral patterns of the customers. In this paper we expand the transaction aggregation strategy, and propose to create a new set of features based on analyzing the periodic behavior of the time of a transaction using the von Mises distribution. Then, using a real credit card fraud dataset provided by a large European card processing company, we compare state-of-the-art credit card fraud detection models, and evaluate how the different sets of features have an impact on the results. By including the proposed periodic features into the methods, the results show an average increase in savings of 13%. (C) 2016 Elsevier Ltd. All rights reserved.
|
|
| 2. |
- Bahnsen, Alejandro, et al.
(författare)
-
Cost sensitive credit card fraud detection using bayes minimum risk
- 2013
-
Ingår i: Proceedings - 2013 12th International Conference on Machine Learning and Applications, ICMLA 2013. - : IEEE Computer Society. - 9780769551449 ; , s. 333-338
-
Konferensbidrag (refereegranskat)abstract
- Credit card fraud is a growing problem that affects card holders around the world. Fraud detection has been an interesting topic in machine learning. Nevertheless, current state of the art credit card fraud detection algorithms miss to include the real costs of credit card fraud as a measure to evaluate algorithms. In this paper a new comparison measure that realistically represents the monetary gains and losses due to fraud detection is proposed. Moreover, using the proposed cost measure a cost sensitive method based on Bayes minimum risk is presented. This method is compared with state of the art algorithms and shows improvements up to 23% measured by cost. The results of this paper are based on real life transactional data provided by a large European card processing company.
|
|
| 3. |
- Yazdani, Mazyar, et al.
(författare)
-
Tear Production Levels and Dry Eye Disease Severity in a Large Norwegian Cohort
- 2018
-
Ingår i: Current Eye Research. - : TAYLOR & FRANCIS INC. - 0271-3683 .- 1460-2202. ; 43:12, s. 1465-1470
-
Tidskriftsartikel (refereegranskat)abstract
- Purpose: To determine if the Schirmer I test (without anesthesia) cut-off value is a predictor of dry eye severity in a large Norwegian cohort of dry eye disease (DED) patients, which are grouped into six levels of tear production. Methods: Patients (n = 1090) with DED of different etiologies received an extensive dry eye work-up: osmolarity (Osm), tear meniscus height (TMH), tear film break-up time (TFBUT), ocular protection index (OPI), ocular surface staining (OSS), Schirmer I test (ST), meibum expressibility (ME), and meibum quality (MQ). Classification of dry eye severity level (DESL) and diagnosis of meibomian gland dysfunction (MGD) were also included. The cohort was divided into six groups: below and above cut-off values of 5 (groups 1 and 2), 10 (groups 3 and 4), and 15 mm (groups 5 and 6) of ST. Mann-Whitney test and Chi-Square test were used for group comparison of parameters (p amp;lt;= 0.05). Results: The groups 1, 3, and 5 had values indicating more severe DED than the groups 2, 4, 6 with significant difference in DESL, Osm, TFBUT, OPI, OSS, and TMH. Regardless of the choice of cut-off values, there was no statistically significant difference in ME, MQ, and MGD between groups below and above selected cut-off value. When gender difference was considered in each group, significant difference was only observed for DESL (groups 2, 4, and 5), TFBUT (groups 2, 4, and 5), OPI (groups 2 and 6), and ME (group1). Conclusions: Schirmer I is a robust discriminator for DESL, Osm, TFBUT, OPI, OSS, and TMH, but not for ME, MQ, and MGD. Patients with lower tear production levels presented with more severe DED at all three defined cut-off values. Interestingly, the differences in the mean values of DESL were minimal although statistically significant. Thus, the clinical value of different Schirmer levels appears to be limited.
|
|
