File Download
There are no files associated with this item.
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1111/jori.12359
- Scopus: eid_2-s2.0-85111089470
- WOS: WOS:000678356200001
- Find via
Supplementary
- Citations:
- Appears in Collections:
Article: Insurance fraud detection with unsupervised deep learning
Title | Insurance fraud detection with unsupervised deep learning |
---|---|
Authors | |
Issue Date | 2021 |
Publisher | Wiley-Blackwell Publishing, Inc. The Journal's web site is located at https://onlinelibrary.wiley.com/journal/15396975 |
Citation | Journal of Risk and Insurance, 2021, v. 88 n. 3, p. 591-624 How to Cite? |
Abstract | The objective of this paper is to propose a novel deep learning methodology to gain pragmatic insights into the behavior of an insured person using unsupervised variable importance. It lays the groundwork for understanding how insights can be gained into the fraudulent behavior of an insured person with minimum effort. Starting with a preliminary investigation of the limitations of the existing fraud detection models, we propose a new variable importance methodology incorporated with two prominent unsupervised deep learning models, namely, the autoencoder and the variational autoencoder. Each model's dynamics is discussed to inform the reader on how models can be adapted for fraud detection and how results can be perceived appropriately. Both qualitative and quantitative performance evaluations are conducted, although a greater emphasis is placed on qualitative evaluation. To broaden the scope of reference of fraud detection setting, various metrics are used in the qualitative evaluation. |
Persistent Identifier | http://hdl.handle.net/10722/305036 |
ISSN | 2023 Impact Factor: 2.1 2023 SCImago Journal Rankings: 1.203 |
ISI Accession Number ID |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Gomes, C | - |
dc.contributor.author | Jin, Z | - |
dc.contributor.author | Yang, H | - |
dc.date.accessioned | 2021-10-05T02:38:50Z | - |
dc.date.available | 2021-10-05T02:38:50Z | - |
dc.date.issued | 2021 | - |
dc.identifier.citation | Journal of Risk and Insurance, 2021, v. 88 n. 3, p. 591-624 | - |
dc.identifier.issn | 0022-4367 | - |
dc.identifier.uri | http://hdl.handle.net/10722/305036 | - |
dc.description.abstract | The objective of this paper is to propose a novel deep learning methodology to gain pragmatic insights into the behavior of an insured person using unsupervised variable importance. It lays the groundwork for understanding how insights can be gained into the fraudulent behavior of an insured person with minimum effort. Starting with a preliminary investigation of the limitations of the existing fraud detection models, we propose a new variable importance methodology incorporated with two prominent unsupervised deep learning models, namely, the autoencoder and the variational autoencoder. Each model's dynamics is discussed to inform the reader on how models can be adapted for fraud detection and how results can be perceived appropriately. Both qualitative and quantitative performance evaluations are conducted, although a greater emphasis is placed on qualitative evaluation. To broaden the scope of reference of fraud detection setting, various metrics are used in the qualitative evaluation. | - |
dc.language | eng | - |
dc.publisher | Wiley-Blackwell Publishing, Inc. The Journal's web site is located at https://onlinelibrary.wiley.com/journal/15396975 | - |
dc.relation.ispartof | Journal of Risk and Insurance | - |
dc.title | Insurance fraud detection with unsupervised deep learning | - |
dc.type | Article | - |
dc.identifier.email | Yang, H: hlyang@hku.hk | - |
dc.identifier.authority | Yang, H=rp00826 | - |
dc.description.nature | link_to_subscribed_fulltext | - |
dc.identifier.doi | 10.1111/jori.12359 | - |
dc.identifier.scopus | eid_2-s2.0-85111089470 | - |
dc.identifier.hkuros | 326291 | - |
dc.identifier.volume | 88 | - |
dc.identifier.issue | 3 | - |
dc.identifier.spage | 591 | - |
dc.identifier.epage | 624 | - |
dc.identifier.isi | WOS:000678356200001 | - |
dc.publisher.place | United States | - |