FRAUD DETECTION IN HEALTHCARE
| dc.contributor.advisor | Gangopadhyay, Aryya | |
| dc.contributor.author | Chen, SongChen, Song | |
| dc.contributor.department | Information Systems | |
| dc.contributor.program | Information Systems | |
| dc.date.accessioned | 2019-10-11T13:57:54Z | |
| dc.date.available | 2019-10-11T13:57:54Z | |
| dc.date.issued | 2017-01-01 | |
| dc.description.abstract | Healthcare Fraud is an important problem that needs more attention from both the federal government and the health services communities. The United States loses at least $60 billion in healthcare fraud every year, and some organizations put the cost as high as 10% of the nation's total healthcare spending, which exceeds $2 trillion in 2014. The federal government relies on its contractors to combat health fraud and requires all insurance companies to have fraud units dedicated to detecting and investigating fraud. In this dissertations paper, we will discuss this research in analytical techniques of healthcare fraud detection. We propose a fraud detection framework which starts with understanding the fraud schemes and their feature sets. This framework is able to detect both known and unknown types of fraud. We explore physician relationships and convert them into a network structure, with which we build a few algorithms to detect relationship-based healthcare fraud. We explore and compare the Social Network Analysis and Predictive Modeling along with the benefits and challenges. In this dissertations paper, we discuss 11 dierent healthcare fraud schemes, case examples and methods to detect them. We develop innovative new algorithms in Pattern Analysis (Chapter 3.4) to detect potential Hit and Run schemes and other suspicous billing behaviors. The community detection algorithms are discussed in Chapter Community Detections (Chapter 3.5) and Chapter Prelimiary Work (Chapter 2.4 and 2.5). To evaluate these algorithms, we develop a method to create a synthesized dataset of any number of providers, and test its similarities with real world health- care claims (Chapter 4.1). We test the Pattern Analysis and show results in Chap- ter 4.2. Lastly, these community detection algorithms prove to be very eective by comparing running time and accuracy with other community detection algorithms (Chapter 4.3). My algorithms can be extended to detect communities of any sizes, and the success rate is better than other algorithms in overlapping communities. | |
| dc.genre | dissertations | |
| dc.identifier | doi:10.13016/m2b6yf-upve | |
| dc.identifier.other | 11726 | |
| dc.identifier.uri | http://hdl.handle.net/11603/15596 | |
| dc.language | en | |
| dc.relation.isAvailableAt | The University of Maryland, Baltimore County (UMBC) | |
| dc.relation.ispartof | UMBC Information Systems Department Collection | |
| dc.relation.ispartof | UMBC Theses and Dissertations Collection | |
| dc.relation.ispartof | UMBC Graduate School Collection | |
| dc.relation.ispartof | UMBC Student Collection | |
| dc.rights | This item may be protected under Title 17 of the U.S. Copyright Law. It is made available by UMBC for non-commercial research and education. For permission to publish or reproduce, please see http://aok.lib.umbc.edu/specoll/repro.php or contact Special Collections at speccoll(at)umbc.edu | |
| dc.source | Original File Name: Chen_umbc_0434D_11726.pdf | |
| dc.subject | community detection | |
| dc.subject | fraud detection | |
| dc.subject | fraud schemes | |
| dc.subject | healthcare | |
| dc.subject | pattern analysis | |
| dc.subject | synthetic dataset | |
| dc.title | FRAUD DETECTION IN HEALTHCARE | |
| dc.type | Text | |
| dcterms.accessRights | Access limited to the UMBC community. Item may possibly be obtained via Interlibrary Loan through a local library, pending author/copyright holder's permission. |