東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁 到查詢結果 [ null ]

切換: 標籤 | MARC模式 | ISBD

Unsupervised Learning for Anomaly De...

Kemp, James,

FindBook

Google Book

Amazon

博客來

Unsupervised Learning for Anomaly Detection in Australian Medical Payment Data /

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Unsupervised Learning for Anomaly Detection in Australian Medical Payment Data // James Kemp.
作者:	Kemp, James,
面頁冊數:	1 electronic resource (181 pages)
附註:	Source: Dissertations Abstracts International, Volume: 86-08, Section: B.
Contained By:	Dissertations Abstracts International86-08B.
標題:	Government agencies. -
電子資源:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31863411
ISBN:	9798302895769

Unsupervised Learning for Anomaly Detection in Australian Medical Payment Data /
Kemp, James,

Unsupervised Learning for Anomaly Detection in Australian Medical Payment Data /James Kemp. - 1 electronic resource (181 pages)

Source: Dissertations Abstracts International, Volume: 86-08, Section: B.

Fraudulent or wasteful medical insurance claims made by health care providers are costly for insurers. Typically, OECD healthcare organisations lose 3-8% of total expenditure due to fraud. As Australia's universal public health insurer, Medicare Australia, spends approximately $A 34 billion per annum on the Medicare Benefits Schedule (MBS) and Pharmaceutical Benefits Scheme, wasted spending of $A1-2.7 billion could be expected. However, fewer than 1% of claims to Medicare Australia are detected as fraudulent, below international benchmarks.Variation is common in medicine, and health conditions, along with their presentation and treatment, are heterogenous by nature. Increasing volumes of data and rapidly changing patterns bring challenges which require novel solutions. Machine learning and data mining are becoming commonplace in this field, but no gold standard is yet available.In this project, requirements are developed for real-world application to compliance analytics at the Australian Government Department of Health and Aged Care (DoH), covering: unsupervised learning; problem generalisation; human interpretability; context discovery; and cost prediction. Three novel methods are presented which rank providers by potentially recoverable costs. These methods used association analysis, topic modelling, and sequential pattern mining to provide interpretable, expert-editable models of typical provider claims. Anomalous providers are identified through comparison to the typical models, using metrics based on costs of excess or upgraded services. Domain knowledge is incorporated in a machine-friendly way in two of the methods through the use of the MBS as an ontology. Validation by subject-matter experts and comparison to existing techniques shows that the methods perform well. The methods are implemented in a software framework which enables rapid prototyping and quality assurance. The code is implemented at the DoH, and further applications as decision-support systems are in progress. The developed requirements will apply to future work in this field.

English

ISBN: 9798302895769Subjects--Topical Terms:

3562805
Government agencies.
Subjects--Index Terms:

Medicare Benefits Schedule

Unsupervised Learning for Anomaly Detection in Australian Medical Payment Data /
LDR:03385nmm a22003973i 4500 001 2437708
005 20251124055000.5
006 m o d
007 cr|nu||||||||
008 261011s2023 miu||||||m |||||||eng d
020 $a 9798302895769
035 $a (MiAaPQD)AAI31863411
035 $a AAI31863411
040 $a MiAaPQD $b eng $c MiAaPQD $e rda
100 1 $a Kemp, James, $e author. $3 3828782
245 1 0 $a Unsupervised Learning for Anomaly Detection in Australian Medical Payment Data / $c James Kemp.
264 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2023
300 $a 1 electronic resource (181 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 86-08, Section: B.
500 $a Advisors: Bain, Michael.
502 $b Ph.D. $c University of New South Wales (Australia) $d 2023.
520 $a Fraudulent or wasteful medical insurance claims made by health care providers are costly for insurers. Typically, OECD healthcare organisations lose 3-8% of total expenditure due to fraud. As Australia's universal public health insurer, Medicare Australia, spends approximately $A 34 billion per annum on the Medicare Benefits Schedule (MBS) and Pharmaceutical Benefits Scheme, wasted spending of $A1-2.7 billion could be expected. However, fewer than 1% of claims to Medicare Australia are detected as fraudulent, below international benchmarks.Variation is common in medicine, and health conditions, along with their presentation and treatment, are heterogenous by nature. Increasing volumes of data and rapidly changing patterns bring challenges which require novel solutions. Machine learning and data mining are becoming commonplace in this field, but no gold standard is yet available.In this project, requirements are developed for real-world application to compliance analytics at the Australian Government Department of Health and Aged Care (DoH), covering: unsupervised learning; problem generalisation; human interpretability; context discovery; and cost prediction. Three novel methods are presented which rank providers by potentially recoverable costs. These methods used association analysis, topic modelling, and sequential pattern mining to provide interpretable, expert-editable models of typical provider claims. Anomalous providers are identified through comparison to the typical models, using metrics based on costs of excess or upgraded services. Domain knowledge is incorporated in a machine-friendly way in two of the methods through the use of the MBS as an ontology. Validation by subject-matter experts and comparison to existing techniques shows that the methods perform well. The methods are implemented in a software framework which enables rapid prototyping and quality assurance. The code is implemented at the DoH, and further applications as decision-support systems are in progress. The developed requirements will apply to future work in this field.
546 $a English
590 $a School code: 0423
650 4 $a Government agencies. $3 3562805
650 4 $a Software. $2 gtt. $3 619355
650 4 $a Deep learning. $3 3554982
650 4 $a Industrial research. $3 3775014
650 4 $a Health care policy. $3 3550686
650 4 $a Ontology. $3 530874
650 4 $a Decision making. $3 517204
650 4 $a Elder care. $3 3696970
650 4 $a Public health. $3 534748
650 4 $a Fraud. $3 721834
650 4 $a Computer science. $3 523869
653 $a Medicare Benefits Schedule
653 $a OECD healthcare
690 $a 0573
690 $a 0984
690 $a 0769
710 2 $a University of New South Wales (Australia). $e degree granting institution. $3 3815488
720 1 $a Bain, Michael $e degree supervisor.
773 0 $t Dissertations Abstracts International $g 86-08B.
790 $a 0423
791 $a Ph.D.
792 $a 2023
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31863411