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

切換: 標籤 | MARC模式 | ISBD

FindBook

Google Book

Amazon

博客來

A Study of Administrative Data Representation for Machine Learning.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	A Study of Administrative Data Representation for Machine Learning./
作者:	Asadzadehzanjani, Negin.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2022,
面頁冊數:	189 p.
附註:	Source: Dissertations Abstracts International, Volume: 84-01, Section: B.
Contained By:	Dissertations Abstracts International84-01B.
標題:	Public health. -
ISBN:	9798834064343

A Study of Administrative Data Representation for Machine Learning.
Asadzadehzanjani, Negin.

A Study of Administrative Data Representation for Machine Learning. - Ann Arbor : ProQuest Dissertations & Theses, 2022 - 189 p.

Source: Dissertations Abstracts International, Volume: 84-01, Section: B.

Thesis (Ph.D.)--George Mason University, 2022.

This item must not be sold to any third party vendors.

Administrative data, including medical claims, are frequently used to train machine learning-based models used for predicting patient outcomes. Despite many efforts in using administrative codes (medical codes) in claims data, little systematic work has been done in understanding how the codes in such data should be represented before model construction. Traditionally, the presence/absence of these codes representing diagnoses or procedures (Binary Representation) over a fixed period (typically one year) is used. More recently, some studies included temporal information into data representation, such as counting, calculating time from diagnosis, and using multiple time windows. However, these methods were not able to comprehensively capture temporal information in data and much of temporal information such as the exact time of the occurrence of an event, and the exact sequence of an event are missed. This dissertation presents the results of development and investigation of two additional methods of administrative data representation (Temporal Min-Max and Trajectory Representation) specific to diagnoses extracted from claims data before applying machine learning algorithms. It then presents a large-scale experimental evaluation of these methods by comparing them with traditional Binary Representation using four classification problems: one-year mortality prediction and high utilization of medical services prediction, prediction of chronic kidney disease and prediction of congestive heart failure. It was shown that the optimal way of representing the data is problem-dependent, thus optimization of representation parameters is required as part of the modeling.

ISBN: 9798834064343Subjects--Topical Terms:

534748
Public health.
Subjects--Index Terms:

Data preprocessing

A Study of Administrative Data Representation for Machine Learning.
LDR:02774nmm a2200361 4500 001 2349715
005 20221003074937.5
008 241004s2022 eng d
020 $a 9798834064343
035 $a (MiAaPQ)AAI29065054
035 $a AAI29065054
040 $a MiAaPQ $c MiAaPQ
100 1 $a Asadzadehzanjani, Negin. $3 3689133
245 1 0 $a A Study of Administrative Data Representation for Machine Learning.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2022
300 $a 189 p.
500 $a Source: Dissertations Abstracts International, Volume: 84-01, Section: B.
500 $a Advisor: Wojtusiak, Janusz.
502 $a Thesis (Ph.D.)--George Mason University, 2022.
506 $a This item must not be sold to any third party vendors.
520 $a Administrative data, including medical claims, are frequently used to train machine learning-based models used for predicting patient outcomes. Despite many efforts in using administrative codes (medical codes) in claims data, little systematic work has been done in understanding how the codes in such data should be represented before model construction. Traditionally, the presence/absence of these codes representing diagnoses or procedures (Binary Representation) over a fixed period (typically one year) is used. More recently, some studies included temporal information into data representation, such as counting, calculating time from diagnosis, and using multiple time windows. However, these methods were not able to comprehensively capture temporal information in data and much of temporal information such as the exact time of the occurrence of an event, and the exact sequence of an event are missed. This dissertation presents the results of development and investigation of two additional methods of administrative data representation (Temporal Min-Max and Trajectory Representation) specific to diagnoses extracted from claims data before applying machine learning algorithms. It then presents a large-scale experimental evaluation of these methods by comparing them with traditional Binary Representation using four classification problems: one-year mortality prediction and high utilization of medical services prediction, prediction of chronic kidney disease and prediction of congestive heart failure. It was shown that the optimal way of representing the data is problem-dependent, thus optimization of representation parameters is required as part of the modeling.
590 $a School code: 0883.
650 4 $a Public health. $3 534748
650 4 $a Artificial intelligence. $3 516317
650 4 $a Health sciences. $3 3168359
653 $a Data preprocessing
653 $a Health informatics
653 $a Medical claims
653 $a Supervised learning
653 $a Temporal machine learning
690 $a 0573
690 $a 0800
690 $a 0566
710 2 0 $a George Mason University. $b Health Services Research. $3 3689134
773 0 $t Dissertations Abstracts International $g 84-01B.
790 $a 0883
791 $a Ph.D.
792 $a 2022
793 $a English