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

Use of the Expectation Maximization ...

Qian, Jiafan .

FindBook

Google Book

Amazon

博客來

Use of the Expectation Maximization Algorithm to Recover Bayesian Network Structure with Incomplete Data.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Use of the Expectation Maximization Algorithm to Recover Bayesian Network Structure with Incomplete Data./
作者:	Qian, Jiafan .
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2020,
面頁冊數:	62 p.
附註:	Source: Masters Abstracts International, Volume: 82-01.
Contained By:	Masters Abstracts International82-01.
標題:	Applied mathematics. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=27960962
ISBN:	9798635226117

Use of the Expectation Maximization Algorithm to Recover Bayesian Network Structure with Incomplete Data.
Qian, Jiafan .

Use of the Expectation Maximization Algorithm to Recover Bayesian Network Structure with Incomplete Data. - Ann Arbor : ProQuest Dissertations & Theses, 2020 - 62 p.

Source: Masters Abstracts International, Volume: 82-01.

Thesis (M.S.)--University of Colorado at Boulder, 2020.

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

Bayesian networks are graphical models that can be used to represent dependencies between random variables. They consist of a directed acyclic graph (DAG) where nodes represent the random variables, and a set of conditional probabilities or probability density functions giving speciﬁc relationships between each random variable and those represented by neighboring nodes. The two main problems to solve for Bayesian networks are estimation of the parameters for the conditional distributions and structure recovery which will aﬀect which nodes are neighbors and therefore aﬀecting other nodes. In this thesis, we address the latter problem of structure recovery speciﬁcally in the presence of missing data. We describe the popular Expectation-Maximization (EM) algorithm and its speciﬁc application to estimating Bayesian network parameters which can then be used for structure recovery. We address three diﬀerent missing data mechanisms, including data that are "missing completely at random" (MCAR), "missing at random" (MAR), and "not missing at random" (NMAR). We describe tests for determining which mechanism is appropriate and then perform network recovery on simulated data so that we may compare our results to ground truth. Finally, we investigate the sensitivity of our results to varying percentages of missing data.

ISBN: 9798635226117Subjects--Topical Terms:

2122814
Applied mathematics.
Subjects--Index Terms:

Bayesian networks

Use of the Expectation Maximization Algorithm to Recover Bayesian Network Structure with Incomplete Data.
LDR:02499nmm a2200361 4500 001 2270268
005 20200921070809.5
008 220629s2020 ||||||||||||||||| ||eng d
020 $a 9798635226117
035 $a (MiAaPQ)AAI27960962
035 $a AAI27960962
040 $a MiAaPQ $c MiAaPQ
100 1 $a Qian, Jiafan . $3 3547643
245 1 0 $a Use of the Expectation Maximization Algorithm to Recover Bayesian Network Structure with Incomplete Data.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2020
300 $a 62 p.
500 $a Source: Masters Abstracts International, Volume: 82-01.
500 $a Advisor: Corcoran, Jem.
502 $a Thesis (M.S.)--University of Colorado at Boulder, 2020.
506 $a This item must not be sold to any third party vendors.
520 $a Bayesian networks are graphical models that can be used to represent dependencies between random variables. They consist of a directed acyclic graph (DAG) where nodes represent the random variables, and a set of conditional probabilities or probability density functions giving speciﬁc relationships between each random variable and those represented by neighboring nodes. The two main problems to solve for Bayesian networks are estimation of the parameters for the conditional distributions and structure recovery which will aﬀect which nodes are neighbors and therefore aﬀecting other nodes. In this thesis, we address the latter problem of structure recovery speciﬁcally in the presence of missing data. We describe the popular Expectation-Maximization (EM) algorithm and its speciﬁc application to estimating Bayesian network parameters which can then be used for structure recovery. We address three diﬀerent missing data mechanisms, including data that are "missing completely at random" (MCAR), "missing at random" (MAR), and "not missing at random" (NMAR). We describe tests for determining which mechanism is appropriate and then perform network recovery on simulated data so that we may compare our results to ground truth. Finally, we investigate the sensitivity of our results to varying percentages of missing data.
590 $a School code: 0051.
650 4 $a Applied mathematics. $3 2122814
650 4 $a Statistics. $3 517247
650 4 $a Statistical physics. $3 536281
653 $a Bayesian networks
653 $a Directed acyclic graph
653 $a Conditional distributions
653 $a Structure recovery
690 $a 0364
690 $a 0217
690 $a 0463
710 2 $a University of Colorado at Boulder. $b Applied Mathematics. $3 1030307
773 0 $t Masters Abstracts International $g 82-01.
790 $a 0051
791 $a M.S.
792 $a 2020
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=27960962