東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Bayesian Latent Space Approaches to ...

Pham, Hanh Thi Duc,

Linked to FindBook

Google Book

Amazon

博客來

Bayesian Latent Space Approaches to Network Analysis /

Record Type:	Electronic resources : Monograph/item
Title/Author:	Bayesian Latent Space Approaches to Network Analysis // Hanh Thi Duc Pham.
Author:	Pham, Hanh Thi Duc,
Description:	1 electronic resource (104 pages)
Notes:	Source: Dissertations Abstracts International, Volume: 86-01, Section: B.
Contained By:	Dissertations Abstracts International86-01B.
Subject:	Biostatistics. -
Online resource:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31235919
ISBN:	9798383188453

Bayesian Latent Space Approaches to Network Analysis /
Pham, Hanh Thi Duc,

Bayesian Latent Space Approaches to Network Analysis /Hanh Thi Duc Pham. - 1 electronic resource (104 pages)

Source: Dissertations Abstracts International, Volume: 86-01, Section: B.

Networks naturally arise in situations where subjects are interconnected rather than being completely independent. Many traditional statistical methods cannot be directly applied to network data due to the correlated nature. This dissertation leverages the framework of latent space models and Bayesian statistics to advance network analysis methodologies in two main areas: network edge clustering and social influence estimation. For the former, we addressed the challenge of selecting the number of clusters, a common issue in clustering problems, including edge clustering. We proposed a novel extension to the state-of-the-art latent space approach for edge clustering. Our extension facilitates the automatic selection of the number of clusters, leading to a significant reduction in computational time required for model-based edge clustering. The latter part of the dissertation employs latent space models to tackle a crucial question in network science: the accurate estimation of social influence, defined as the impact of an entity on the behaviors of their connections, in cross-sectional data. We utilized latent space models to estimate latent homophily, the tendency for similar entities to form connections, and incorporated this effect into the social influence estimation. We detailed the theoretical foundation of our proposed models, demonstrating its relationship with the network autocorrelation model (NAM), a widely-used model for social influence in cross-sectional data. Through extensive simulation studies, we demonstrated the effectiveness of these homophily-adjusted models in providing deeper insights into network dynamics, particularly emphasizing the significance of accounting for latent homophily in social influence estimation. We included an R package for each method presented, ensuring straightforward implementation for users.

English

ISBN: 9798383188453Subjects--Topical Terms:

1002712
Biostatistics.
Subjects--Index Terms:

Networks

Bayesian Latent Space Approaches to Network Analysis /
LDR:03289nmm a22004213i 4500 001 2429328
005 20250430100800.5
006 m o d
007 cr|nu||||||||
008 261011s2024 miu||||||m |||||||eng d
020 $a 9798383188453
035 $a (MiAaPQD)AAI31235919
035 $a AAI31235919
040 $a MiAaPQD $b eng $c MiAaPQD $e rda
100 1 $a Pham, Hanh Thi Duc, $e author. $3 3817108
245 1 0 $a Bayesian Latent Space Approaches to Network Analysis / $c Hanh Thi Duc Pham.
264 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2024
300 $a 1 electronic resource (104 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: 86-01, Section: B.
500 $a Advisors: Sewell, Daniel K. Committee members: Roberts, Emily K.; Brown, Grant D.; Oleson, Jacob J.
502 $b Ph.D. $c The University of Iowa $d 2024.
520 $a Networks naturally arise in situations where subjects are interconnected rather than being completely independent. Many traditional statistical methods cannot be directly applied to network data due to the correlated nature. This dissertation leverages the framework of latent space models and Bayesian statistics to advance network analysis methodologies in two main areas: network edge clustering and social influence estimation. For the former, we addressed the challenge of selecting the number of clusters, a common issue in clustering problems, including edge clustering. We proposed a novel extension to the state-of-the-art latent space approach for edge clustering. Our extension facilitates the automatic selection of the number of clusters, leading to a significant reduction in computational time required for model-based edge clustering. The latter part of the dissertation employs latent space models to tackle a crucial question in network science: the accurate estimation of social influence, defined as the impact of an entity on the behaviors of their connections, in cross-sectional data. We utilized latent space models to estimate latent homophily, the tendency for similar entities to form connections, and incorporated this effect into the social influence estimation. We detailed the theoretical foundation of our proposed models, demonstrating its relationship with the network autocorrelation model (NAM), a widely-used model for social influence in cross-sectional data. Through extensive simulation studies, we demonstrated the effectiveness of these homophily-adjusted models in providing deeper insights into network dynamics, particularly emphasizing the significance of accounting for latent homophily in social influence estimation. We included an R package for each method presented, ensuring straightforward implementation for users.
546 $a English
590 $a School code: 0096
650 4 $a Biostatistics. $3 1002712
650 4 $a Bioinformatics. $3 553671
653 $a Networks
653 $a Traditional statistical methods
653 $a Network autocorrelation model
653 $a Bayesian statistics
690 $a 0308
690 $a 0800
690 $a 0715
710 2 $a The University of Iowa. $b Biostatistics. $e degree granting institution. $3 3817109
720 1 $a Sewell, Daniel K. $e degree supervisor.
773 0 $t Dissertations Abstracts International $g 86-01B.
790 $a 0096
791 $a Ph.D.
792 $a 2024
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31235919