國立高雄大學圖資館 |

語系: 繁體中文

說明(常見問題)

圖資館首頁

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

On Estimation of Contagion-based Soc...

North Carolina State University.

On Estimation of Contagion-based Social Network Dependence with Event Time Data.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	On Estimation of Contagion-based Social Network Dependence with Event Time Data.
作者:	Yu, Lin.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, 2018
面頁冊數:	77 p.
附註:	Source: Dissertation Abstracts International, Volume: 80-03(E), Section: B.
Contained By:	Dissertation Abstracts International80-03B(E).
標題:	Statistics.
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=11007248
ISBN:	9780438599444

On Estimation of Contagion-based Social Network Dependence with Event Time Data.
Yu, Lin.

On Estimation of Contagion-based Social Network Dependence with Event Time Data. - Ann Arbor : ProQuest Dissertations & Theses, 2018 - 77 p.

Source: Dissertation Abstracts International, Volume: 80-03(E), Section: B.

Thesis (Ph.D.)--North Carolina State University, 2018.

As the advances of a wide variety of data collection methods and the emerging of a growing number of social networking services, the study of social networks has received much attention. In this dissertation, we analyze two different types of social network datasets and propose new methods to address them respectively. The first one refers to a social network with time-to-event data, and the second one considers a sequence of recurrent events data within a social network group. Our objective for both studies is to model the contagion-based social network dependence by analyzing how users behave through the influence of social events.

ISBN: 9780438599444Subjects--Topical Terms:

182057
Statistics.

On Estimation of Contagion-based Social Network Dependence with Event Time Data.
LDR:03132nmm a2200313 4500 001 547667
005 20190513114600.5
008 190715s2018 ||||||||||||||||| ||eng d
020 $a 9780438599444
035 $a (MiAaPQ)AAI11007248
035 $a AAI11007248
040 $a MiAaPQ $c MiAaPQ
100 1 $a Yu, Lin. $3 827053
245 1 0 $a On Estimation of Contagion-based Social Network Dependence with Event Time Data.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2018
300 $a 77 p.
500 $a Source: Dissertation Abstracts International, Volume: 80-03(E), Section: B.
502 $a Thesis (Ph.D.)--North Carolina State University, 2018.
520 $a As the advances of a wide variety of data collection methods and the emerging of a growing number of social networking services, the study of social networks has received much attention. In this dissertation, we analyze two different types of social network datasets and propose new methods to address them respectively. The first one refers to a social network with time-to-event data, and the second one considers a sequence of recurrent events data within a social network group. Our objective for both studies is to model the contagion-based social network dependence by analyzing how users behave through the influence of social events.
520 $a This dissertation contains three parts. Chapter 1 gives an overall introduction of social network analysis and overview of the underlying basis models for both studies. In Chapter 2, we extend generalized linear transformation model to study the social network influence with network-based time-to-event data. A time-varying covariate is proposed to incorporate network structure into the model and quantify the contagion-based social correlation. We further introduce a novel data generation procedure in simulations and establish the asymptotic properties of the proposed estimators. The numerical performances of the estimators are demonstrated via both simulation studies and a real-world application.
520 $a In Chapter 3, we propose a new way to model the contagion-based social network correlation with recurrent events data. It is well known that individuals are influenced through the network ties. In particular, the future actions of individuals depend not only on their own past behaviors, but also on their friends' past activities. Thus, we generalize the Hawkes self-exciting point process to model both self and mutually exciting influence in a social network dataset. A semi-parametric estimation method is considered for model flexibility. Consistency and asymptotic normality of the proposed estimators are further established. Both simulation studies and an analysis of an online social network dataset are provided to illustrate the empirical performance of the proposed method for the parameters estimation and the influential group detection.
590 $a School code: 0155.
650 4 $a Statistics. $3 182057
650 4 $a Behavioral psychology. $3 827055
650 4 $a Computer science. $3 199325
690 $a 0463
690 $a 0384
690 $a 0984
710 2 $a North Carolina State University. $b Statistics. $3 827054
773 0 $t Dissertation Abstracts International $g 80-03B(E).
790 $a 0155
791 $a Ph.D.
792 $a 2018
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=11007248