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

Spread of Opinion in a Social Network!

Mallick, Shailaja.

FindBook

Google Book

Amazon

博客來

Spread of Opinion in a Social Network!

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Spread of Opinion in a Social Network!/
作者:	Mallick, Shailaja.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2023,
面頁冊數:	109 p.
附註:	Source: Dissertations Abstracts International, Volume: 84-10, Section: A.
Contained By:	Dissertations Abstracts International84-10A.
標題:	Computer & video games. -
電子資源:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30364030
ISBN:	9798377682301

Spread of Opinion in a Social Network!
Mallick, Shailaja.

Spread of Opinion in a Social Network! - Ann Arbor : ProQuest Dissertations & Theses, 2023 - 109 p.

Source: Dissertations Abstracts International, Volume: 84-10, Section: A.

Thesis (Ph.D.)--North Carolina State University, 2023.

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

When a new product enters a market already dominated by an existing product, will it survive and coexist with this dominant product? Most existing research has represented the coexistence of two competing products spreading or being adopted as overlaid graphs of social networks with a common set of individuals. However, for the survival of a weaker product on the same graph, it has been established that the stronger product dominates and eliminates the weaker competition. In this thesis, we take a step towards narrowing this gap so that even a new or weaker product may survive with a positive market share. Specifically, we attempt to identify a locally optimal set of users over time to form an influencer community that can be targeted by a company launching a new product under a given budget constraint. We model the system as a competing Susceptible-Infected-Susceptible (SIS) epidemic and employ perturbation techniques that identify budget parameters to produce a positive market share in a cost-efficient manner. Our simulations with real-world graph datasets and performance comparisons against standard centrality measures demonstrate that with our choice of target nodes, a new product can establish itself, rather than being dominated and eventually pushed out of the market under the same budget constraint. Although SIS-based models determine users' opinions towards a product using exogenous model parameters, online forums or reviews are the most common real-world medium for users to express their opinions. This in turn helps other potential purchasers before making a buying decision. Important features of a product evaluated in reviews make it easier for a user to build an informed decision. We take advantage of online reviews posted in e-commerce websites to analyze their sentiment and visualize the results as a node-link graph highlighting how reviews help users form opinions. We construct a sentiment-weighted property list assigning 1, 0, or −1 to positive, neutral, or negative senti-ment for each important product property discussed in the reviews. This labeled dataset acts as input to machine learning algorithms that can predict a user's preferences based only on property sentiment extracted from their reviews. These preferences are integrated into our SIS model to optimize which users to target and which product features to emphasize. To better represent our findings, we employ visualization to provide information about the structure of the users and their relationships. Our sentiment visualization uses node-link graphs to present users, their relationships, and the spread of influence through review sentiment about product features. This illustrates how potential purchasers form opinions on the basis of a product's features combined with influence from their neighbors. Our novel contributions include: (1) perturbation to modify bi-SIS epidemic model parameters in a competitive product domain; (2) optimal budget allocation for advocate community creation; (3) sentiment analysis of review datasets to choose and weight important product aspects and properties for bi-SIS elements; and (4) a text-to-visual representation of the integrated bi-SIS/sentiment model showing how influence is spread between potential product adopters.

ISBN: 9798377682301Subjects--Topical Terms:

3548317
Computer & video games.

Spread of Opinion in a Social Network!
LDR:04406nmm a2200361 4500 001 2395658
005 20240517104944.5
006 m o d
007 cr#unu||||||||
008 251215s2023 ||||||||||||||||| ||eng d
020 $a 9798377682301
035 $a (MiAaPQ)AAI30364030
035 $a (MiAaPQ)NCState_Univ18402040246
035 $a AAI30364030
040 $a MiAaPQ $c MiAaPQ
100 1 $a Mallick, Shailaja. $3 3765169
245 1 0 $a Spread of Opinion in a Social Network!
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2023
300 $a 109 p.
500 $a Source: Dissertations Abstracts International, Volume: 84-10, Section: A.
500 $a Advisor: Eun, Do Young;Healey, Christopher.
502 $a Thesis (Ph.D.)--North Carolina State University, 2023.
506 $a This item must not be sold to any third party vendors.
520 $a When a new product enters a market already dominated by an existing product, will it survive and coexist with this dominant product? Most existing research has represented the coexistence of two competing products spreading or being adopted as overlaid graphs of social networks with a common set of individuals. However, for the survival of a weaker product on the same graph, it has been established that the stronger product dominates and eliminates the weaker competition. In this thesis, we take a step towards narrowing this gap so that even a new or weaker product may survive with a positive market share. Specifically, we attempt to identify a locally optimal set of users over time to form an influencer community that can be targeted by a company launching a new product under a given budget constraint. We model the system as a competing Susceptible-Infected-Susceptible (SIS) epidemic and employ perturbation techniques that identify budget parameters to produce a positive market share in a cost-efficient manner. Our simulations with real-world graph datasets and performance comparisons against standard centrality measures demonstrate that with our choice of target nodes, a new product can establish itself, rather than being dominated and eventually pushed out of the market under the same budget constraint. Although SIS-based models determine users' opinions towards a product using exogenous model parameters, online forums or reviews are the most common real-world medium for users to express their opinions. This in turn helps other potential purchasers before making a buying decision. Important features of a product evaluated in reviews make it easier for a user to build an informed decision. We take advantage of online reviews posted in e-commerce websites to analyze their sentiment and visualize the results as a node-link graph highlighting how reviews help users form opinions. We construct a sentiment-weighted property list assigning 1, 0, or −1 to positive, neutral, or negative senti-ment for each important product property discussed in the reviews. This labeled dataset acts as input to machine learning algorithms that can predict a user's preferences based only on property sentiment extracted from their reviews. These preferences are integrated into our SIS model to optimize which users to target and which product features to emphasize. To better represent our findings, we employ visualization to provide information about the structure of the users and their relationships. Our sentiment visualization uses node-link graphs to present users, their relationships, and the spread of influence through review sentiment about product features. This illustrates how potential purchasers form opinions on the basis of a product's features combined with influence from their neighbors. Our novel contributions include: (1) perturbation to modify bi-SIS epidemic model parameters in a competitive product domain; (2) optimal budget allocation for advocate community creation; (3) sentiment analysis of review datasets to choose and weight important product aspects and properties for bi-SIS elements; and (4) a text-to-visual representation of the integrated bi-SIS/sentiment model showing how influence is spread between potential product adopters.
590 $a School code: 0155.
650 4 $a Computer & video games. $3 3548317
650 4 $a Competition. $3 537031
650 4 $a Computer science. $3 523869
650 4 $a Information technology. $3 532993
650 4 $a Recreation. $3 535376
650 4 $a Web studies. $3 2122754
690 $a 0984
690 $a 0489
690 $a 0814
690 $a 0646
710 2 $a North Carolina State University. $3 1018772
773 0 $t Dissertations Abstracts International $g 84-10A.
790 $a 0155
791 $a Ph.D.
792 $a 2023
793 $a English
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30364030