語系:
繁體中文
English
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
Investigating Sentiment, Homophily, ...
~
Yuan, Guangchao.
FindBook
Google Book
Amazon
博客來
Investigating Sentiment, Homophily, and Location for Understanding User Interactions in Social Media.
紀錄類型:
書目-電子資源 : Monograph/item
正題名/作者:
Investigating Sentiment, Homophily, and Location for Understanding User Interactions in Social Media./
作者:
Yuan, Guangchao.
出版者:
Ann Arbor : ProQuest Dissertations & Theses, : 2016,
面頁冊數:
105 p.
附註:
Source: Dissertation Abstracts International, Volume: 78-08(E), Section: B.
Contained By:
Dissertation Abstracts International78-08B(E).
標題:
Computer science. -
電子資源:
http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10585486
ISBN:
9781369639582
Investigating Sentiment, Homophily, and Location for Understanding User Interactions in Social Media.
Yuan, Guangchao.
Investigating Sentiment, Homophily, and Location for Understanding User Interactions in Social Media.
- Ann Arbor : ProQuest Dissertations & Theses, 2016 - 105 p.
Source: Dissertation Abstracts International, Volume: 78-08(E), Section: B.
Thesis (Ph.D.)--North Carolina State University, 2016.
With the rapid development of Web 2.0 technologies, smart phones, and global position system (GPS), location sharing services have become prevalent on social media. Instead of passively absorbing information, users have become active participants in social media via various user interactions.
ISBN: 9781369639582Subjects--Topical Terms:
523869
Computer science.
Investigating Sentiment, Homophily, and Location for Understanding User Interactions in Social Media.
LDR
:03904nmm a2200361 4500
001
2124580
005
20171030113307.5
008
180830s2016 ||||||||||||||||| ||eng d
020
$a
9781369639582
035
$a
(MiAaPQ)AAI10585486
035
$a
AAI10585486
040
$a
MiAaPQ
$c
MiAaPQ
100
1
$a
Yuan, Guangchao.
$3
3286585
245
1 0
$a
Investigating Sentiment, Homophily, and Location for Understanding User Interactions in Social Media.
260
1
$a
Ann Arbor :
$b
ProQuest Dissertations & Theses,
$c
2016
300
$a
105 p.
500
$a
Source: Dissertation Abstracts International, Volume: 78-08(E), Section: B.
500
$a
Adviser: Munindar Singh.
502
$a
Thesis (Ph.D.)--North Carolina State University, 2016.
520
$a
With the rapid development of Web 2.0 technologies, smart phones, and global position system (GPS), location sharing services have become prevalent on social media. Instead of passively absorbing information, users have become active participants in social media via various user interactions.
520
$a
We divide user interactions into two categories: interactions with other users and interactions with information objects. The information objects can be locations, blogs, photos, and so on. Understanding user interactions from the massive amount of available data in social media can help bridge the gap between users' online and offline activities and thereby improve the quality of recommender systems and search engines.
520
$a
There are two common characteristics of user interactions. First, user interactions usually involve data sources across multiple layers: content, social, geography, and time. Second, even though a huge amount of data is generated every day, it is sparse. Therefore, how to exploit data from various layers and overcome the data sparsity is important in understanding user interactions. We propose solutions mainly from two perspectives: exploiting (1) content, (2) homophily---similarity between nodes breeds connections between them.
520
$a
With the ultimate goal of better understanding user interactions, this dissertation makes three main contributions. The first contribution is a framework of exploiting sentiment homophily for link prediction, with theoretical modeling and empirical evaluation. This framework helps answer the question of whether applying the homophily principle to the content would improve link prediction.
520
$a
The second contribution is an approach of estimating the location of where a message originated. Due to the importance of geo-tagged messages (e.g., advertising), whether we can exploit homophily and the large amount of available content to overcome the sparsity of locations is important. Evaluation results on a Twitter dataset demonstrate that (1) the prediction error could be greatly reduced by applying homophily to both the social and the geographical layers, (2) our proposed approach of relating one user's content to another user's locations is effective in reducing prediction error.
520
$a
The third contribution lies in how to exploit content and geographical homophily to improve the performance of point-of-interest (POI) recommendation. Motivated by the sparsity of the user-POI check-in matrix, we propose a context-aware framework to improve recommendation quality. The context of a POI includes (1) aspect-based sentiment extracted from reviews and, (2) environmental context created by its surrounding POIs (neighborhoood effect). Our work could provide insights about how to measure a user's preference for a POI by modeling aspect-based sentiment, and how to model the neighborhood effect in general.
590
$a
School code: 0155.
650
4
$a
Computer science.
$3
523869
650
4
$a
Information science.
$3
554358
650
4
$a
Social research.
$3
2122687
690
$a
0984
690
$a
0723
690
$a
0344
710
2
$a
North Carolina State University.
$3
1018772
773
0
$t
Dissertation Abstracts International
$g
78-08B(E).
790
$a
0155
791
$a
Ph.D.
792
$a
2016
793
$a
English
856
4 0
$u
http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10585486
筆 0 讀者評論
館藏地:
全部
電子資源
出版年:
卷號:
館藏
1 筆 • 頁數 1 •
1
條碼號
典藏地名稱
館藏流通類別
資料類型
索書號
使用類型
借閱狀態
預約狀態
備註欄
附件
W9335192
電子資源
01.外借(書)_YB
電子書
EB
一般使用(Normal)
在架
0
1 筆 • 頁數 1 •
1
多媒體
評論
新增評論
分享你的心得
Export
取書館
處理中
...
變更密碼
登入