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

Learning-Based Social Recommender Sy...

Liu, Xiang.

FindBook

Google Book

Amazon

博客來

Learning-Based Social Recommender Systems and Applications.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Learning-Based Social Recommender Systems and Applications./
作者:	Liu, Xiang.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2017,
面頁冊數:	90 p.
附註:	Source: Dissertation Abstracts International, Volume: 78-10(E), Section: B.
Contained By:	Dissertation Abstracts International78-10B(E).
標題:	Computer science. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10255390
ISBN:	9781369840063

Learning-Based Social Recommender Systems and Applications.
Liu, Xiang.

Learning-Based Social Recommender Systems and Applications. - Ann Arbor : ProQuest Dissertations & Theses, 2017 - 90 p.

Source: Dissertation Abstracts International, Volume: 78-10(E), Section: B.

Thesis (Ph.D.)--Polytechnic Institute of New York University, 2017.

With the rapid growth of many online social networking services, recommender systems have emerged to play an essential role in aggregating information from large-scale data. However, the heterogeneous entities and relations in such networks make the recommendation task very challenging. To fully represent and exploit the diverse information, this dissertation focuses on graph-based approaches to integrate different types of social objects, and on machine learning methods to explore item-based recommendation. By analyzing two different types of social networks, a co-authorship network, and an event-based social network, we study three different facets of top-N recommendation.

ISBN: 9781369840063Subjects--Topical Terms:

523869
Computer science.

Learning-Based Social Recommender Systems and Applications.
LDR:04320nmm a2200349 4500 001 2200116
005 20181214114458.5
008 201008s2017 ||||||||||||||||| ||eng d
020 $a 9781369840063
035 $a (MiAaPQ)AAI10255390
035 $a (MiAaPQ)poly:10338
035 $a AAI10255390
040 $a MiAaPQ $c MiAaPQ
100 1 $a Liu, Xiang. $3 1275849
245 1 0 $a Learning-Based Social Recommender Systems and Applications.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2017
300 $a 90 p.
500 $a Source: Dissertation Abstracts International, Volume: 78-10(E), Section: B.
500 $a Advisers: Nasir Memon; Torsten Suel.
502 $a Thesis (Ph.D.)--Polytechnic Institute of New York University, 2017.
520 $a With the rapid growth of many online social networking services, recommender systems have emerged to play an essential role in aggregating information from large-scale data. However, the heterogeneous entities and relations in such networks make the recommendation task very challenging. To fully represent and exploit the diverse information, this dissertation focuses on graph-based approaches to integrate different types of social objects, and on machine learning methods to explore item-based recommendation. By analyzing two different types of social networks, a co-authorship network, and an event-based social network, we study three different facets of top-N recommendation.
520 $a Automatic expert assignment is a common problem encountered in both industry and academia. In general, for each expert recommendation task, several aspects are usually jointly considered to make the final decision. In this dissertation, we focus on the problem of paper-reviewer recommendation, which has been studied by several other researchers. Most existing papers have focused on improving the relevance between the paper and experts, thus expertise is considered as the main criterion for relevance. This dissertation proposes an automatic paper-reviewer recommendation system that propagates the paper query over a co-authorship network and considers aspects of expertise, authority, and diversity. We show that the proposed method obtains performance gains over state-of-the-art reviewer recommendation systems in terms of expertise, authority, diversity, and, most importantly, relevance as judged by human experts.
520 $a Event-based online social networks, which are used to maintain interest-based groups and to distribute and organize offline events, have recently increased in popularity. Such platforms are used to announce millions of offline events every month. Thus users are easily to be overwhelmed by a lot of online notification streams. To alleviate this information overload, this dissertation investigates the event recommendation problem on Meetup, a social event organizing service, using a meta-path-based approach that integrates content-based and collaborative-filtering-based methods. We show that the model with latent meta-path-based features achieves better performance than a state-of-the-art matrix factorization approach.
520 $a Finally, we conduct a large-scale analysis of an event-based social network to explore if it is possible to automatically detect group failure. Unlike the existing work, which models social group evolution as an information diffusion process and measures group growth from the membership growth point of view, we first justify a definition of group failure as no events being organized or no event RSVPs being collected for twelve months. We then investigate both the statistical and structural features of the social groups and find that event features play an important role in distinguishing social groups with different topics and categories. We use two different feature selection methods and build a model to predict which groups will fail over a period of time. The experimental results show that social group failures on Meetup can be predicted with high accuracy, and that member features contribute significantly to the success of social groups.
590 $a School code: 1540.
650 4 $a Computer science. $3 523869
650 4 $a Information technology. $3 532993
650 4 $a Web studies. $3 2122754
690 $a 0984
690 $a 0489
690 $a 0646
710 2 $a Polytechnic Institute of New York University. $b Computer Science and Engineering. $3 2103713
773 0 $t Dissertation Abstracts International $g 78-10B(E).
790 $a 1540
791 $a Ph.D.
792 $a 2017
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10255390