抄録
According to the remarkable progress of Web technologies, people search various types of items such as movies, books, and CDs on the Internet. Items that are returned as search results would be ranked based on objective criteria such as rating scores. Such rankings are not personalized to a user based on his/her interests and preferences. In this paper, we introduce a method for ranking items that are specified by users. Our method does not request users to specify their interests and preference manually. Our method focuses on the users who have posted tweets about the items and extracts features of the users. Items are ranked according to the preferences of the user based on the similarities of twitter users. The effectiveness of the proposed method is evaluated with experimental results on subjects.
| 元の言語 | 英語 |
|---|---|
| DOI | |
| 出版物ステータス | 出版済み - 1 1 2014 |
| イベント | 8th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2014 - Siem Reap, カンボジア 継続期間: 1 9 2014 → 1 11 2014 |
その他
| その他 | 8th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2014 |
|---|---|
| 国 | カンボジア |
| 市 | Siem Reap |
| 期間 | 1/9/14 → 1/11/14 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Computer Networks and Communications
- Information Systems
これを引用
A method for personalized ranking of items based on similarity between twitter users. / Ushiama, Taketoshi; Tominaga, Kazushige.
2014. 論文発表場所 8th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2014, Siem Reap, カンボジア.研究成果: 会議への寄与タイプ › 論文
}
TY - CONF
T1 - A method for personalized ranking of items based on similarity between twitter users
AU - Ushiama, Taketoshi
AU - Tominaga, Kazushige
PY - 2014/1/1
Y1 - 2014/1/1
N2 - According to the remarkable progress of Web technologies, people search various types of items such as movies, books, and CDs on the Internet. Items that are returned as search results would be ranked based on objective criteria such as rating scores. Such rankings are not personalized to a user based on his/her interests and preferences. In this paper, we introduce a method for ranking items that are specified by users. Our method does not request users to specify their interests and preference manually. Our method focuses on the users who have posted tweets about the items and extracts features of the users. Items are ranked according to the preferences of the user based on the similarities of twitter users. The effectiveness of the proposed method is evaluated with experimental results on subjects.
AB - According to the remarkable progress of Web technologies, people search various types of items such as movies, books, and CDs on the Internet. Items that are returned as search results would be ranked based on objective criteria such as rating scores. Such rankings are not personalized to a user based on his/her interests and preferences. In this paper, we introduce a method for ranking items that are specified by users. Our method does not request users to specify their interests and preference manually. Our method focuses on the users who have posted tweets about the items and extracts features of the users. Items are ranked according to the preferences of the user based on the similarities of twitter users. The effectiveness of the proposed method is evaluated with experimental results on subjects.
UR - http://www.scopus.com/inward/record.url?scp=84899709888&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84899709888&partnerID=8YFLogxK
U2 - 10.1145/2557977.2558031
DO - 10.1145/2557977.2558031
M3 - Paper
AN - SCOPUS:84899709888
ER -