Abstract
Social media websites allow users to exchange short texts such as tweets via microblogs and user status in friendship networks. Their limited length, pervasive abbreviations, and coined acronyms and words exacerbate the problems of synonymy and polysemy, and bring about new challenges to data mining applications such as text clustering and classification. To address these issues, we dissect some potential causes and devise an efficient approach that enriches data representation by employing machine translation to increase the number of features from different languages. Then we propose a novel framework which performs multi-language knowledge integration and feature reduction simultaneously through matrix factorization techniques. The proposed approach is evaluated extensively in terms of effectiveness on two social media datasets from Facebook and Twitter. With its significant performance improvement, we further investigate potential factors that contribute to the improved performance.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 88-101 |
| Number of pages | 14 |
| Journal | Frontiers of Computer Science in China |
| Volume | 6 |
| Issue number | 1 |
| DOIs | |
| State | Published - Feb 2012 |
Fingerprint
Keywords
- matrix factorization
- multi-language knowledge
- short texts
- social media
- text representation
ASJC Scopus subject areas
- Computer Science(all)
- Theoretical Computer Science
Cite this
Enriching short text representation in microblog for clustering. / Tang, Jiliang; Wang, Xufei; Gao, Huiji; Hu, Xia; Liu, Huan.
In: Frontiers of Computer Science in China, Vol. 6, No. 1, 02.2012, p. 88-101.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Enriching short text representation in microblog for clustering
AU - Tang, Jiliang
AU - Wang, Xufei
AU - Gao, Huiji
AU - Hu, Xia
AU - Liu, Huan
PY - 2012/2
Y1 - 2012/2
N2 - Social media websites allow users to exchange short texts such as tweets via microblogs and user status in friendship networks. Their limited length, pervasive abbreviations, and coined acronyms and words exacerbate the problems of synonymy and polysemy, and bring about new challenges to data mining applications such as text clustering and classification. To address these issues, we dissect some potential causes and devise an efficient approach that enriches data representation by employing machine translation to increase the number of features from different languages. Then we propose a novel framework which performs multi-language knowledge integration and feature reduction simultaneously through matrix factorization techniques. The proposed approach is evaluated extensively in terms of effectiveness on two social media datasets from Facebook and Twitter. With its significant performance improvement, we further investigate potential factors that contribute to the improved performance.
AB - Social media websites allow users to exchange short texts such as tweets via microblogs and user status in friendship networks. Their limited length, pervasive abbreviations, and coined acronyms and words exacerbate the problems of synonymy and polysemy, and bring about new challenges to data mining applications such as text clustering and classification. To address these issues, we dissect some potential causes and devise an efficient approach that enriches data representation by employing machine translation to increase the number of features from different languages. Then we propose a novel framework which performs multi-language knowledge integration and feature reduction simultaneously through matrix factorization techniques. The proposed approach is evaluated extensively in terms of effectiveness on two social media datasets from Facebook and Twitter. With its significant performance improvement, we further investigate potential factors that contribute to the improved performance.
KW - matrix factorization
KW - multi-language knowledge
KW - short texts
KW - social media
KW - text representation
UR - http://www.scopus.com/inward/record.url?scp=84863014790&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84863014790&partnerID=8YFLogxK
U2 - 10.1007/s11704-011-1167-7
DO - 10.1007/s11704-011-1167-7
M3 - Article
AN - SCOPUS:84863014790
VL - 6
SP - 88
EP - 101
JO - Frontiers of Computer Science
JF - Frontiers of Computer Science
SN - 2095-2228
IS - 1
ER -